Home · Register · Search · View Winners · Software · Hosting · Software · Join Upload & Sell

Moderated by: Fred Miranda
Username   Password

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Alternative Gear & Lenses | Join Upload & Sell

  

Archive 2013 · Sigma DP Merrills now supported by Iridient Raw Developer
  
 
HelenB
Offline

Upload & Sell: Off
p.1 #1 · p.1 #1 · Sigma DP Merrills now supported by Iridient Raw Developer


I was pleased to see that version 2 of Raw Developer, now called Iridient Developer, supports the DP Merrills. I'm just downloading it now and will try it ASAP.


Jan 20, 2013 at 08:10 PM
mpmendenhall
Offline
• • • •
Upload & Sell: Off
p.1 #2 · p.1 #2 · Sigma DP Merrills now supported by Iridient Raw Developer


I've been experimenting with the demo.

It's got several things I like (especially compared to SPP):
- Fast operation
- Retina-display aware
- Much more flexibility/control over choices in sharpening, noise reduction, etc.
- Creates more "natural" (or at least "closer to original sensor data") results, not enforcing Sigma's idiosyncratic processing "look" (e.g. crushed, desaturated shadow tones to hide chroma noise)
- Fast operation

There are also a couple areas where I prefer SPP's processing:
- Detail for distant subjects. SPP seems to preserve a lot of detail through distance/haze, where Iridient results are mushy (similar to non-Foveon results).
100% crops example, SPP:






Iridient (note how the mid-frame trees are complete mush, compared to the shading SPP was able to extract):






- Black and white from red-filtered photos comes out dull/noisy from Iridient (no matter what conversion settings I choose).
100% crops example, SPP:






Iridient:






If Iridient was better on distance and color-filtered B&W conversions (two areas where I think the Foveon sensor has the greatest advantage over other cameras), it would be a no-brainer purchase to replace SPP. As is, I'd still need to use SPP for many of my shots, though purchasing Iridient might still be worthwhile for avoiding SPP in other cases.



Jan 21, 2013 at 08:58 AM
mpmendenhall
Offline
• • • •
Upload & Sell: Off
p.1 #3 · p.1 #3 · Sigma DP Merrills now supported by Iridient Raw Developer


Another Iridient issue: strong color shifts. DP2M images from SPP often have moderate color shift towards the edges, but SPP appears to (partially) correct for this. Iridient doesn't "know" about this issue, and produces images with full-blown color shifts.

SPP:





Iridient:







Jan 21, 2013 at 09:57 AM
Tariq Gibran
Online
• • • • • •
Upload & Sell: On
p.1 #4 · p.1 #4 · Sigma DP Merrills now supported by Iridient Raw Developer


Pretty much what I'm seeing as well. There is a lot of non user controllable processing going on behind the scenes with Sigma Photo Pro. If SPP were able to read Tiff's, it would be interesting to see if the results on traditional bayer sensor images were similar to what we see with Foveon images processed with SPP.


Jan 21, 2013 at 02:02 PM
bgbg
Offline

Upload & Sell: Off
p.1 #5 · p.1 #5 · Sigma DP Merrills now supported by Iridient Raw Developer


A couple notes here,
1) For monochrome conversions be sure sure to checkout the "Sigma XDR Monochrome" option found under the Input Profile popup menu near the bottom of the In tab pane. This monochrome processing is completely unique to the Foveon sensor used in the Sigmas and should give better results than the standard monochrome modes found on the Mono tab, at least in regards to getting the most possible data out of the sensor data... there are various special monochrome conversions styles available on the Mono tab, such as custom channel mixers, that would obviously give a different look. So "better" here may be an individual preference.

2) Regarding distant detail, SPP uses a lot of sharpening in its defaults. Some seems to be applied even with sharpening set to its minimum -2.0 value. If you want a crisper, sharper result, especially with distant objects, try increasing the amount of sharpening in Iridient Developer or try one of the other sharpening methods. My R-L Deconvolution sharpening method in particular can really pull an amazing amount of detail out of these Sigma files.

