Thanks to Bob Kouré and Alan Sharkey for proof-reading.
Thanks to Wil Smith over in Austin, Texas for help in rewriting the manual.

Introduction

MRWFormat is a plug-in for Adobe Photoshop (and other compatible software) to open and pre‑process raw images taken with Minolta DiMAGE digital still cameras. The main goal of this plug-in is to allow the user to access all the benefits of shooting in RAW format, giving one complete control over every step of image processing.

The software supplied with the camera, DiMAGE Viewer (a.k.a. DiVU), is designed for an occasional user. DiVU is very easy to use and there are some basic image editing functions built-in. However even a slightly advanced user will start to miss many of the important features. One can use DiVU just as a converter from RAW, and then do the image editing in another software. Typically some users eventually find that shooting in RAW has so many benefits that they want to shoot in RAW exclusively. This is when one chooses to leave DiVU, and find a faster processor with more robust controls.

I was in this situation some time ago. I was looking for a decent converter from RAW to TIFF that would just allow me to pre-process an image, as fast as possible, and then work on it in Photoshop later. Unfortunately there was no tool that met my needs. So I decided to create my own tool.

During my development Adobe announced CameraRAW, and some time later it was released. I was disappointed at first for I thought that all my work was in vain. But now that I have seen it, I am sure that there is still a place for my plug-in. Adobe CameraRAW is definitely a nice product. It is both quite fast and handsome. But it has some drawbacks too. While it can work with RAW files from most cameras on the market, its capabilities are reduced to those that are common to all supported cameras. And most importantly it contains only one interpolation algorithm. It seems to be very good, but my tests have concluded that there is no all inclusive "best" algorithm for RAW processing. The algorithm that produces the finest result with one picture might be very lacking with another. Beauty is in the eye of the beholder and one should test different algorithms to find the result that meets your goals.

Here’s an image of my plug-in’s interface with all its various screens displayed, and some of Photoshop’s pallets on the right.

Requirements

This plug-in should work on any computer that is able to run Adobe Photoshop. There are really no specific requirements, but the faster your processor(s) the faster it will work.

PC version

·         Pentium processor required; Pentium III or higher recommended

·         64 MB of RAM; 256 MB recommended

·         a screen resolution at least 1024x768 in 24 bit colors

·         Windows 95 or higher, Windows XP recommended

·         supported cameras:
Minolta DiMAGE 5
Minolta DiMAGE 7
Minolta DiMAGE 7u (this is DiMAGE 7 with firmware upgraded to version 2.xx)
Minolta DiMAGE 7i
Minolta DiMAGE 7Hi
Minolta DiMAGE A1
Minolta DiMAGE A2

·         supported host applications:
Adobe Photoshop CS (8)
Adobe Photoshop 7
Adobe Photoshop 6
Adobe Photoshop Elements 1
Adobe Photoshop Elements 2
Adobe ImageReady 3
Adobe ImageReady 7
These applications were tested with the plug-in. There may be other applications that will work too. This version of the plug-in will not work with Corel PhotoPaint, JASC PaintShopPro or any other non-Adobe host software.

MacOS X version

·         PowerPC® processor (G3, G4, G4-dual, G5, G5-dual)

·         128 MB of RAM; 256 MB recommended

·         a screen resolution at least 1024x768 in 24 bit colors

·         MacOS X v.10.1.3 or higher

·         supported cameras:
Minolta DiMAGE 5
Minolta DiMAGE 7
Minolta DiMAGE 7u (this is DiMAGE 7 with firmware upgraded to version 2.xx)
Minolta DiMAGE 7i
Minolta DiMAGE 7Hi
Minolta DiMAGE A1
Minolta DiMAGE A2

·         supported host applications:
Adobe Photoshop CS (8)
Adobe Photoshop 7
Adobe Photoshop Elements 2
Adobe ImageReady 8
Adobe ImageReady 7

MacOS 9 version

Not planned at this moment.

 

 

Installation

You should download the file named MRWFormat.zip, or the MRWFormat.dmg.gz file for Macintosh users.

Unpack it in any location on your hard disk.

The package contains three files:

·         MRWFormat.8bi
This is the plug-in itself.

·         ReadMe.txt
The text file containing the description of the latest changes in the plug-in. This information was not included in this User Guide. You should always read this file as it might contain very important information.

·         MRWFormat UserGuide.pdf
Documentation in Adobe Acrobat format. It is very much to your advantage to fully read this document for there are many tips, and very functional explanations that will greatly help your use and understanding of this plug-in.

Now you must locate the folder where Photoshop stores its plug-ins. It is a folder called something like C:\Program Files\Adobe\Photoshop 6.0\Plug-Ins\. On a Macintosh it is usually Applications‑>Adobe Photoshop 7->Plug-Ins.

 If you are not sure which folder it is you can start Adobe Photoshop and from the menu choose Edit‑>Preferences‑>Plug‑Ins & Scratch Disks. A window will open and the name of the required folder is displayed in its upper part. On a Macintosh the path is: from the Menu bar choose Photoshop‑>Preferences‑>Plug‑Ins & Scratch Disks.

Copy the file MRWFormat.8bi to this folder.

Start Adobe Photoshop (you have to close Photoshop and start it again if it was running at the moment you copied the files in plug-in folder).

Verify that the plug-in was loaded by Photoshop from the menu choose Help‑>About Plug‑in‑>MRWFormat…
On a Macintosh from the menu bar choose Photoshop‑>About Plug‑In‑>MWRFormat… This should display the About dialog.

Tip for Adobe Camera Raw (Photoshop CS) users:
If you have Adobe Camera Raw plug-in installed on your computer, (it comes bundled with Photoshop CS,) then MRWFormat will take precedence and your Minolta raw images will be opened using MRWFormat. The MRWFormat plug-in will be used for displaying the thumbnails and metadata in Photoshop’s Browser as well.

If you would like to open .mrw file in Adobe Camera Raw plug-in then you can use the function File->Open As… and select file type Camera Raw. If you need to open and process more files using Camera Raw instead of MRWFormat then you can temporarily disable MRWFormat. To do so quit Photoshop, locate the file MRWFormat.8bi in Photoshop’s plug-ins folder and precede its name with tilde, renaming it to ~MRWFormat.8bi

Obtaining updated versions

The development of MRWFormat is definitely an on going project. I release the new versions quite often and the updates are free of charge. The updated versions will be available to download from the internet using the support forum (mentioned in the next section). To make sure that you will not miss the new version releases please do the following:

1.       Visit the support forum at http://www.dalibor.cz/support

2.       Click on Register in the upper part of the page

3.       You have to agree with the terms of the forum

4.       Enter you chosen username, password and valid e-mail address
(you can fill out the other details as well)

5.       Then click on Submit at the bottom of the page

6.       Note this username and password and store it on a safe place as you will need it in the future

7.       Then e-mail me the following information on my address mrw@dalibor.cz :
- your real name
- your chosen username
- the model of the camera you are using (e.g. DiMAGE A1)
- the version of the software you are using (e.g. Adobe Photoshop CS)
- the version of the operating system (e.g. Windows XP Home, MacOS X 10.2.6)
- where have you found the information about my plug-in (e.g. dpreview.com)

8.       When I will receive your email with this information I will grant you an access to the Download Area in the forum. I will send you a confirmation email as well.

All this has to be done only once and all the information you will send me will not be given to anyone else.

When the new version will be released I will send an e-mail notification to all the users who have sent me this information. You will then do the following:

1.       Visit the support forum http://www.dalibor.cz/support again

2.       Click on Log in in the upper part of the page,

3.       Enter your own username and the password and click on Log in button

4.       Enter the forum section "Download Area"

5.       Open the post named “Download link to the updated versions”

6.       Follow the link to the download page

7.       Choose the latest version you want to download

8.       Save it on your computer and unpack it

9.       Close Adobe Photoshop

10.   Overwrite the old file MRWFormat.8bi located in Photoshop’s Plug-ins folder with the downloaded one

11.   Start Adobe Photoshop

Support

You will probably want to ask few questions about MRWFormat or Minolta raw files in general. Please do not use my e-mail as a primary method of getting the support. There are quite a lot of MRWFormat users and it is very time consuming to answer all your questions by e-mail, especially because the vast majority of the questions keep repeating over and over. Therefore I have created web support forum. You can find it on the address http://www.dalibor.cz/support and I assume that you have created your own username there already as this is the place where you can get the latest version of my software.

So if you want to know something about MRWFormat, Minolta DiMAGE Viewer (DiVU) or .mrw files please visit this forum. Before you post your question into one of the sections there, try to search the forum as it is possible you’re your question was asked and answered already. If it was not then feel free to post it. You should log in to this forum before you post your question. If you do so then you will be notified by e-mail when someone posts an answer to your question.

The development of this plug-in is definitely an on going project. I have many ideas for future improvements, and you are very welcome to send me all your comments and suggestions. My intention is to make this plug-in as user friendly as possible so I will seriously evaluate all your comments and try to include them in future releases. Please write all your comments and proposals to the forum’s Wish List section.

Quick start

After successfully installing the plug-in in Adobe Photoshop you can start processing your images. Just select from Photoshop’s menu File‑>Open and browse for your raw image with .mrw extension. When you have found the image you want to process just highlight it and click Open.

If you have a newer version of Photoshop (version 7 or higher) you can use the Browser function also. Just click File‑>Browse and Photoshop will display a directory tree such that you can see all your images as thumbnails. Just double-click the thumbnail of the .mrw file you want to process.

The main dialog of the plug-in will open and you can make changes to get the best combinations of settings for your image. The recommended procedure is:

a)       Display the preview of the image by clicking on the Show button.

b)      Find the best white balance for your image either by using White Balance radio buttons or by CTRL+click (Option+click or Command+click on Macintosh) on a neutral (gray) area of the preview image.

c)       Find the best setting for the Exposure slider. I suggest clicking the Max button, which will give you a good starting point.

d)      Set Contrast & Gamma to Precise.

e)       Play with Gamma and Contrast sliders until you get the best combination for your image.

f)        Darken the Shadows and brighten the Highlights.

g)      Enable the Interpolation check box (and perhaps the Zoom too).

h)      Find the best interpolation algorithm for your image by trying different ones listed in the Interpolation menu. You should choose a few areas in your image and see the effect of different interpolations.
If you do not want to bother with trying different algorithms then the best single choice is Color Correction I.

i)        Click the Open button and the image will be processed according to your settings and it will be passed to Photoshop where you can continue editing, and eventually save the image.

After you have processed your first image successfully you should play with the rest of the controls. I cannot stress enough please read all this manual, since it contains excellent information, and tips for the best results with this plug-in.

Detailed description of controls

This section contains a detailed explanation of each control. I will include my personal tips for the best settings.

Main dialog

The main dialog opens after you choose the raw image (.mrw extension) from Photoshop’s File‑>Open menu or when you double-click the raw image in Photoshop’s Browser. There are two different layouts from which you may choose.

 

 

 

 

 

 

 

 

The choice of layouts is set in the Preferences dialog described later.

 

 

 

 

 

There are many controls in the main dialog, and it may appear quite confusing. Do not worry; it is not as confusing as it first appears.

