New profile making software: Lumariver Profile Designer

[Rado]

I've just snatched the Repro version to create some C1 profiles. Happy to support a fellow programmer/photographer.

[David Good]

I picked up the Repro version as well for my Sony and Capture One. Excellent work Anders,

Dave

[kwalsh]

It is an interesting subject, would a 3000K-5000K be better than 2700K-6500K profile for example? Maybe, I don't know as I haven't done enough experiments. To test the simulation capabilites of DCamProf using SSF would be a place to start. I personally use the combo StdA-D50 or D55, I think D65 is unnecessarily high. It happens quite a lot from 3000K down to 2700K (the temp scale is not linear when it comes to color) so I'm thinking maybe one should step up a bit on the low, but I haven't made any serious testing there. It would be interesting to hear if someone else has some experience or ideas to share.

Thanks Anders for the detailed reply.  Very helpful!

It will probably be late summer before I have time to setup any controlled testing - but if I do end up trying out some different dual illuminant setups I'll be sure to share my results here.

[photocor]

When I want to download my reference file made with BasicColor, this window appears:
The reflectance spectra are missing and the illuminant is unknown or unsupported, so the colors can not be identified. Add an "ILLUMINATION_NAME" entry to the file (must be either D50 or D65). [/ I]
I solve this problem by adding the proposed ILLUMINATION_NAME "D50" text just after the CREATED draft.

But this information already exists in the field CONDITIONS OF MEASUREMENT inthe BasicColor file. Could you add this condition to avoid rewriting the information?

Regards

[torger]

When I want to download my reference file made with BasicColor, this window appears:
The reflectance spectra are missing and the illuminant is unknown or unsupported, so the colors can not be identified. Add an "ILLUMINATION_NAME" entry to the file (must be either D50 or D65). [/ I]
I solve this problem by adding the proposed ILLUMINATION_NAME "D50" text just after the CREATED draft.

But this information already exists in the field CONDITIONS OF MEASUREMENT inthe BasicColor file. Could you add this condition to avoid rewriting the information?

Regards

Yep, added to the TODO list. Thanks for reporting.

(I'm not reading as often here currently. I'm reading frequently in some periods, other periods not so much. So if anyone has a specific support issue you can drop an email too, and indeed I got this issue via mail too and saw it there first.)

[Hening Bettermann]

Hi Anders!

I have recently bought some of these lamps:

https://store.yujiintl.com/collections/high-cri-led-lights/products/bc-series-a60-high-cri-remote-phosphor-led-bulb-unit-2-pcs

which offer 5000K and D50 in an E27 socket and have impressive curves. They were originally thought just for enlightening my "darkroom" for tonal editing. Then I thought they might as well be used for shooting the Color Checker for creating color profiles for all my lenses. They would be much more comfortable to work with than the Solux.

For the shooting, I set the camera to 5000K. Iridient calls that 5150K (non-DNG figures).

However, when I view the resulting shots on my calibrated screen, the middle gray patch reads 96-97-96 in Solux light, and 96-96-91 in Yuji light. So it looks like the Yuji lamps are lacking in blue.

Question: Does this make a difference for the profiles? Or will it disappear because  dcamprof/Lumariver Profiler calculates an ideal white anyway? I can not look through this and would appreciate enlightenment.

Best regards - Hening

I sent a copy of this post as e-mail to your PM address.

[jc1]

  Sharing my Velvia Builds


NikonD810-jc1-D810-velvia.icc  < Link >

FujiXT2-jc1-XT2-velvia.icc  < Link >

[jc1]

Velvia Color Justification

Reference Raw: Checker24  in < Imaging Resource  >
                       
Reference file:  < here >
Derived by converting Reference Raw to Tiff with Fujifilm RFC ver 4.2.7.0 and with simulated color set to Velvia.
Reference color was extracted with PatchTool.

Based on test images, new profile yields more pleasing blue sky, though old copy has lower dE.
Curve to be set to AUTO or standard film curve, just likes with C1 generic profile.

NikonD810-jc1-D810-velvia build 002     < here >

[jrp]

The Lumariver profile designer is worth every € over dcamprof for ease of use.  Indeed it is so easy to create new profiles that putting some work into enhancing the facilities for documenting the settings used to generate a particular profile would be welcome.

I find that the differences between different "looks" is very subtle and so probably not worth refining much further.  My pictures certainly look much more natural than with the base Adobe profile.

One characteristic of the Natural+ profiles for the camera that I have worked most on (Leica SL) is that saturated highlights (e.g., on flowers) get lightened and desaturated.  So I often have to reduce the highlights considerably in ACR/Lightroom.  This may be a side-effect of the way that highlight roll-off is handled by the profiler.  Is there any way of producing a profiles that produces darker, more constrasty, highlights?

It's sometimes instructive to compare what the Lightroom "Auto tone" setting does under different profiles. In my case, it often raises the shadows and increases the exposure.  It would be good to have a tone curve in Lumariver that was closer to what ACR auto-tone produces, as I often find that to be a good starting point for further adjustment (typically for the Black / White points, e.g.)