Brian Griffith
Iridient Digital
Iridient Developer Developer



Jan 25, 2013 at 03:11 AM
bgbg
Offline

Upload & Sell: Off
p.1 #6 · p.1 #6 · Sigma DP Merrills now supported by Iridient Raw Developer


mpmendenhall wrote:
Another Iridient issue: strong color shifts. DP2M images from SPP often have moderate color shift towards the edges, but SPP appears to (partially) correct for this. Iridient doesn't "know" about this issue, and produces images with full-blown color shifts.


Regarding color casts, I actually do apply correction processing for these images using embedded X3F metadata and without correction things would actually be worse (at least in all test images I have at this time)... That said some users, particularly it seems with the DP2M model, seem to be still getting images with significant color cast, such as this example.

This is the very first release I've made with DP Merrill support so something may not be getting handled with these cameras as well as it should. In 2.0 I also made improvements for the color cast issue for the SD1 which in my testing seemed to apply well to the DP1M/DP2M despite the format of the metadata being changed slightly with these models. So far user feedback with the SD1 Merrill seems to indicate things have gotten much better in version 2.0, not sure why some of the DP2M images are not looking so good...

With some cameras and/or some settings options (aperture, focus distance, ISO, firmware, etc) these issues seem to be worse, oddly some Sigma users seem to be very happy with the latest Sigma changes made in v2.0 so my guess is this may be related to something camera (like firmware version) or settings specific. It's still early and I'm not sure why some images or camera bodies seem to suffer from the issue so much worse than others, but I'm looking into it and hope to be able to continue to make refinements to my Sigma processing for future versions.

If you use a Sigma and see strong color cast issues and have a test image or even a few please send an email to me at support. The more test cases for this issue I can get the faster I can hopefully get it resolved.

Thanks,
Brian Griffith
Iridient Developer Developer



Jan 25, 2013 at 03:25 AM
Tariq Gibran
Online
• • • • • •
Upload & Sell: On
p.1 #7 · p.1 #7 · Sigma DP Merrills now supported by Iridient Raw Developer


bgbg wrote:
A couple notes here,
1) For monochrome conversions be sure sure to checkout the "Sigma XDR Monochrome" option found under the Input Profile popup menu near the bottom of the In tab pane. This monochrome processing is completely unique to the Foveon sensor used in the Sigmas and should give better results than the standard monochrome modes found on the Mono tab, at least in regards to getting the most possible data out of the sensor data... there are various special monochrome conversions styles available on the Mono tab, such as custom channel mixers, that would obviously give a different look.
...Show more

Hi Brian,

Thanks for the more detailed information...and thanks for supporting these cameras! I have been using your great software for many years.

I was curious about how your software deals with noise in Foveon Merrill files as compared to how SPP does so. At first, I thought SPP simply crushed the blacks to hide noise but it appears there is further noise suppression going on automatically with SPP in the shadows. Sean Reid recently posted an example showing this - much more noise showing up with Iridient Developer versus SPP.

As an aside, maybe I missed the change, but just wondering why the change in name from Raw Developer to Iridient Developr. Perhaps to avoid confusion with all the current developers with "raw" in their name?



Jan 25, 2013 at 02:18 PM
bgbg
Offline

Upload & Sell: Off
p.1 #8 · p.1 #8 · Sigma DP Merrills now supported by Iridient Raw Developer


Tariq Gibran wrote:
Hi Brian,

Thanks for the more detailed information...and thanks for supporting these cameras! I have been using your great software for many years.

I was curious about how your software deals with noise in Foveon Merrill files as compared to how SPP does so. At first, I thought SPP simply crushed the blacks to hide noise but it appears there is further noise suppression going on automatically with SPP in the shadows. Sean Reid recently posted an example showing this - much more noise showing up with Iridient Developer versus SPP.

As an aside, maybe I missed the change, but just wondering
...Show more

I changed the name to Iridient Developer for a couple reasons:
1) It's no longer just for RAW images anymore. Many standard image formats (TIF, JPG, PNG, PS, others) and a couple HDR formats as well (OpenEXR and Radiance HDR) are now supported.
2) Yes, the RAW Developer name was just way too generic and when people mentioned "RAW Developer" it often was mistaken for any old raw processor not specifically the RAW Developer program from Iridient.