Open Button

This button applies the current settings, processes the image and passes it to Photoshop, where you can continue editing it using Photoshop’s powerful tools, and ultimately save it.

Tip: keyboard shortcut

You can hit the key Enter instead of clicking on OK button.

Cancel Button

This button cancels the opening of the image, and the plug-in is closed. Note that you can select multiple images in File‑>Open dialog, and the selected images will be processed one by one. In this case the Cancel button will not only cancel the current image but all the images selected.

Also you should know that when you press Cancel then the current setting of the plug-in is not saved and all possible changes done in presets or the white balance menu will be lost.

Tip: keyboard shortcut

You can hit the key Esc instead of clicking on Cancel button.

Reset Button

The Cancel (Reset) button functions the same as in Adobe Photoshop. Some people over look this, but it is quite convenient. When one holds ALT key (Option on Mac) down then the Cancel button will change to Reset allowing you to revert to the default settings.

If you want to define your own default settings which would be used after you press Reset then you have to create your own preset named Default. For more information see section Presets.

Show Button - Preview Window

The preview window is used to watch your image as you find the best settings to process your raw image. To display it click the Show button. The preview window will display on the right side of the main dialog (or under it if the dialog orientation is horizontal). To hide it again click the same button again, which will now be labeled Hide.

In the beginning the preview window is “glued” to the main dialog, so if you move the main dialog the preview will follow. If you move the preview window it is “unglued”, and it stays in its position even if you move the main dialog. This is very convenient for those with dual monitors for they can move the preview window to the second monitor, and enlarge it. When you drag the window close to its original location it will be “glued” back to main dialog.

Windows only: When you click on the image in the preview window the RGB values of the selected pixel will be shown in the title bar of the preview window. Additionally a red cross + will be displayed marking the selected pixel. The values are displayed as rgb: (R_raw, G_raw, B_raw) ‑> (R_result, G_result, B_result). The first three are the R,G,B values as they are read from the raw file therefore they are in range 0-4095. The next three are the values that are the results of processing of the image; they are in range 0-255. If Color Depth is set to 48 bits then the values in range 0-65535 are displayed as well. Sometimes the values are followed by exclamation mark. The reason for displaying the exclamation mark is explained in the paragraphs Warning (pre) and Highlights fix. (This feature works on Windows only as there is no title bar in the preview window on Macintosh.)

Tip: keyboard shortcut

You can quickly show or hide the preview window by pressing the key P.

Info Button – Info Window

By clicking this button you can display the info window. The image histogram and some other interesting information about the image itself are displayed there. The info window is originally located under the preview window and it is “glued” to it. When the window is displayed, the text on the button is changed from Info to Hide Info and you can use the same button to close it again. I will describe the info window in detail later in the section Info window.

Tip: keyboard shortcut

You can quickly show or hide the info window by pressing the key I.

Preferences Button – Preferences Window

This button shows the preferences window where you can fine tune various settings of MRWFormat plug-in. It will be described later in the section Preferences.

Rotation

When your image was taken in portrait mode then you can rotate it in appropriate direction by using the buttons CCW and CW.

Tip: keyboard shortcut

You can rotate the image quickly by pressing the key R or T.

If you do use Photoshop CS you can use the rotation functions in Photoshop Browser. When you select the function File->Browse from the Photoshop menu then you will see the thumbnails of the mrw files. You can use the functions to rotate the image in this view and MRWFormat will respect this rotation. There are a few drawbacks in this approach though. First minus is that the information about image rotation is not stored in the picture itself but it is saved in Photoshop’s cache. When this cache is purges or the image or folder is renamed then the rotation is lost. The second problem is that Photoshop does tell MRWFormat to rotate the image when the image is opened. MRWFormat does rotate it when it displays the image but when you are finished with adjusting the settings and you press OK button then MRWFormat has to rotate the image back. The reason is that Photoshop will rotate the image when it will receive it from MRWFormat. So this way the rotation is carried out three times.

The same functions are in the Browser in Photoshop 7 and Photoshop Elements 2 but they cannot be used effectively. Although those applications will rotate the image after processing it with MRWFormat they do not tell MRWFormat what rotation was chosen by the user. Therefore MRWFormat will open and display the picture un-rotated.

The best way of rotating the mrw files is to use the software Dalifer. Dalifer is able to rotate the image and write the information about the requested rotation straight into the image itself. This rotation is not lost by renaming or moving the file. The best news is that this rotation is not respected only by MRWFormat but also by Camera Raw and Minolta DiVU. There are much more functions in Dalifer; for example displaying of the mrw thumbnails is Windows Explorer or older versions of Photoshop. If you want to find out more about Dalifer then visit the page http://www.dalibor.cz/minolta/dalifer.htm The users of MRWFormat can get Dalifer for half price.

Preview window size

The preview window cannot be resized freely by just dragging. This limitation is intentional and results in faster response to the user’s actions. The computer can process the image in a fixed size window much faster. There are four radio buttons that control how large the preview image will be. The 1/2 size means that every pixel of the preview image is created from four pixels of the original raw image. So the resulting preview image has half width and half height of the original raw file. The same logic applies to the other sizes, up to 1/16 where each pixel is an average of 256 raw pixels.

Of course the smaller fractional choices result in the preview window being redrawn faster, and the user’s interaction will be faster. The best size choice for you is dependant on the computing power of your processor(s). If you decide that you want the initial size of the preview to always be for example 1/4, then you can set it to this value and then go to Preferences and click Save which will save the layout as a default.

There are other preview sizes available also which do not have the corresponding radio buttons. You can use the keyboard shortcuts to select them. The resize the preview window using keyboard hold down the Ctrl or Alt key and press the key + or -. This way you can select additional preview sizes 1/6, 1/12 and 1/16. If the preview size is set 1/2 and you will hit + again then the preview will switch to Interpolated Preview and the next hit of + key will zoom in the interpolation preview area. The same logic applies to the shortcut Alt- in the opposite direction.

Tip: choosing the preview size

At the beginning of this process, when one is finding the best values for the sliders for an image conversion, use the small preview size. This will cause the window to be redrawn faster, and the controls will react more smoothly. When you reach the most pleasing result then increase the preview window size, and fine-tune your settings.

White Balance

The human eye can compensate for differences in lighting situations such that a white piece of paper looks white even under reddish or bluish light. Digital cameras don’t have this ability so, for instance, white paper under tungsten light will be red. The camera has to post process the image and get rid of this red cast. This is done by changing the ratio of red, green and blue values of pixels. The amount of red is reduced by some coefficient and the amount of blue is increased. This way the paper in the image will end up with equal values of R, G and B and it will look white on a monitor. The coefficients used to white balance the image are displayed in the R, G, G’ and B edit boxes. You can try to edit the values in those edit boxes directly and see what is their effect on the image. If you wonder why there are two coefficients for green color then the reason is simple. The CCD of your camera is composed from red, green and blue cells (there is a picture of the CCD layout at the beginning of the Interpolation algorithms section). The amount of blue and red cells is the same but there are twice as many green cells. One set of green cells is located beside red pixels and the second is beside the blue pixels. MRWFormat has the ability to treat those two sets of green in a different way, therefore there are two edit boxes named G and G’. But unless you are experimenting you should keep the value G’ equal to G.

There are three ways the camera selects the white balance for an image. In auto mode the camera examines the scene, and tries to guess the proper white balance coefficients. This is very a complicated task, but the camera has a very intelligent algorithm built-in. However occasionally the algorithm will fail in a difficult lightning situation. In this case one has to select one of the fixed white balance coefficients (daylight, cloudy, fluorescent, etc.). The most precise, and most complicated is custom white balance. In this mode you point the camera at something white and the camera will measure the proper white balance coefficients, and use them for subsequent pictures.

When taking a normal image the camera modifies the picture data by the selected white balance coefficients, and then stores that modified data. This is especially disastrous if the white balance algorithm fails, or if the user has mistakenly selected the wrong white balance preset. In these cases the image will look horrendous, and it is very time consuming to restore it. The resulting restoration is a very inferior final image compared to what would have been possible had the data not been altered. This is the most important advantage of shooting images using RAW. For the white balance coefficients are not used to alter the data before storing the image. The white balance information is just stored with the other image data, and you can easily change it later without affecting the quality of the image!

There is a white balance menu displayed in the main dialog. From this menu you can choose one of the fixed white balance settings which will be the best for your image. I will describe the settings in this menu and the buttons Save and Delete in the section White balance later.

Neutralize

The best way to white balance your image is to use the Neutralize button. Background info: the camera’s built-in algorithm has to struggle hard to find the area in a picture that should be converted to white. The human brain has no problem finding the white objects in a photographed scene so with a little of your help the plug-in can find the best white balance coefficients. Just open up the preview window, click the Neutralize button and then click in the area of the picture which should have neutral color. The plug-in will then adjust the color of the selected area to be a neutral gray (R=G=B).

After the white balance coefficients are calculated this way the results are displayed in the edit boxes. The white balance mode will then be switched to Custom. You can use the Save button and store this white balance as a new entry in the white balance menu. Then you can use it for other pictures taken in the same conditions.

You can achieve the best results if you select a gray area that is not too dim, and not too bright. Selecting an almost black or an almost white area will not give you the most accurate result. Look for and select mid-tones.

Tip: Neutralize shortcut

There is a shortcut for the neutralization feature. You can hold CTRL (Command or Option on Mac) and click inside the preview image, without clicking the Neutralize button first. Now you can rapidly try different locations in the picture, and select the one you like the best.

Tip: The precision of neutralization

Sometimes it is hard to neutralize the image as the white balance coefficients are calculates from single pixel you click on. When the area you are clicking on is contaminated by noise then you can get weird results. There is no way to specify the amount of the surrounding pixels to be taken into calculation in this version of the plug-in. But there is workaround. You can switch the size of the preview image to a smaller size (e.g. 1/16). In this mode every pixel of the preview is an average of more raw pixels, and therefore the effects of noise and local irregularities are minimized. This will result in a better calculation of the white balance coefficients.

Temp & CCIndex

Use can use the Temp slider to correct the white balance of your image by shifting it warmer or cooler tone. The values of the light temperature are the same as displayed by DiVU and they are calculated using the calibration factors specific for your own camera. This means that the R,G,G’,B white balance coefficients calculated from given light temperature are unique to your own camera only. If you would enter the same light temperature when processing the mrw file from other camera then the resulting coefficients would be different although the look of the image should be the same.

The CCIndex slider should be used to shift the colors from green to magenta. At this moment I do not know the formulas necessary to calculate the white balance coefficients from CCIndex and therefore this slider is not enabled.

Saturation

You can saturate or de-saturate the colors of the image with this slider and edit box. When you set the slider to the minimum value -70 you will get a black & white image. Please use the saturation slider with care for adding too much saturation can ruin the picture.

The MRWFormat plug-in does this saturation adjustment after all color conversions and interpolation. So it doesn’t matter if you adjust saturation here or in Photoshop. The results will be the same.

If you are using color managed workflow then make sure to adjust the saturation only with the check box Color manage enabled. Otherwise the colors of the final image will be different in Photoshop.

