Welcome to my RSP profiles page
This page contains profiles for use with a D70 in Raw Shooter Premium 2006 from http://www.pixmantec.com.
The reason for this profile is to make untweaked pictures from RSP look more like the out-of-camera JPEGs (in mode Ia). It has been created by Per Baekgaard (baekgaard at-sign b4net the-dot dk, http://www.b4net.dk) using argyll (http://www.argyllcms.com) and various custom made software pieces and small tweaks, and dozens of test shots of IT8 targets (from Wolf Faust, http://www.coloraid.de).
It will be especially helpful for skintones, where the built-in profile to my taste gives too red colours, and the ETC profiles sometimes results in an even more “grayish” look.
These profiles also has a slightly higher contrast than the other profiles mentioned above.
[Compared to some of my pre-release profiles, it is not as contrasty as the “mode II” profile. It also has whitebalance problems largely fixed, and I have removed the slightly warm colourcast that was present.]
To illustrate, I’ve included some portraits showing quite clearly the differences (on a calibrated screen!). These are shot under 3-4 different conditions — some formal, some not, obviously. To me, the Internal profile (the first column) is clearly too saturated red in the skintones. The ETC profiles (two last columns) in general do quite well here, although I find them a bit too “grayish” for some of the portraits.
If you haven’t invested in the ETC profiles, these are clearly recommended.
Why not just use the one from Nikon Capture?
It is true that NC 4.x comes with a set of profiles for the D70, and if you’re technically skilled (or good at using google ;-), you can easily find these profiles. You would then need to “patch” them to the proper type and checksum/length fields…
These profiles, however, do not match RSP, and although it is technically possible to tweak the input shaper curves to make them somewhate useable, they will still not give the right look due to what appears to be internal differences in postprocessing in NC 4.x compared to RSP. The resulting images will be too contrasty (for my taste) and there may be strange artifacts where almost identical colours suddenly map into rather different colours, depending on white balance settings.
To install, unzip the file
from the zip archive and copy it to the directory on your windows machine that holds the ICC profiles, usually
Copyright and Distribution
To the best of my knowledge, the profile attached is only copyrighted by me, and should not infringe on anyones intellectual property.
From my side, I allow anyone to use the profile for private or business use without limitations on its use within RSP, but it cannot be redistributed. Anyone that wishes to use this profile must download the latest version directly from http://www.b4net.dk/rsp
If anyone feels this profile infringes on their intellectual property rights, please let me know, and I’ll stop distributing it and recall any rights to use it.
Is it really free
Yes, it is free to use and you don’t have to pay to use it.
I have spent more than 100 hours working on D70 profiles and have invested in IT8 targets and other needed equipment. If you want to offset some of these expenses, feel free to donate any small amount you’d like to my PayPal account (email@example.com).
As is, the profile has a few limitations — or in other words, it is not completely up the quality I’d like it to have.
I hope to have these problems fixed in coming releases.
- One area that still needs refinement (although it will not in the usual cases result in any diminising of the value of using it) is that gray tones do not always map to gray tones; there can be a difference of +/- one RGB value. Thus, where you may expect to see 240,240,240, you could end up with 240,241,239 (in the worst cases).
- Another area, although of less practical implication, is the choice of blue chromaticity. The current profile, 3.00.053-m, has a negative y value for “true” blue (similar to what dcraw and also apparently PhotoShop) has. This is caused by the way the blue channel has to affect the other channels, but is not strictly needed. It can potentially cause some too dark blue colours in the near-UV spectrum, but I’m yet to see this in practice.
- Possibly related to this, there are other areas of local colour space distortions. Although these are now reasonably well under control compared to some of my previous attempts, there can still be found problems in certain outer parts of the gamut (in other words slight discolouration in shadows or highlights).
I think I know how these can be fixed, but it will require a bit more time to get it properly under control.
- Lastly, one “quirk” is that it usually clips pure white at 254, not 255… so pure white almost never is marked when pressing ctrl. It has no visible effect otherwise, but watch out for clipping highlights, so that you don’t accidently miss them if you use the ctrl feature often!
If you find any other problems or have comments to the profiles, please let me know by email to baekgaard at-sign b4net the-dot dk.