There are complex noise issues with these Sigma cameras due to the way the Foveon sensor design works. I have a number of completely unique Foveon specific noise processing options available in Iridient Developer to help deal with the noise. I do make different noise tradeoffs in my processing than SPP does, and its just the nature of RAW image processing in general that no 2 programs will ever be exactly the same in output, but there are a number of ways for the user to adjust noise reduction to best suit there personal preferences in Iridient Developer.

In particular for Foveon noise:
1) Don't completely disable all noise reduction, this just makes no sense with these cameras. This is like cranking sharpening all the way up as high as it will possibly go and saying their are sharpening artifacts. I repeat do not completely disable noise reduction for these cameras.

Same goes for sharpening, Iridient Developer offers some wonderful sharpening options. Completely turning off sharpening just tosses this functionality out the window and is not generally recommended with any RAW file from any camera. I leave this up to the user, but if you turn off sharpening and then find the images look soft compared to another program the reason is most likely because you turned sharpening off.

I think I mentioned the RL Deconvolution option before, if you have a fast processor this sharpening option can give amazing results, some users absolutely love it. Sharpening looks are a personal preference and that's why I offer 4 options. There is also a great high quality (floating point LAB colorspace based) unsharp masking option available for a more traditional sharpening appearance and can give a bit more "edge" than RL Deconvolution does. The other 2 are nice too...

Many raw processors even with sharpening set to none or zero or -2 or -100 will still apply some default amount of "sharpening" type processing automatically to images regardless.

1) The Early Stage Noise Reduction slider --> this slider is completely customized for Foveon sensor. My personal preference with noise tends to allow for more noise than many other processors and my own primary output target is print which will tend to mask noise much more than closeup viewing on a computer monitor.

If you are going to be primarily viewing images on computer monitor at very close range. Go ahead and turn this adjustment way up, it's not going to radically destroy detail even at a value of 20.

2) Even the above still doesn't take care of the noise enough try enabling the "Soft Look" checkbox. With Bayer cameras this uses an alternate demosaic method. With the Foveons this changes color processing to a lower noise option. You may prefer the lower noise option and this processing will automatically vary based on image ISO.

3) If the above 2 still aren't enough there are both chroma and luminance smoothing sliders. Nothing specific here for Foveon, the same processing is available for Bayer and TIFF/JPEG files, but this can allow for some even further noise reduction as well.

Turning off absolutely all of this noise reduction and then claiming the processing is noisy is kind of pointless, but in the example you mention from Sean Reid this is what he did.

Yes, these cameras are noisy (very) I think just about everyone is in agreement on that. SPP applies a whole bunch of noise reduction and other processing that cannot be turned off by the user. Iridient Developer does a lot of this too, however in some areas like sharpening and noise reduction I may allow a little more user level control over the process than other software. This extra user level control can allow for some ugly results in some cases, but I tend to leave more of the decision making up to the user for better or worse.

Brian Griffith
Iridient Digital



Jan 25, 2013 at 04:47 PM
 

Search in Used Dept. 



mpmendenhall
Offline
• • • •
Upload & Sell: Off
p.1 #9 · p.1 #9 · Sigma DP Merrills now supported by Iridient Raw Developer


Thanks for the information -- the "Sigma XDR Monochrome" setting gives great results for B&W conversion (maybe this should appear under the "Mono" tab settings just for X3F files to make it easier to find?). I've just purchased an Iridient license to support what seems like a nice replacement for SPP.

For the detail at distance issue, I'm not sure that this is simply a sharpening issue, at least not on the same size scale that the "Sharpen" slider in SPP (or the various sharpening options in Iridient) address. In my sample above, it's a fairly coarse scale of contrast/detail that only SPP is recovering from the distant tree line (though I can match the finer detail sharpening). If I were going to guess what SPP is doing, perhaps it is relying more on the bottom/red sensor layer for luminance info in these areas of the image (like shooting through a haze-cutting orange/red filter)?