Warning!

Color management is a very complex area, and there is lot confusion about it. Before you start experimenting with ICC profiles please make sure that your monitor’s ICC profile is correct, and check the color management settings in Photoshop. There are very nice step-by-step instructions on this web site http://www.computer-darkroom.com/ explaining how to set-up your computer. Bruce Fraser’s articles on http://www.creativepro.com/author/home/40.html are definitely also worth reading.

 

 

Exposure

This slider allows you to adjust the exposure of the image in a vast range (± 8 EV). When you move the slider the effect can be seen immediately in the preview window. Also the actual exposure correction will be displayed in the edit box in units of exposure (EV or f-stops). This control has the same effect as the exposure correction control built in the camera. Obviously you can use this to fix under or overexposed pictures.

Remember that increasing the exposure merely multiplies the values of the raw pixel data by some value. So if you raise the exposure then you are effectively increasing the values of the noise pixels too. This makes them more visible. And vice versa decreasing the exposure darkens them and effectively makes them seem to go away.

Also remember when you increase the exposure of an image you will loose some detail in highlight areas, because you are pushing some values above the maximum value allowed. Remember you can see which areas you are being clipped by enabling the Warning (pre) check box.

Another problem which will exhibit when you will reduce exposure is described in the section Highlights fix.

ISO Pushing

This is an interesting technique proposed by Russell Bozian. You can use it for photos taken in dark locations without tripod. If the camera wants to use a long exposure time, and you are afraid of ruining the picture from handshake, then select a shorter time in M mode and underexpose the picture. You can later correct the underexposure using this plug-in and the Exposure slider. The result would be almost the same as if the camera was equipped with ISO 1600, 3200, … Of course the resulting picture will be very noisy, and you will have to use a noise removal filter on it.

Exposure Max

This button sets the exposure to such a level that no clipping of the highlights will occur (with the given white balance). To make this button selectable you must open the preview window otherwise it stays grayed out. The sequence would be: click the Preview button, and select the correct white balance for your image (as described above). Then click the Max button and the plug-in will select the maximum possible exposure that can be used without any highlights being clipped. This will usually result in the exposure being lowered.

You can use this exposure setting as a starting point for finding the best exposure. I suggest you enable the Warning (pre) check box to see what areas you are going to clip, and then raise the exposure slowly and see the effect on the image. You have to find a reasonable balance between good exposure and highlights clipping.

Contrast & Gamma mode

The plug-in has three different modes for adjusting the contrast and gamma of an image. The first one is called Precise and then there are two modes called DIVU ±3 and DIVU ±5.

DIVU ±3 – In this mode you can select the contrast with slider in the range of -3 to 3. This setting will produce the exact same result as processing the image with Minolta DiMAGE Viewer (DiVU). The gamma is fixed in this case, and you cannot adjust the Shadows and Highlight slider. This is because the image processing curves are exactly the same as those used in DiVU and therefore no modification is possible.

DIVU ±5 – This mode is the same as the above one. This time you can set the slider in the range of -5 to 5. The curves used to process the image are the same as used by DiVU for processing the images from A1 and A2 cameras.

Warning!

You should avoid using the DIVU modes as there are serious bugs in DiVU’s math. The curves used in DiVU are not smooth and can decrease the quality of the picture, because they actually introduce more noise. Also, they reduce the dynamic range of the image slightly. The only advantage to using them is that the resulting image is the same as if it was processed through DiVU. I suggest that you use them just to see the results of a DiVU conversion, and then switch to Precise mode and find the settings that match the result of DiVU conversion closely.

 

Precise – This mode gives you complete control over the curve that is used to process the image. The curve is calculated according to the settings of the sliders. (i.e. Contrast, Gamma, Shadows, and Highlights). You have much more flexibility in this mode and you can adjust all the controls in very fine steps (read below for more detail).

Tip: understanding the curves

You can see the actual curves used to process your image in the Histogram window. Display the histogram by clicking the Info button, then enable the Show curves check box and play with the sliders. The curves are colored red, blue, and green. This is a very good way to learn the effect of these controls as you can immediately see the effect of every slider. It is helpful to set the Gamma to 1 as the effect will be better visible.

Contrast

The slider labeled Contrast is used in to alter the contrast of the image. The default value of 50 was chosen to match the results from DiVU.

Gamma

The Gamma slider controls the gamma of the image processing curve. This control can be thought of as a brightness control. The default value is 2.64. Again it was chosen to match the DiVU result.

There is one technique of post processing which starts with linearly processed raw files. If you want to use this method of post processing then set the gamma to 1 and the contrast to 0.

Shadows & Highlights

These sliders enable you to modify the darkest and the brightest parts of your image. The Shadows slider defines the black point of the image (will make shadows darker). Similarly the Highlights slider defines the white point (makes the highlights brighter). This control adjusts all the color channels by the same amount so the overall color relationship in the image is preserved.

For example, entering the number 2 in the shadows edit box sets the threshold so that every pixel value lower than 2% of the maximum range will turn to zero (the threshold value in this case is 0.02*4095 = 81).

Even though these sliders are very similar to Photoshop’s Levels command their affect is dramatically different. The processing is done before the interpolation of the image. The adjustment of these sliders could have a huge impact on the final image (the same holds true for almost every control in the plug‑in).

Note: you can use the Shadows slider very effectively to get rid of the noise in the black background of some images.

Interpolation

You can select the desired interpolation algorithm from this menu. There are many algorithms built‑in because there is no one algorithm that would give the best possible result for every picture. There are some better ones though. For a description of the implemented algorithms see section Interpolation algorithms.

If you do not want to bother with selecting the interpolation algorithm then you should use Variable Number of Gradients or Color Correction I. Those algorithms are rated best for general use.

Demo version

The best algorithms are available in the DEMO version, but they will open just the middle area of the image. You can tell if an algorithm will run in demo mode by observing the Image Dimensions, the word (DEMO) will be displayed there. To remove this limitation you have to purchase the full version from SWREG. There is a link to the internet shop on my page http://www.dalibor.cz/minolta/plugin_order.htm

Extra size

The size of the interpolated image is displayed below the Interpolation menu. The original RAW image of DiMAGE 7Hi has 2568x1928 pixels, but because some of them are lost during the interpolation, the resulting size of the image is less. Normally the images from the camera are interpolated to 2560x1920 pixels. But some interpolation algorithms are able to produce larger images. If you enable the Extra size check box then MRWFormat will create the largest possible image. You can see the dimension of the image displayed below the Interpolation menu. For the compulsive among you some interpolation algorithms can give you up to 31 thousand more pixels that DiVU! And with DiMAGE A2 it is even more - 40 thousand more pixels!

Warning (pre)

One of the problems with digital photography is clipping of the highlights. The AD (analog to digital) converter built in the camera has a limited range. For instance, when you take a picture the camera tries to capture as much of the scene as possible, but the camera’s range is much narrower than the dynamic range of the human eye therefore some of the information is lost. If there were some areas in the picture that were too bright to capture because they exceeded the limits of the camera, the details in these areas will be lost, and the pixel values will be set to maximum. After the picture has been recorded there is no way to correct the blown out highlights, because these details were NEVER recorded.

When you click on any pixel in preview its values will be displayed in the title bar of the preview window (only in Windows version). The first set of the raw values will be followed by exclamation mark in case that the corresponding color channel was “blown” at this pixel.

There is another step in image processing that causes blown (lost) highlights. It is the process of white balancing the picture. Imagine that there is a bright object in the scene having the raw pixel values R=2750, G=G’=3965 and B=2959. And imagine the white balance coefficients, which are correct for this picture and produce a good result, are WB_R=1.8, WB_G=WB_G’=1 and WB_B=1.4. Then after white balancing the values become R=2750*1.8=4950, G=G’=3965*1=3965 and B=2959*1.4=4142. Now we have the value of R and B over the maximum allowable value of 4095 thus the resulting excessive values are clipped to the maximum level; R=4095, G=G’=3965 and B=4095. So you can see that one will lose some information just by white balancing the picture!

The same logic applies when you are increasing the exposure of the image by using the Exposure slider (see below). When you increase the exposure by 1 EV you are multiplying the pixel values by two, and some of them will be larger than the maximum allowed value, and clipping will occur.

The areas of highlights that would be clipped by white balancing and/or exposure adjustment can be displayed by checking Warning (pre). This function substitutes a zero for every pixel value larger than the maximum allowed. So the areas where all three color channels are clipped will turn to black. The areas where only one or two channels are clipped will turn to a different color (blue, magenta, red, …) depending on which channels are clipped. This allows you to quickly see where details will be lost in the highlights. The size of these areas is affected by the color balance and exposure settings only. No other controls have any effect on this.

Tip: keyboard shortcut

You can quickly change the state of this pressing the key H or W.

Highlights fix

As you probably know one of the biggest problems of digital camera pictures is blown highlights. The camera is not able to capture all the range of light intensity between the dark and highlight areas in the scene due to its limited dynamic range in many cases. Therefore it sets the exposure to capture the details in the midtones and the details in highlights are lost. When the data from the CCD are digitized then all the values higher than the maximum value are clipped to this maximum. If you have exposed the picture properly than this loss of detail in highlights is acceptable and you usually do not expect to see anything there. But if you see when processing the raw image that the exposure you have chosen was too high and you want to decrease it MRWFormat then problem will arise.

In many cases not all the details from the highlights are lost. Usually one of the color channels was clipped sooner than the other two. From this point of view there are three different areas in the image. In the first one all the pixel values were under the maximum threshold and the details were captured fine. Another area is where all the three channels are clipped. This area is white only and nothing can be recovered from here. But there is a third area in between of those two. In this area there is only one (green usually) or two channels clipped, containing no information but some of the details is preserved in the other channels. This area will cause problems when you try to reduce the exposure. You can see this area when you set Exposure to zero and enable the Warning (pre) function.

The check box Highlights fix controls the way MRWFormat deals with partially blown highlights when you are reducing the exposure. If you leave exposure set to zero or you if you increase it then this check box has no effect. If you keep this checkbox unchecked and start to reduce the exposure then you will see that the pixels start to change the color in the area of partially blown highlights. The reason is that the clipped channel(s) is not in the correct proportion with the un-clipped channel(s) which will create an imbalance in the color and this leads to magenta cast usually (the actual color cast depends on the white balance setting of your picture).

If you enable the Highlights fix checkbox then MRWFormat will clip all the channels proportionally. This means that the color of the blown highlights will stays consistent with the rest of the picture. On the other hand this will lead to the loss of the details which were contained in the un-clipped channels.

To summarize it: When you are reducing the exposure of a picture containing the blown highlights you have two options to choose from. Either you keep as much detail as possible but you introduce a color cast in those areas or you eliminate the color cast at the cost of removing the details from affected areas.

It seems that it would be possible to add a new function in the future versions of MRWFormat called Highlights recover. This feature would be able to try to guess the missing values in the clipped channels by observing the detail in the un-clipped channels. Then you would have a third option in this situation and MRWFormat will be able to recover the details in blown highlights area to some extent.