[Tim Lookingbill]

One characteristic of the Natural+ profiles for the camera that I have worked most on (Leica SL) is that saturated highlights (e.g., on flowers) get lightened and desaturated.  So I often have to reduce the highlights considerably in ACR/Lightroom.  This may be a side-effect of the way that highlight roll-off is handled by the profiler.  Is there any way of producing a profiles that produces darker, more constrasty, highlights?

Assuming you're working on Raw data instead of cooked pixels in tiff format, the highlight contrast for more defined highlights may be caused by the editing app itself which from what you describe is compression or flatness and dullness to colored flower petal highlights in the app's attempt to reduce the color gamut (primarily controlled by luminance) and thus dynamic range.

With ACR/LR's PV2012 you have to brighten with the White slider by first pulling back Highlights to -100. This puts back the linear scaling toward max clipping of highlights that was in PV2010's Exposure slider  and thus preserves highlight definition.

I can't speak for how to do this in C1 or other Raw editing apps. This behavior with PV2012 is why I've stuck with PV2010.

I don't believe you can expect a camera profile to control this kind of behavior to highlights on a consistent basis since scene dynamic range with daylight lit colored flowers varies drastically both in color gamut and color clipping near full saturation.

[jrp]

Thanks.  Yes, it's PV2012 that I am using.

I though that I remember reading somewhere in the manual that a setting raises and desaturates the highlights.  I wonder whether it's possible to reverse that effect.

I can see that a universal profile may be impossible, but I would have hoped that it might be possible to generate a profile against which you don't have to fight to get a good starting point.

[Tim Lookingbill]

Thanks.  Yes, it's PV2012 that I am using.

I though that I remember reading somewhere in the manual that a setting raises and desaturates the highlights.  I wonder whether it's possible to reverse that effect.

I can see that a universal profile may be impossible, but I would have hoped that it might be possible to generate a profile against which you don't have to fight to get a good starting point.

I've found for preserving detail and definition in saturated bright surfaces to make them as bright as possible is to use HSL panel in ACR/LR. See the A/B comparison of only adjusting yellow luminance, but I used a dual illuminant DNG profile made in Adobe DNG Profile Wizard which always saturates these types of colors. You can also use the adjustment brush on just the areas of detail you want to preserve.

[dwalme]

Velvia Color Justification

Reference Raw: Checker24  in < Imaging Resource  >
                       
Reference file:  < here >
Derived by converting Reference Raw to Tiff with Fujifilm RFC ver 4.2.7.0 and with simulated color set to Velvia.
Reference color was extracted with PatchTool.

Based on test images, new profile yields more pleasing blue sky, though old copy has lower dE.
Curve to be set to AUTO or standard film curve, just likes with C1 generic profile.

NikonD810-jc1-D810-velvia build 002     < here >

Thank you for posting this.   I am interested in doing this myself for a couple of my cameras.

Are there any reference sites you can link to that step through the process a bit more in detail?

[jc1]

Very brief summary in < here >
Too few feedback so I am unsure if these custom profiles work satisfactorily for general photography.

[daicehawk]

Will buy a lite version that allows:

• Opening multiple raw images at once
• Selecting a monitor profile different from the system one
• Selecting\adjusting a display curve applied to open raw images
• Selecting the workspace or it being ProPhoto or similar (larger than Adobe RGB)
• WB of each image by a right-click and selecting the weighting area square like the RT
• Exposure compensation +-3stops to set a middle gray as one wants to (the middle gray in the histogram being the default)
• Being able to tweak RGB primaries (CMY not needed) from zero by hue, saturation and lightness. I DO understand that tweaking one value will affect all other colors in the same way as the LUT tweaking does (in a worse way regarding the robustness of the profile)
• Grey axis stays grey

Basically I need an extended version of Adobe DNG Profile editor. What bothers me now I do not know whether the start matrix profile (the more so the LUT profile which I would like to strip the LUT off of) is good enough to my taste, and the controls I have are only hue and saturation without lightness. Ok, lightness might be absent as a control since it will be interconnected with the hue and saturation, but I want to be able to tweak the hue and saturation from zero with keeping the grey axis grey.

[digitaldog]

Will buy a lite version that allows:

• Selecting a monitor profile different from the system one

Why would you want to do that?

[daicehawk]

Why would you want to do that?

A somehow valid use case is different monitor profiles Print White Point D50 (or whatever the visual WP of the lightbox on a particular paper is) vs sRGB Web D65 profile. Of course you need to relod the calibration externally. I am just not aware how Windows behaves when switching profiles via 3rd part apps like DispCAL profile loader. I suppose Win does not assign the default profile to the one you have switched to.  I agree it is not a critical feature.

[digitaldog]

A somehow valid use case is different monitor profiles Print White Point D50 (or whatever the visual WP of the lightbox on a particular paper is) vs sRGB Web D65 profile. Of course you need to relod the calibration externally.

But the 'system' profile for display you always use is still (and alway) in effect so how can this work? How and why use a display profile that doesn't define the current display conditions?

[scyth]

But the 'system' profile for display you always use is still (and alway) in effect so how can this work?

/off topic/ on Windows platform it is up to a specific software code what to use (if at all)...

[Alexey.Danilchenko]

/off topic/ on Windows platform it is up to a specific software code what to use (if at all)...

Sure but that makes what you looking at even less predictable - I am also puzzled as to why would anyone need it...