Is there a way to export the original 3-layer sensor RAW data (before processing to RGB color, similar to "dcraw -d") as a TIFF16? If not, that would be an interesting feature.



Jan 25, 2013 at 05:52 PM
bgbg
Offline

Upload & Sell: Off
p.1 #10 · p.1 #10 · Sigma DP Merrills now supported by Iridient Raw Developer


mpmendenhall wrote:
Is there a way to export the original 3-layer sensor RAW data (before processing to RGB color, similar to "dcraw -d") as a TIFF16? If not, that would be an interesting feature.


No and I don't plan to offer such an option. However I believe in addition to dcraw there are some other open source tools specific to the Foveon sensors that do allow for this such as:
http://www.proxel.se/x3f.html

Brian Griffith
Iridient Digital



Jan 25, 2013 at 06:53 PM
bgbg
Offline

Upload & Sell: Off
p.1 #11 · p.1 #11 · Sigma DP Merrills now supported by Iridient Raw Developer


mpmendenhall wrote:
For the detail at distance issue, I'm not sure that this is simply a sharpening issue, at least not on the same size scale that the "Sharpen" slider in SPP (or the various sharpening options in Iridient) address. In my sample above, it's a fairly coarse scale of contrast/detail that only SPP is recovering from the distant tree line (though I can match the finer detail sharpening). If I were going to guess what SPP is doing, perhaps it is relying more on the bottom/red sensor layer for luminance info in these areas of the image (like shooting through a
...Show more

I'd need to see an example X3F file to really know what's going on there. There are differences between how micro detail, very fine details, grain, textures, etc and larger macro edge details like buildings, large tree branches would be handled between various sharpening options in Iridient Developer though. RL Deconvolution tends to be spectacular at resolving micro detail, traditional unsharp mask can be better at really sharpening up the appearance of larger macro edges. With larger details sometimes a simple global contrast change or tone curve will give the appearance of more overall sharpness too.

Brian Griffith
Iridient Digital




Jan 25, 2013 at 06:59 PM
Tariq Gibran
Online
• • • • • •
Upload & Sell: On
p.1 #12 · p.1 #12 · Sigma DP Merrills now supported by Iridient Raw Developer


Thanks Brian for the detailed explanation. The RL Deconvolution sharpening offered in your raw developer is actually what most impressed me with it years ago...along with it's very decent interpolation capability (which adds a little noise/ grain in the process I believe). For landscape type imagery with distant detail, I have found it to be one of the better sharpening methods to use with traditional bayer based sensors (I haven't found it to be quite as good with foveon based images, probably because they don't suffer the typical issues of AA softening to begin with).


Jan 25, 2013 at 10:12 PM
Chrissearle
Offline
• •
Upload & Sell: Off
p.1 #13 · p.1 #13 · Sigma DP Merrills now supported by Iridient Raw Developer


I'm running ID2.2 and making careful comparisons with SPP5.5 Both (obviously) have pro's and cons, however, unlike other commentators I find less of a magenta cast in challenging light with ID than with SPP. Distant detail is indeed less well defined in ID however I sharpen BUT the overall 'Foveon look' is definitely not lost. I'll probably end up using both dependant on what each file needs and who knows, maybe even blending in PS. Fun times....


Oct 11, 2013 at 04:30 AM
mcbroomf
Online
• • • •
Upload & Sell: Off
p.1 #14 · p.1 #14 · Sigma DP Merrills now supported by Iridient Raw Developer


Anyone tried 2.3 yet?
http://diglloyd.com/blog/2013/20131010_4a-IridientDeveloper.html

Mike



Oct 11, 2013 at 08:10 AM
SoulNibbler
Offline
• •
Upload & Sell: Off
p.1 #15 · p.1 #15 · Sigma DP Merrills now supported by Iridient Raw Developer


I know its unlikely but is there ever going to be linux support. Alternatively is there cute work around to run this on linux without resorting to using a mac VM?


Oct 11, 2013 at 01:11 PM





FM Forums | Alternative Gear & Lenses | Join Upload & Sell

    
 

You are not logged in. Login or Register

Username   Password    Reset password