Note that Warning (pre) has a higher priority for the display than the Highlights fix so you will not be able to see the effect of Highlights fix in the preview window if both checkboxes are enabled at the same time.

Tip: keyboard shortcut

You can quickly change the state of this pressing the key F.

Warning (post)

Application of the curve (which is defined by the parameters set with the Contrast, Gamma, Shadows and Highlights) could also lead to some loss of image information. In the example where one set the shadows to a value of 2, all the pixel values in range from 1 to 81 would be set to zero. You can use Warning (post) to check the areas of the image you are affecting by the curve. If this check box is enabled then all the pixels having minimal value 0 or maximum value 255 (or 65535 in case of 48-bit colors) are displayed with inverted color.

You should note that even the unprocessed image can contain pixels with minimum or maximum values. So the fact that some pixels will be highlighted when you enable this feature doesn’t mean that the picture itself was bad from the beginning. And vice versa not all the pixels are highlighted as an effect of your adjustments. You should enable this warning just before you start adjusting the sliders and watch the effects of your adjustments.

You should first select the required Color Depth before you examine the Warning (post) because the result of the warning feature is different in 24 and 48 bits.

Color Depth

The plug-in is capable of working in two color depth modes. Either it will process the resulting image to have 24 bits colors (8 bits per channel) or 48 bits (16 bits per channel). The difference between these two choices is hardly noticeable if you compare the images in Photoshop. But, if you plan on doing intensive editing of the picture afterward then you should choose 48 bits as the final result might be better. Otherwise use 24 bits for it is enough in most instances, and the image will convert faster.

Color Space

The colors of an image are adjusted during processing. There are two known methods of color conversion: the original one used for DiMAGE 5 and 7 cameras (also 7u), and the new one for DiMAGE 7i and newer cameras. DiMAGE Viewer (DiVU) does this adjustment automatically when it processes an image, and you are not given any choice. In this plug-in, however, you can choose what method to use. If you do not like the results you are getting with the default method for your camera then you should choose the other methods to see if they will give you more pleasing colors. As the general ICC profiles for digital cameras are vastly simplified there is nothing wrong with processing A2 images with D7i profile if you like to results better.

The other meaning of this setting is to define what ICC profile will be attached to the image processed. My plug-in does use modified profiles based on the original profiles supplied with DiVU. This modification was suggested by Jeff Yigdall.

When you are evaluating the results of the various settings of Color Space then you should have the checkbox Color manage enabled. Otherwise the colors you will see in Photoshop later will be different.

 

 

 

Default – Chooses the color conversion and ICC profile depending on what camera was used to take the picture. If you are not satisfied with the result you can override the selection by selecting one of the following options.

D5 / D7 – This option forces the old color conversion used for D5, D7 and D7u cameras. The D7 ICC profile will be attached.

D7i / D7Hi – This will force the new conversion and the ICC profile for D7i will be attached.

A1 – Similarly this option forces the new color conversion and the A1 ICC profile will be attached. Using the A1 ICC profile does result in slightly more saturated colors then with D7i/D7Hi.

A2 – Similarly this option forces the new color conversion and the A2 ICC profile will be attached. The main difference to A1 is in deep blue areas.

None – When this option is selected then absolutely no color conversion is done. The color conversion is processor intensive task therefore the plug-ins reaction will be much faster when it is off. But when it is disabled then the Color manage check box will be unavailable and the Saturation slider as well.

Color manage

When you enable this check box then the colors of the preview will be color managed by the internal color management module. If you do use a color managed workflow then you should enable this check box otherwise the colors of the image shown in preview window will be different to the colors displayed in Photoshop later.

Color management is relatively complicated area and a lot of people make mistakes here. In order to get good results please do read the section Color management.

The color management in MRWFormat is done on the fly and it requires a lot of processor time. The reactions of the user interface will be slowed down considerably when Color manage is enabled. I recommend enabling the color management of the preview window only at the times when it is really necessary. You can do the contrast, gamma and other adjustment without it. But when you are white balancing the image, adjusting saturation and Color Space then you should enable it.

When Color Space is set to None then you cannot enable color management of the preview window. The reason is that the colors are undetermined from the colorimetric point of view and therefore they cannot be managed.

 

Tip: keyboard shortcut

You can quickly change the state of this pressing the key M.

Interpolation preview

With this box checked you will see the results of the various interpolations as each type is selected. If the preview window is displayed, and you enable the Interpolation checkbox then the preview window will split into two parts. In the upper part is the normal preview image, but its size is reduced, and in the lower part is the interpolated image. If you click in the upper (preview) image a blue box is displayed. The area contained by this blue box is the interpolated part of the picture that is displayed in lower section of the display.

If you want to test the interpolation of another part of the image, either click inside the preview image or click-drag the mouse inside the interpolated image (the interpolation preview is redrawn after you release the mouse button).

To display the RGB values of the pixels in an interpolated image just click on them (Windows only).

Note: you cannot use the Neutralize feature in the interpolation preview area.

Tip for speed

The interpolation preview slows down the responsiveness of the user interface. When choosing the best interpolation algorithm it is best to fine-tune all the sliders first (gamma, shadows, etc.) without interpolation preview selected and then select the interpolation preview.

 

Zoom

Because the differences between the various interpolation algorithms are very subtle it is sometimes hard to see them in the preview window. To help with this there is a Zoom checkbox. Selecting it will cause the interpolation image to be enlarged. This will help one examine the results of the interpolation algorithm very closely.

You can also change the zoom ratio by clicking inside the interpolation image while holding Alt or Ctrl. Ctrl+click zooms in and Alt+click zooms out. (On Macintosh Command+click , and zoom out using Shift+click or Option+click.) You can use the keys + and – to zoom in and out as well (on Windows you have to hold Alt or Ctrl pressed). The current zoom ratio in range 100% to 1600% is displayed as a label of a Zoom checkbox.

Average Interpolation

As this interpolation does produce the images of half width and half height compared to other interpolation methods the zoom ratio is automatically doubled when you select this interpolation. This allows easier comparison to the other algorithms.

White balance

As I have explained before the digital camera does behave in very different way than a human eye. When you will be observing the white sheet of a paper in direct sunshine or under tungsten or fluorescent light the paper will be always white. The reason is that the human brain does compensate the different color of the light. Digital cameras do not have such ability and the color of white paper will be different under the different light. This means that digital cameras must be able to simulate the behaviour of the human brain and change the colors of the image to look as we expect. To archive this digital cameras change the ratio of the R, G and B color channels in the image. This process is called white balancing. The process of finding the correct white balance for the given image is very easy task for a human brain but it is very complicated task for the camera and the algorithm built in the camera can be fooled very easily giving you wrong results.

One of the biggest advantages of shooting in raw is that the white balancing is done during the post processing of the image and you can choose try several different ways of white balancing the image and then choose the best one. At the end of the white balance process you will find four numbers called white balance coefficients which specify the correct rations of red, green and blue colors. The coefficients for the first and second green cells should be equal and they have the value 1 in most cases. The correct white balance is achieved by altering the R and B channels only. MRWFormat does display the actual coefficients use in the four edit boxes named R, G, G’ and B.

There is a white balance drop down menu which stores all the white balance coefficients you can use. You can define your own white balances and add them to this menu as well using the buttons Save and Delete below this menu.

There are several ways how the four white balance coefficients are calculated and how you can work with them in MRWFormat.

Camera setting

This is the white balance that was set in the camera when you took the picture. As it is the most used type of white balance it is always the first entry displayed in the white balance menu. The actual white balance mode set on camera is displayed after the words Camera Setting in brackets. If the camera was set to Auto white balance then Camera Setting (auto) will be displayed. This means that when the camera was taking the shot it was examining the scene trying to guess the correct white balance. This white balance is recorded inside the mrw file but it is not applied on the image data as it is done when you shoot JPEG or TIFF. When you select this white balance MRWFormat will read the coefficients from the mrw file display them in corresponding edit boxes and apply them on the picture.

Of course you can set the camera to one of the fixed white balance modes. Depending on you camera type those are Daylight, Cloudy, Tungsten, Fluorescent, Shade and Flash. On newer models there are several variations of those fixed coefficients, for example Daylight -3 or Tungsten 2. If one of those coefficients was selected than its name will appear after the Camera Setting in brackets as well. Then there is a hidden white balance called Sunset. You are not able to choose this preset in the white balance menu of the camera but you have to use Digital Subject Program instead.

The last variant is the Custom white balance. This white balance is measured using the white sheet of paper and it is the best method to white balance the scene properly.

All those white balance coefficients are defined as four numbers R, G, G’ and B. The Auto and Custom settings are variable of course as their values depend on the light in the scene. The other settings (Daylight, Cloudy, ...) are constant but they vary from camera to camera.

Every camera is a little bit different because of variances in the manufacturing process. Cameras are calibrated in the factory to produce the same result. This is probably done by taking a series of pictures under different lightning sources, and then the white balance presets are calculated and stored in the camera memory. Unfortunately there is no known way to read those presets from a camera. Therefore if you are satisfied with the result you are getting when you switch your camera to for example Daylight mode and you want to you the same set of white balance coefficients in MRWFormat then you have to copy this preset to MRWFormat. I will explain how to achieve this later.

Tip for processing a series of pictures taken under the same lighting

Browse all the pictures taken in the same lighting situation and find the one that has the best white balance and the most the realistic colors. Then store the white balance of the chosen picture as a user defined white balance. Then use the stored white balance for every picture in the series.

 

Tip for using camera white balance setting

It makes no sense to select any of the fixed white balance presets when you shoot in raw mode. It is better to set the camera to Auto mode. This way the camera will try to decide what the correct white balance is and it will store its guess into the mrw file usually creating a good starting point. Even better is to use the white card and measure the white balance using Custom white balance function. As only one white balance can be stored in the mrw file by the camera it is wiser to let it write the measured white balance in there and not one of the fixed white balance presets as their values are the time for your camera all the time.

Custom white balance – defined by coefficients

This white balance is defined straight by the four white balance coefficients for the R, G, G’ and B color channels. It is the second entry in the white balance menu and it is called Custom.

When you use the Neutralize function then the white balance menu will switch to the Custom automatically and the white balance values calculated as the result of Neutralize will be saved into this preset. Also when you enter or edit the white balance displayed in the R, G, G’ and B edit boxes manually then MRWFormat will switch to the Custom as well.

There is only one custom preset so if you use the Neutralize function again ,the new results will overwrite the old ones, unless you save it as a user defined preset first. If you want to save the custom white balance as for the future use then please click Save button below the white balance menu. The dialog window will appear asking you to give the name to newly created user preset. The predefined named is New white balance and you should change it to something reasonable. Then this preset will be added as the last entry in the white balance menu and you can recall the stored set of the four white balance coefficients any time later.

Custom white balance – defined by temperature

This white balance is similar to the previous one but this time it is defined by color temperature (and CCIndex in the future). If you use the Temp slider or enter some temperature directly into Temp edit box then the white balance mode will change to the Custom Temp mode and it will be indicated in the white balance menu as well.

As I have mentioned earlier, every camera is a little bit different and they are calibrated in the factory. The result is that when you select the color temperature 5500 degrees Kelvin for two pictures from different cameras then the R, G, G’ and B coefficients will not be the same. MRWFormat is able to calculate precisely the white balance coefficients from entered temperature for every piece of camera. So you can create a user defined preset name Custom Temp 5500 and when you use it for your pictures, the calibration of your camera will be taken in account and if you process the picture taken by your friends’ camera and use the same preset then the coefficients will be recalculated to match this other camera giving you the same result.

Therefore if you change the white balance of the picture by using the Temp slider and then you click Save button to store this preset, MRWFormat will save not the four white balance coefficients but it will save the entered color temperature instead. To emphasize this fact the predefined name in the save dialog will be Custom Temp xxxx.

When you select this created preset from the white balance menu the next time then MRWFormat will first determine the correct calibrations factors from the mrw file and calculate the four R, G, G’, B coefficients accordingly.

White balance stored in the raw file

This is the last type of white balance preset. In every raw file are stored six different white balance presets. Those preset have the constant values for one camera but they vary from camera to camera. Those presets are listed in the white balance menu as Daylight (raw), Cloudy (raw), Tungsten (raw), Fluorescent (raw), Flash (raw) and Unknown (raw).

Those coefficients are used by DiVU (DiMAGE Viewer) when processing the raw pictures. DiVU displays a dialog called Open Minolta‑RAW file where a white balance box lists the options. If you choose one of the presets then DiVU uses one of those (raw) coefficients. Please note that there is not 100% match between the presets in DiVU and in MRWFormat. DiVU’s behaviour was changed several times during its development and it depends on your camera type as well. But the basic principle is the same.

For some strange reason those coefficients are usually not exactly the same as the coefficients stored in the camera. This means that when you take a picture with your camera’s white balance set to Cloudy and you then open the picture in MRWFormat and then you will probably find that the white balance Camera setting (cloudy) is different to Cloudy (raw) although the difference will be quite small.

If you think that those (raw) coefficients have no use for you then you can disable them in Preferences and they will not appear in white balance menu.

Interesting fact

Those six white balance coefficients stored in every raw picture are different with every camera so they could be thought of as a “camera signature”. I have never seen two cameras with the same “signature”. Heads up you cloak and dagger types: a public .mrw file can be matched to YOUR camera. If stealth is your modus operandi, change the file to .jpg before going public.

Duplicating fixed white balance presets from your camera

As I have mentioned earlier the fixed white balance presets you can set in your camera are unique to your own camera. If you are satisfied with them then you might want to duplicate them into MRWFormat so you can use them for any raw picture. Unfortunately MRWFormat is not able to determine them automatically for you but it will help you to get them as smooth as possible. To duplicate the fixed presets from your camera to MRWFormat please follow these steps:

1.       Switch your camera to RAW mode

2.       Set the white balance to Daylight on your camera

3.       Take a picture (It doesn’t really matter what will be on the picture. The lens cap can stay on if you wish).

4.       Set the white balance to Cloudy on your camera

5.       Take another picture

6.       Repeat for all the fixed white balance presets you want to duplicate to MRWFormat. Don’t forget that Sunset is fixed white balance preset as well. But you have to use Digital Subject Program to select it. It usually makes no sense to duplicate the Cloudy +3 to Cloudy -3 variants if your camera has them. The only preset where capturing the variants makes sense is Fluorescent.

7.       Download all the pictures to your computer

8.       Open the first picture and select Camera Setting from the white balance menu

9.       Click Save button below the white balance menu

10.    A dialog box will open and the preset will be named automatically according to the white balance mode set on the camera so just click Save again

11.    Press OK to process the picture. Do not click Cancel as MRWFormat will not save the newly created preset if you would cancel the opening of mrw file.

12.    Repeat for the other pictures taken

Now you have taught MRWFormat all the camera white balance presets you wanted to. As it is quite boring procedure I do recommend finding the .ini file where those newly created presets are stored and back it up on the safe place. The .ini file is named MRWFormat.ini and it is stored in the same directory as the plug-in itself is on Windows. Usually it is c:\Program Files\Adobe\Photoshop XX\Plug-ins. On Mac it is MinoltaRaw.ini and it is kept in Library->Preferences in your Home folder. If you are not able to locate it then just do a search on your hard drives for MRWFormat.ini or MinoltaRaw.ini

Processing images from a camera belonging to someone else

If you have taught MRWFormat the set of the coefficients to match your camera (which you should do if you are following my instructions), and you need to process the pictures from someone else’s camera then it is best to use the white balance settings named Something (raw) . Those are the ones embedded in the raw image. This will usually produce better results for they are the ones stored with the picture itself from the camera that made the picture. Remember each camera is uniquely different and without a basis set taken from your friend’s camera you will be “shooting in the dark” as the Americans say.

Delete white balance preset

Any of user defined white balance preset can be deleted as well. Just select the preset you want to delete from the White balance menu and then click the Delete button. The preset will be deleted and the last one from the menu is selected. You cannot delete any of the following: Camera setting, Custom and all the (raw) presets.

There is no confirmation dialog, but if you make a mistake then you can just click Cancel in the main dialog and the white balance preset will not be deleted. The reason is that all the changes are saved in the .ini file only when you click the OK button.

Presets

All the settings you can adjust in MRWFormat which affect the look of the image produced can be saved into so called presets. If you like the result you are seeing in the preview window but you want to try slightly different setting to see if it will not be even better then you can save the current settings under chosen name. Then you can go on with trying to find better settings. If you then want to revert back to the previous state you can recall it by loading the stored preset. All this is done using the Presets menu located below the Preferences button.

To save your current setting as a new preset just click the arrow by the text Presets. The pop-up menu will appear showing you the currently stored presets (if there are any). In the lower part separated by the line are displayed the commands to save and delete presets. Select Save As… from this part of menu. The Save Preset dialog will be displayed and you can assign a name to this preset. When you do so and click Save, the newly created preset will be added to Presets menu for future use.

To load one of the presets you have stored before, just open the Presets menu again and click one of the presets listed. The preset will be activated and the bullet sign will appear beside its name. Be careful when loading presets as there is no confirmation and the current settings of sliders will be lost. You might want to save the current settings as a new preset before loading the saved one.

When you have loaded one of the presets and you change some of the settings in MRWFormat then Save command in Presets menu will be enabled. If you will think that the setting you just changed is better than the one stored inside the preset then you can save the current settings back to selected preset using this Save command.

You can also delete the currently selected preset by using the command Delete.

All the changes in presets will be saved into the .ini file and they will be available next time you will start MRWFormat. But you should know that the .ini file is saved only when you close the MRWFormat dialog using the OK button. If you make some changes in preset and then you close MRWFormat by clicking the Cancel button then all the changes will be lost. If you want you can save the current settings into the .ini file by opening the Preferences dialog and clicking Save button there which will save all the changes immediately.

When you process the image using the OK button then the preset used for this mrw file is sent to Photoshop scripting. The reason for this is that Photoshop can record the settings used into an action. This action can be then used by a Batch or Droplet functions. When MRWFormat plug-in is started by Photoshop next time to process another image then Photoshop will send back the last used settings to MRWFormat. Therefore all the sliders in MRWFormat will have the same values as the last time you have pressed OK. This behavior is fine for some users and it is welcomed when you process a series of images taken in the same conditions.

However some users do prefer to start with the same setting all the time and find the best setting for every image separately. If this is your case then you can set MRWFormat to disregard the parameters set by Photoshop and load the default values for every image at start. This is controlled by the checkbox Set parameters to default at start in the Preferences dialog. When this checkbox is enabled then before the main dialog is shown, MRWFormat does search the stored presets for the one named Default . If the preset with this name is found then it is loaded automatically. If there is no such preset then the “factory” defaults hard coded in the plug-ins is used.

Therefore if you want to override the default settings for the plug-in then set all the sliders in MRWFormat as you wish. Then open the Presets menu and click on Save As… command. Enter the name Default in the Save Preset dialog which will appear. Do not forget then to process the image so the newly created preset is saved into the .ini file.

As I have mentioned before only the parameters directly affecting the look of the final image are stored in the preset. To be exact the stored parameters are:

·         White balance (can be defined as camera setting; by R,G,G’,B coefficients or color temperature)

·         Exposure

·         Saturation

·         Contrast (either precise or DiVU)

·         Gamma

·         Shadows and Highlights

·         Interpolation (including Extra Size option)

·         Color Depth

·         Color Space

·         Highlights Fix

·         Clip values > 3965

·         Convert processed image to monitor color space

The last two settings are set in the Preferences dialog but they are saved into preset. All other settings are “global” and they are common for all presets.

Color management

Color management is relatively complex area and explaining it in detail is beyond the scope of this user guide. It is very easy to make a mistake in the settings of the color management options in Photoshop and MRWFormat. I can give you here only a basic guidelines which you should follow. If you want to learn more about ICC color management then you should be able to find a lot of information on internet or in the bookstores.

Please do not overestimate the significance of ICC profiles for digital camera. It is not possible to create a precise universal ICC profile, which would be suitable for all lightning situations. If you need a very precise ICC profile then the only way to get it is to take one picture of a color chart, and use a specialized tool for creating ICC profiles. Then shoot in that same controlled environment (studio). The ICC profile created probably would even then only be valid for that session.

Setting up Photoshop

Photoshop is very powerful tool with advanced color management options. It is very easy to set them wrong and then complain about the colors of your pictures. To get good results the following requirements must be met:

·         Color manage check box in MRWFormat must be enabled.

·         You have to let MRWFormat to attach the ICC profile (in MRWFormat Preferences dialog).

·         Photoshop must be set correctly to respect the embedded profiles.

·         You have to set the correct profile for your monitor in your operating system.

When you process the image in MRWFormat then Photoshop does present you with the following options: Use the embedded profile, Convert document’s color to the working space and Discard the embedded profile. The option you should choose here is Use the embedded profile. If you have a good reason then the second option could be used as well, but it is better to avoid any unnecessary conversions. You should never use the third option.

If you do not want to answer this question every time you process the raw file then choose Edit->Color Settings from the Photoshop menu. In this menu set the option Color Management Policies – RGB: to Preserve Embedded Profiles. Also uncheck the Profile mismatches: Ask when opening.

Always embed the ICC profile when you are saving the post-processed image from the Photoshop into the TIFF or JPEG file.

If you have set all the above correctly then the colors you are going to see in MRWFormat preview will be exactly the same as seen in Photoshop.

Setting up Photoshop Elements

Adobe has decided to limit the color management features built-in Photoshop Elements. The bad news is that you cannot convert your image into another color space. The good news is that you do not have to pay attention to the color settings of Elements as they cannot be set wrong. So only the following has to be met:

·         Color manage check box in MRWFormat must be enabled.

·         You have to let MRWFormat to attach the ICC profile (in MRWFormat Preferences dialog).

·         You have to set the correct profile for your monitor in your operating system.

Always embed the ICC profile when you are saving the post-processed image from the Photoshop into the TIFF or JPEG file.

If you have set all the above correctly then the colors you are going to see in MRWFormat preview will be exactly the same as seen in Photoshop Elements.

Intended use of the image

It is very important to choose the correct color space for your images based on their intended use:

·         Archiving the images
If you are not sure for what purpose the image will be used then it is recommended to keep it in the color space assigned by MRWFormat and save it in the TIFF file with the profile attached. You can always open this image in Photoshop later and convert it into proper color space.

·         Printing on cheap ink jet printer
Many cheap ink jets printers assume that the image you are printing is in sRGB color space. Therefore to get a good print you have to convert your image into sRGB color space first. You can check if your printer does expect sRGB data easily. Print one image from Photoshop as it is and then convert the image into sRGB and print it again. If the colors of the second print will be better then you know that you should convert your images before printing to sRGB space.

·         Publishing on the web
If you are going to publish your image on the internet then it has to be in sRGB color space. So you have to convert the image to this color space before uploading it on the internet.

·         Viewing on the monitor
If you are using Windows then you have to pay attention to the fact that the vast majority of the picture viewing applications are not ICC profiles aware. They will read and display the image without taking the attached ICC profile in account. Therefore the colors you will see in Photoshop will be very different from the colors displayed by ACDSee, IrfanView, xnView or other software. To see the good colors you have to convert your images into sRGB space.
On Macintosh is the situation different. The color management is built inside the operating system and therefore all the applications on Macintosh should display the correct colors according to the ICC profile embedded in the image.

Conversion to sRGB space

If you are using the full version of Photoshop then this is very easy task. Just open your image and then select from the menu Image->Mode->Convert to profile...and select sRGB as the Destination Space. Then save the converted image and name it for example PICT0032srgb.jpg.

Of course one may think that it is better to convert all the images into sRGB immediately after processing them by MRWFormat and work in sRGB workflow only. Actually this is a quite acceptable approach giving good results. But there is one drawback in it. sRGB space is intended for viewing on monitor, publishing on the web and printing on cheaper printers. But the expensive printers are able to print more colors than those covered by sRGB space. If you convert your image to sRGB then those colors lying outside sRGB gamut are lost forever. Therefore I do recommend archiving your images in the original camera color space as it will save all the colors in your image. You can then create a copy in the sRGB space when necessary.

If you are using Photoshop Elements than the situation is little bit complicated. When the image is processed by MRWFormat then the ICC profile describing the camera color space is attached to it and it will be respected by Elements, but you will have no chance to convert it into sRGB space. So if you need to have your image in sRGB space for publishing on the web or printing then you have to set this in MRWFormat. There is a checkbox named Convert image into monitor color space in MRWFormat Preferences dialog. When this checkbox is enabled, the image will be converted into the color space similar to sRGB before it will be passed to Elements. So if you have no other way how to convert your images to another color space then just enable this check box and stay in sRGB only workflow.

Setting up color unmanaged workflow

If you want you can try to use the non-color managed workflow as it might give you pleasing results as well. You will probably have to boost saturation of the images in MRWFormat or later in Photoshop (Elements) as the images might be little bit pale using this workflow. To turn off the color management features you have to do this:

·         Uncheck Color manage checkbox in MRWFormat

·         Uncheck the checkboxes Attach ICC profile and Convert processed image into monitor color space in MRWFormat Preferences dialog.

·         Photoshop Elements – set the Color Settings to Limited color management

·         Photoshop – open the Color Settings dialog and set Working Spaces RGB to sRGB and Color Management Policies to Off.

 

When using this workflow all the images are thought to be in sRGB space which is standard for color management unaware users. They should look exactly the same in MRWFormat, in Photoshop, in ACDSee and in the internet browser. This workflow will probably make experts laugh or shake their heads but if it will give you pleasing results than it is good.

 

Preferences

When you click the Preferences button in the main dialog the following window will appear where you can fine tune the behavior of MRWFormat.

Dialog Orientation

Here you can select if you want the main dialog to be oriented horizontally or vertically. I find the vertical layout easier to use but it takes a lot of space and it might not fit onto your screen. If you want to switch to the horizontal layout then do so. As the change will be applied at the next start of the plug-in then please click Save button and then OK button. Click Cancel in the main dialog and open the same picture again. The dialog will be displayed in newly selected orientation.

Live preview

This check box controls how the preview window reacts when the user is dragging the sliders or editing the values in edit boxes. When the Live Preview box is checked the window is updated as you drag the slider with your mouse (or immediately as you change a value in the edit box). When it is unchecked the change takes place after you release the mouse (or after your cursor leaves the edit box).

Of course the live preview slows down the responsiveness of the sliders. So if you use a larger preview window size on a slower computer, it is best to turn Live preview off. For smaller preview window sizes it is better to keep it on.

Display Focus Area in preview

Some cameras store information in the EXIF data noting where the camera was focused while one was taking the picture. You can display this region in the preview window by enabling this checkbox. There will be a red box drawn on the preview image. If the camera was out-of-focus or the manual focus was used then the red box is drawn around the whole image. If this checkbox is grayed out, and cannot be selected then the image does not contain information about the focus area.

Attach ICC profile

If you enable this check box an ICC profile will be attached to the processed image, and passed to Photoshop. The choice of the attached profile depends on the Color Space setting. For D5/D7 the MLTDiM7.icm profile is attached. This profile was published by Minolta. Other profiles used are based on profiles supplied with DiVU slightly modified by Jeff Yigdall. In Default mode the choice of the profile depends on the type of camera that was used. In None mode the plug-in will attach one of the Jeff Yigdall’s profiles too.

You should have a very good reason to not attach the ICC profile to your images. If you do not attach it then the colors displayed in MRWFormat and in Photoshop will not match. Also you should let Photoshop use this profile attached. See the section Color management for more details.

All profiles included are quite general profiles, and I cannot guarantee that those are the best ones possible. If you have a profile that you like better then you can manually assign it later in Photoshop by choosing Image‑>Mode‑>Assign Profile…

Convert processed image into monitor color space

When this checkbox is checked then MRWFormat performs the same conversion on the final image as it does for color managing the preview image. In another words it converts the output data from the camera ICC profile to a standard monitor profile. The profile I use and attach is taken from Minolta DiVU. This profile is not exactly the same as sRGB but it is quite close. If you are using the full version of Photoshop then you should never enable this checkbox as you can use color profile conversion function in Photoshop which is more precise.

The only reason to use this checkbox is when you are using Photoshop Elements only. There are no color profile conversion functions in Elements. So if you do need to publish your images on the web or print in on standard inkjet printer then you could enable this checkbox and let MWRFormat convert the image to something close to sRGB. See the section Color management for more details.

Note that unlike the other settings in the Preferences dialog, this one is not global. It is saved in the preset.

Set parameters to default at start

When the MRWFormat dialog is shown Photoshop sets it to the last setting used. If you do not like this behavior then you can override it with this checkbox. When it is checked then MRWFormat will set all the parameters to default values. First it tries to find and load the preset called Default. If there is no preset named Default then it uses the “factory” defaults. For more details see section Presets.

Include WB presets read from raw file

This checkbox controls if the white balance presets located in every mrw file will be displayed in the White balance menu. If you never use those white balance presets, whose name ends with (raw), then you can clear this checkbox. See section White balance for more details.

Use internal raw thumbnails in browser

MRWFormat is responsible for generating the thumbnails of mrw files for Photoshop. When you choose File->Browse in Photoshop then Photoshop will ask MRWFormat to provide him with the thumbnail and the metadata (mainly EXIF tags) for every mrw file displayed.

There are two ways of generating this thumbnail. The first one, selected when this checkbox is unchecked, is to process the mrw file as usually and then resize it down and pass it back to Photoshop. Although I have paid a lot of attention to optimize this process still it requires a lot of calculation and therefore it is quite slow.

The much faster method is to use the JPEG thumbnail embedded in every raw file which happens when this checkbox is checked. There is a slight problem in this method. When you will take the image in RAW mode but you set the camera to Solarize or B&W mode then the internal thumbnail will be solarized or black&white. But as MRWFormat ignores those modes the image you will open will be normal and therefore different from the thumbnail. But as I do not believe that this will really bother anyone then you should keep this checkbox enabled to benefit from much higher speed.

Don’t use internal raw thumbnails for D5 and D7

Another problem related to the use of internal thumbnails is for images taken with DiMAGE 5 and 7 (this does not apply to the DiMAGE 7 upgraded to software version 2.xx, called DiMAGE 7u). Those cameras were not white balancing the thumbnails stored in the mrw file and the thumbnails have unrealistic colors. MRWFormat does try to restore the color but the results are not very pleasant all the time and the thumbnails generated using the slower method are better.

If you do check this checkbox then MRWFormat will use the internal thumbnails only if they do not need to be corrected. And if it finds that the image was taken with the camera with this problem then it will use the slower method of generating the thumbnail from by reading and processing the whole mrw file.

Users of DiMAGE 5 and 7 are advised to try both states of this checkbox to see if the colors of the faster thumbnails are acceptable for them or not. Do not forget that when you change the method of generating thumbnails you have to purge the cache of the Photoshop Browser to force it to regenerate the thumbnails.

Clip values > 3965

Normally the highest number of a pixel captured is 3965. But the camera occasionally captures a picture whose maximum value is higher then 3965, up to 4095. This is to allow for “vignetting compensation” that occurs when shooting with an aperture of f2.8. Yes, the camera vignettes at this aperture. The vignetting is hard to see in the corners because the camera multiplies based on a mask to brighten the corner pixels. After this compensation there might be some pixels with values greater than 3965 (those are the pixels in the corners that were close to maximum before multiplication based on the mask). The mask looks like this:

This checkbox allows you to set all the pixels with values higher than 3965 to 3965 and observe the difference. Though it is very hard to find a picture, and the corresponding white balance in which you could see the change. You might set the contrast to higher values to emphasize the effect. Anyway consider this check box experimental and leave it un-checked.

Note that unlike the other settings in the Preferences dialog, this one is not global. It is saved in the preset.

Enable logging into text file

If you enable this checkbox then a text log file will be created for you. MRWFormat will store the information about every processed mrw file into this log file. The following information will be recorded as well:

·         the original filename and the full path to the mrw file

·         the date and time

·         the version of the MRWFormat plug-in used

·         all the setting affecting processing of the file (in other words text dump of the preset used)

·         the name of the ICC profile attached

·         the time of the processing in milliseconds (not counting your own interaction with the plug-in)

The log file is named MRWFormat.log and it is stored in the same folder as the .ini file.

Open log

If you have enabled the logging feature and the log was created already then this button will be enabled. If you click it your default text editor (Notepad usually) will open displaying you the contents of the log file.

Delete log

You can delete the log file, if it exists, using this button.

Resolution

Here you can set the resolution which should be passed to Photoshop together with the image. Photoshop uses this number to determine the default printing size of the image. The value 300 DPI is used for high quality offset printing. The value in range 120 – 150 DPI should give you fine results for normal ink jet printers.

This is only an informational number and no resizing is done when you change it and therefore no information is gained or lost. You can of course change it later in Photoshop. But you should pay the attention to this value anyway as it is important for evaluating the results of the sharpening. When you sharpen the image in Photoshop then you should view it on 100% zoom and the resolution should be calculated according to the requested size of the print. The relation of image dimensions to the printing size is:
Printing size in inches = Image dimension in pixels / resolution in DPI
So if the image produced by MRWFormat is 2560 x 1920 pixels and the resolution is set to 150 DPI then the printing size is set to 2560 / 150 = 17 inches wide and 1920 / 150 = 12.8 inches height.

You should calculate the DPI which corresponds to your usual printing size and enter it here.

Increments

You can fine tune the behaviour of the sliders and edit boxes in this area. You can specify three different numbers here for every slider/edit box pair.

Small – when you click on the arrow at the end of the slider than the value will be changed by this amount.

Large – when you click inside the slider, between its thumb and the arrow then the values will be changed by this amount.

Number – when you click inside the edit box by the slider and then press the up or down arrow on the keyboard then the value will be changed by this amount

About

When you press this button the information dialog box is displayed. It is the same dialog which you can display by choosing Help->About Plug-in->MRWFormat from the Photoshop’s menu (or Photoshop‑>About Plug‑In‑>MWRFormat on Macintosh). The most important information displayed here is the version number of the plug-in installed on your computer. There is also a link to my web site which you might click to get there.

Save

All the changes you do in this window will be applied automatically when you close this dialog using the OK button. However the changes will be saved into the .ini file for future use only when you hit OK in the main MRWFormat dialog as well. When you cancel opening the mrw file by hitting Cancel button then the .ini file is not updated and your changes will be lost. The same applies to creating/deleting new presets and white balances. So if you wish to write the new setting in the .ini immediately then you should click the Save button.

 

 

Info window

Clicking the Info button will toggle this Info window on and off. Initially it will be “glued” to the other plug-in windows, but you can move it to any other location.

 

Histogram

This graph shows the distribution of pixel values in your image. It is very helpful when you are trying to find the best settings of various sliders and you do not want to loose too much image information by clipping. The histogram is live (it is redrawn after every change of the controls).

Below is a description of each mode of this histogram:

Original – Displays the original values that are in the raw image before any processing.

Processed – Displays the values that result from processing the image. This is the default mode.

Show curves – Displays the RGB curves inside the histogram window. Observing the changes of these curves is a good way to learn the effects of all the sliders. The display of the curve is a normal mathematical graph. On x axis is the original value of the pixel and then on y axis is the resulting value (after processing).

Tip for learning the effect of the sliders

Set the Gamma to 1, and play with Contrast, Shadows, Exposure and other sliders to see their effect on the curves displayed.

Color radio buttons – You can display all the color channels in one histogram, or you can select any one with these buttons.

Range – Normally the full range of values is displayed. But you can zoom both the highlights and shadows areas. This is useful if you want to expand the histogram right to the edge. The range of values is displayed on the right and left edge of the histogram. These values are represented as percentages on the left/right side of the histogram.

The text ”values 4064-4095, 1.76%” means that the leftmost line of the histogram is the count of the pixels having values from 4064 to 4095, and 1.76% of all pixels in the image have a value in this range.

“Min 139/8” and “Max 4095/255” tells you that the minimum value in the original image was 139 and the maximum was 4095. The minimum in processed image will be 8 and the maximum will be 255.

“Shad. 40/2” and “High. 4004/249” means that all the values lower than 40 (or 2 in 0-255 range) will be forced to zero by the Shadows slider. And values larger than 4004 will be set to the maximum value (this is the result of using the Highlights slider).

Scale to fit – If this box is checked the histogram is automatically scaled to fit the box surrounding it. If it is not checked then the histogram will use a fixed ratio. The lines that would not fit into the surrounding box are represented by the end line(s) being red. I prefer the display without scale to fit.

Camera settings

Below the histogram is a display of the camera settings that were in effect when you took the picture. These values are taken from a “MakerNote” field stored with every picture. Note: These may not be the same as the values displayed by your favorite EXIF viewer. This is because some of the values in EXIF tags are rounded in an imprecise way. The values displayed in the Info window are definitely more trustworthy than the usual EXIF information! You can find more details about MakerNote on my website.

Note for owners of older camera models

Not all available lines will display values. Some of the older cameras do not store this information (Focus mode, Color mode, EV=SV+BV)., or the recorded value is questionable (Focus distance). In these cases an n/a will be displayed. This is another reason for owners of DiMAGE 7 to upgrade their firmware to 2.0 (DiMAGE 7u).

The unusual camera setting is displayed in bold.

Below are the settings that might benefit from further discussion:

Aperture & Shutter speed

I do not use the standard values. I am using the exact values, which are based on power of 2. I believe that even in classic photography the value of 1/500s is rounded to the decimal system, and the real shutter time is 1/512s.

ISO

The values here might look incorrect. For instance it is not a bug if ISO 133 (set to auto) is displayed. The camera has the ability to use values other than 100, 200, 400 and 800. This is not surprising for the native ISO of this CCD chip is about 130. The other ISO values are just multiplication products of the values read from the CCD by some coefficient.

Another interesting case might be ISO 1037 (set to 800). Even when you set the ISO to a fixed value the camera can alter it. But, this happens only in A and S mode.

Focal length

The values here are a bit different from the values stored in EXIF. The reason is that Minolta “adjusts” the focal length values to end up with 7x zoom. But in reality the real zoom of this camera is just 6.5x. The values displayed here do not conform to marketing numbers but rather to reality.

Digital zoom

One of four values will be displayed here: off, digital zoom, elec. magnification or unknown (3). The elec. magnification means that you took the shot with the image enlarged in the EVF. At this time I do not know the meaning of the value unknown (3).

APEX values

The three lines starting with EV display the exposure values in APEX units (Additive System of Photographic Exposure). APEX is a convenient unit for expressing exposure (Ev). The basic rule for calculating the exposure in every camera is Ev = Av + Tv = Bv + Sv. When the shot is exposed properly then Av+Tv is equal to Bv+Sv.
 Ev       – exposure value
 Av      – aperture value, calculated from the selected aperture; Av = 2 * log₂(F_number)
 Tv       – time value, calculated shutter speed; Tv = - log₂(shutter speed)
 Bv       – brightness value, measured by exposimeter; Bv = log₂(brightness in foot lambert)
 Sv       – speed value, calculated from film sensitivity; Sv = log₂(0.32*ISO)

The first line shows the exposure value of the shot calculated from aperture and shutter speed taking into account the optional exposure compensation set on the camera. The second line shows the exposure value calculated from ISO value and the brightness of the scene as measured by camera. The third line shows the difference of these two.

If the picture is exposed properly then the difference will be very small. But if you were shooting in M mode then this difference could be larger, and you could use it as a guide for adjusting the Exposure slider. Note: When shooting with a flash there is always a difference because the brightness of the scene is measured before the flash fires.

Drive mode

If you took the picture using Bracketing, then you will see what was bracketed, and by what amount for example “bracketing exposure ±1/3 EV”.

 

Data Imprint

If you have DiMAGE A1 or A2 camera then you can set it so it records some additional information into the EXIF data. This information will be then displayed here. My favorite setting is to set Data Imprint to Text+ID# and let it imprint to EXIF only. I set the text to my own name and of course I keep the File Number Memory set to On. Then my pictures have unique ascending numbers and it is easier for me to catalogue them easily.

 

Saturation, Contrast, Sharpness, Color Mode

This information displayed here is just for your information. MRWFormat does only display them and none of these settings affects the processing of the image. They are just silently ignored.

Interpolation algorithms

The RAW image is a copy of the values captured from the CCD sensor with very little modification. The CCD sensor is used in combination with a Color Filter Array. Every cell of the CCD is covered by red, green or blue filter. This is called a Bayer pattern.

A single sensor is able to capture all the color information this way, and the designers don’t have to use three different CCD sensors (one each for red, green and blue). The disadvantage is that every pixel contains just the value of one color and the other two are missing. One has to process the RAW image with an interpolation algorithm that tries to guess the missing color information from pixels in the neighborhood.

The choice of the interpolation algorithm is the most important step in the process of converting RAW files. You are advised to spend time with this selection. A sequence would be: open the preview, enable both Interpolation and Zoom check boxes, then examine closely the effect of the interpolation algorithm in various areas of the image. The most important areas to evaluate are:

·         The edges of the objects: should be clean and not jaggy. Also, the colors of adjacent areas should not mix into the edge, contaminating solid black with dots of different colors (rainbow effect).

·         The flat areas contaminated with noise: Some algorithms are very sensitive to noise, and will increase it a lot.

·         Thin lines: Usually the “zipper effect” will be seen here. The algorithm could create isolated dots (brighter or darker than the surrounding area) along edges, and at the end points of lines.

·         Areas of the same color with dramatic differences in luminance: the algorithm could create dots of a color that do not match the object at all. (For example a picture of silver jewelry would have areas that are very bright and areas that are much darker, but these wide variances of luminance all should be close to the same color and not contain dots of color that are way out of range.)

·         Areas with a lot of detail: Ideally interpolation should not cause detail to be lost and should not cause artifacts. Look for this in areas of fur or hair.

Many of the implemented algorithms are based on Ting Chen’s study, which can be found at http://ise0.stanford.edu/~tingchen/

None

This gives you a clue of what the unprocessed RAW image looks like. No interpolation is done, and the pixels are displayed in the original Bayer pattern (however the color conversion is done and the curves are applied). The resulting image could be useful someday if someone creates a stand-alone program for Bayer pattern interpolation. (It would obviously have to accept and process plain bitmaps.)

Relative speed: 3.3 (This speed designation is relative to the Average algorithm. The bigger the number the slower it is.)

 

None (gray)

This is the same as None except the resulting image is black & white. This was inspired by Vector, who claims that in this mode one can find dead pixels easily. But, I can see a possibility of using this interpolation very successfully for infrared photography. (I intend to explore this possibility sometime.)

Relative speed: 3.2

Average (half size)

This is in fact not a real interpolation. One pixel is created from every four R, G, G’ and B pixels. The resulting image is only half size in both height and width of the starting image (resulting in 1280x960 pixels for 5MP camera). Because there is no interpolation done there are no interpolation artifacts. So, if this reduced sized image is enough for you then do not hesitate to use it. Usually this is the best way to create images that will be viewed on screen or the web.

If you process the image with any of the advanced interpolation algorithms, and then scale it down you will probably end up with a lower quality image than this process would produce.

Relative speed: 1

+ very fast algorithm

+ the best possible results

–  produces only half sized images

Nearest Neighbor

This is a very primitive set of algorithms. They are very fast but the results are almost unusable. The missing color information is duplicated from the nearest neighbor pixel that has this missing color. There are four neighbor pixels from which the algorithm can choose. Therefore the variants are: SouthWest, NorthWest, NorthEast, NorthWest. These four variants will differ slightly on the edges.

Relative speed: 3.2

–  this interpolation set produces so many artifacts its use is not recommended

Bilinear

This is a very simple but effective algorithm. The missing color information is averaged from the neighbor pixels of the same color. It is not affected by noise and even reduces it. The drawback is that the details are lost as the image is unsharpened a little bit.

Relative speed: 3.4

+ fast algorithm

+ reducing noise

– some detail is lost

– rainbow effect

– zipper effect

Smooth Hue Transition

This algorithm uses the fact that there are more green pixels than red and blue pixels and it handles them differently. First it interpolates green pixels by bilinear interpolation, and then it calculates the blue and red values trying to deliver a smooth transition in hue from pixel to pixel.

Relative speed: 5.2

+ no rainbow effect

– sensitive to noise (effectively multiplying it)

– slow algorithm

– lot of zipper effect

Smooth Hue Transition Logarithmic

This is simple variation of the previous algorithm. The smoothness of hue transition is considered in logarithmic scale.

Relative speed: 5.7

+ no rainbow effect

– sensitive to noise (effectively multiplying it)

– slow algorithm

–  lot of zipper effect

Edge Sensing I

This is the first adaptive algorithm in the list. Adaptive algorithms examine the image and try to choose the best method for pixel value estimation. This algorithm tries to find edges in green pixels and process them better. For red and blue values a smooth hue transition in logarithmic scale is used. Sometimes the algorithm falsely detects an edge in a flat area (especially when it is contaminated with noise).

Relative speed: 5.8

+ edges are better and the result looks sharper

– slow algorithm

–  spots in flat areas

– a lot of isolated dots

– very sensitive to noise (effectively multiplying it)

Edge Sensing II

This is a slight variation of the previous method as it interpolates green pixels. For the red and blue pixels a bilinear interpolation of the color difference is used.

Relative speed: 3.7

+ edges are very good and the result looks sharper

+ fast algorithm

+ immune to noise

–  isolated dots

Color Correction I

The full name of this algorithm is “Linear Interpolation with Laplacian second-order correction terms I”. This algorithm was designed to produce very good looking vertical and horizontal edges. Despite its long name the algorithm is very fast, and produces very good results.

Relative speed: 3.9

+ fast algorithm

+ very good at flat areas

+ very good at edges

+ no zipper effect

+ immune to noise

–  isolated dots (but relatively few)

Pixel Grouping

This algorithm was invented by Chuan-kai Lin. It combines the ideas from some other interpolation algorithm and the result is surprisingly good. A detailed description of it can be found at: http://www.csie.ntu.edu.tw/~b6506063/demosaic/

The algorithm is very good at interpolating flat areas of solid color and thin lines. There are no zipper effect or isolated dots. It does not perform very well on hair or fur (detail).

Relative speed: 5.3

+ very good at flat areas

+ very good at edges

+ no zipper effect

+ no isolated dots

+ immune to noise

– problems with very fine details (fur, hair)

– sometimes destroys the edge (mixing in the color of the adjacent area)

–  slow algorithm

Variable Number of Gradients

This is a very sophisticated algorithm that evaluates gradients of 5x5 neighbor pixels. Then it tries to choose the most appropriate gradient to determine the missing colors.

Relative speed: 6.0

+ very good at flat areas

+ very good at edges

+ immune to noise

–  a few isolated dots

–  slow algorithm

Variable Number of Gradients (P. Porlan)

This is a modification of the original Variable Number of Gradients created by Patrick Porlan. This interpolation is used in his software Image Shifter. The algorithm produces fewer artifacts around edges than the original algorithm. On the other hand the very fine details are softened a little bit.

Relative speed: 5.9

+ very good at flat areas

+ very good at edges

+ immune to noise

– rainbow effect

– slow algorithm

Variable Number of Gradients (DJ)

This is a slight modification of the original Variable Number of Gradients. There is a slight difference in interpolation of metallic areas.

Relative speed: 6.0

+ very good at flat areas

+ very good at edges

+ immune to noise

–  a few isolated dots

–  slow algorithm

 

 

Summary

In the most cases the last six algorithms will perform very well. Every one of them has its weaknesses. You have to decide which is best based on the content of the image, combined with trial and error. Pixel Grouping performs very well on images containing areas of solid color, and not too much detail. Variable Number of Gradients can handle details very well. The Color Correction I is the fastest of these and handles both detail and solid areas well. So, when in doubt use it.

 

 

Scripting and batch processing

The plug-in is fully scripting aware so you can use all the power of Photoshop’s scripting system to automate repeating tasks.

The process is to create an action is:

1.       Start Photoshop and open the Actions palette.

2.       Create a new set called My MRW Presets.

3.       Create a new action in this set and name it, (in this example Preset 1) and hit Record.

4.       Click file open and then open a .mrw file.

5.       Set all the settings in plug-in that you want and then click Open.

6.       You can add additional steps in Photoshop as Levels, Auto Color, ...

7.       Click the Stop button to stop recording the action.

 

Now you can then use a very handy thing called Batch. Select File‑>Automate‑>Batch… The Batch dialog will appear. From the Set menu select My MRW Presets and in the Action menu choose the stored preset. The Source is usually a folder. You should Choose the folder where the raw files you want to batch convert are stored. Make sure you enable Override Action “Open” Commands. Then click OK, and all the files found in the selected folder will be processed one by one with your chosen settings. You will then be asked how to save them.

Of course you can create a more complicated action which would include the Save command too. Then it would be fully automatic. You could even add some filters or other image processing in the action. Please refer to the Photoshop documentation.

If you use the Batch command regularly, you should create a Droplet. Creating the Droplet is very similar to creating a Batch described above. But the result is saved as an executable file that you can store on your desktop. The created Droplet will automatically process all the files you drag and drop on it. This is very useful if you need to process bunch of mrw files to jpeg. Using the Droplet will be many times faster then doing the same thing in Minolta DiVU.

Tip for batch processing in Photoshop Elements

The batch processing in Photoshop Elements is intentionally limited by Adobe. However still there is a way to batch process many mrw files at once. It requires some manual editing of the .ini file so only experienced uses should try it. The details how to set-up the batch in PS Elements can be found on my support forum at http://www.dalibor.cz/support

Frequently Asked Questions

1.      Why is there no thumbnail in File‑>Open menu?

There is a file size limit in Adobe Photoshop. If the file you have selected in File‑>Open dialog exceeds this limit then Photoshop doesn’t ask the plug-in for an image thumbnail. Unfortunately all .mrw files exceed this limit, and there is no known way of increasing this limit. But in newer versions of Photoshop you can use File‑>Browse feature which displays thumbnails correctly.

If you really miss the thumbnails then you can try my other software Dalifer which is able to display the thumbnails of .mrw files in the File->Open menu. If you want to find out more about Dalifer then please visit http://www.dalibor.cz/minolta/dalifer.htm You can find the description of all the features built-in this little utility and you can download and test the Demo version as well. As a MRWFormat user you are entitled to 50% discount if you decide to purchase it.

2.      Why is there no sharpening in your plug-in?

Sharpening of the image should always be the very last step in processing of your image. It should come after all editing, color adjustments and noise removal. The image should be sharpened just before printing. Also the sharpening for monitor viewing and printing are different so sharpening parameters should be based on the end use (display/print). Therefore sharpening in the early stages of image processing is not very clever.

3.      Why is there no noise removal feature in your plug-in?

There are many very sophisticated tools for noise removal, and I do not want to compete with them at this time. A very good tool is Neat Image (http://www.neatimage.com). I suggest that you give it a try for I believe that it is the best tool available. Neat Image can also intelligently sharpen your images taking noise into consideration (it can distinguish between noise and image details and sharpens just the image and not noise).
Wil Smith recommends taking a look at Grain Surgery 2 (http://www.visinf.com/gs/ps/).

4.      Why doesn’t the Cloudy (raw) in your plug-in match the cloudy preset in my camera?

This is explained in the section White balance.

5.      Why does your Info dialog display different EXIF values than my favorite EXIF viewer?

The EXIF information is stored in every picture. And, there is a special tag called MakerNote. Inside this tag is stored the complete information about the camera settings which were in effect when taking the picture. This tag is the ultimate source of the information as it is much more precise than the normal EXIF tags. The information I display in Info window is based on MakerNote tag. If you find any value that is different in EXIF tag (and there are many of them) I believe the Info window as it is more precise.
You can find more information on my website http://www.dalibor.cz/minolta/makernote.htm and there is also a small DOS command line utility, exiftool, to decode this tag. It works on TIFF and JPEG files too (http://www.dalibor.cz/minolta/exiftool.htm).

6.      Why is CCIndex grayed out and zero always?

Unfortunately the formulas for these are not known at this moment. Bradley Broom has discovered how to calculate Temp from the white balance coefficients. But the formula for CCIndex and both the reverse formulas are still unknown. As soon as I know how to calculate the CCIndex from white balance coefficients (and white balance coefficients from Temp and CCIndex) I will update the plug-in and enable both sliders.

7.      How do I batch convert a series of images?

Easy, using the Batch (or Droplets) features of Photoshop. See the section “Scripting and batch processing”.

8.      How do I switch between MRWFormat and Adobe Camera Raw plug-in?

If you own the Adobe Camera Raw plug-in (it is supplied together with Photoshop CS) then it is effectively disabled for .mrw files by installation of MRWFormat. MRWFormat has a higher priority than Camera Raw and therefore it is used for all operations with Minolta raw files, including the displaying of thumbnails and metadata in Photoshop’s browser.
If you want to open a single file using Camera Raw then the easiest way is to use the command File->Open As and select Camera Raw from the Open As menu. This will override Photoshop’s priorities and it will open the selected file in Camera Raw. Note that not all versions of Photoshop have File->Open As command.
If you want to open more files in Camera Raw then it is probably better to disable MRWFormat temporarily. To do so close Photoshop and locate its Plug-ins folder where you have copied the file MRWFormat.8bi file during the installation. Rename MRWFormat.8bi to ~MRWFormat.8bi and start Photoshop again.

9.      Do I have to keep the thumbnail files PICTxxxx.thm?

The short answer is: No, they are useless. Neither MRWFormat not Adobe Camera Raw use them.
The longer answer is: If you do use DiVU regularly then you should keep them. The only use of those .thm files is in DiVU. DiVU uses them to quickly display the thumbnails of the raw files in the folder. But even DiVU can live without them as when it doesn’t find them along with the .mrw file then it generates the thumbnail from the .mrw file itself. This takes quite a long time but if you use DiVU only occasionally and you can live with the additional delay to generated the thumbnail then you can delete .thm files straight away.

10.    MRWFormat does behave in a strange way after upgrading to version 2.xx

I had to change the format of the .ini file between version 1.x and 2.x. The new version will normally load the old format and convert it to the new one. However, the conversion might occasionally fail. If you get issues, try to locate the .ini file and move it onto your desktop and restart Photoshop. If the problem disappears then the conversion has probably failed and you would have to set up the plug-in's preferences again. The .ini file is named MRWFormat.ini on Windows and MinoltaRaw.ini on Mac.

11.    There are some features not described in this User Guide

This user guide was written for MRWFormat version 2.10. If you are using newer version then there might be new features added which are not covered here. I always add a text file ReadMe.txt to the MRWFormat package. The new features will be described in this text file.

12.    Why is my question not listed here?

You haven’t asked it yet. Ask it using the support forum at http://www.dalibor.cz/support and I’ll add it here.

13.    Is that you in the picture in this User Guide?

Yes, that’s me.