Hi
Took your advice…got rid of the space between Macintosh and HD…
I would say at least part of the solution.
Thanks
Hi Xavier and Everyone of you!
Problem cracked…here is what got me going on the right track and it was a comment from Xavier.
In that example you’re really adding 2 new custom profiles: ACES and ACEScg.
There are only 10 slots in the Working Spaces Profiles template in RT…
You will notice that the top two are Named;
Acesp0
Acesp1
What I did was to create a new Json with the DCam3 file in it and named it Acesp1. As I saved this Acesp1 a warning came up that I was entering a privileged site ! I went ahead and entered Acesp1 with Xavier’s much tighter DCam3 script and Paperdigits hunch on tightening Macintosh HD to MacintoshHD. MacOs Finder then confirmed that DCam 3 profile was now lodged in Acesp1 as a working profile !
Deducting from that experience it seems to me that Custom Working Profiles can overwrite Acesp1and Acesp2 just by overwriting them…
i definitely would never have cracked that one on my own
Take a bow everyone and
Thank You
matt
ps off to bed (ireland)
1 Like
Hi @keentolearn,
Sorry to disappoint you, but what you describe doesn’t make much sense to me… It’s been a while since I took a look at that code (which I wrote fwiw) but I don’t remember that you can overwrite the built in profiles. And there’s no limit on how many of them you can have, unless this was changed recently… I’ll have another look later today to double check just to be sure
So, I just downloaded the free “Ekta Space PS 5” profile, and it works as expected in RT 5.8. So I suspect that you are either not editing the workingspaces.json file correctly, or you are putting the file in the wrong location.
workingspaces.json should be placed in the directory of your ICC profiles, which you can set here:
After doing that, it shows up as expected:
Here’s my workingspaces.json:
{"working_spaces": [
{
"name" : "Ekta Space PS 5, J Holmes",
"file" : "/path/to/iccprofiles/Ekta Space PS 5, J Holmes.icc"
}
]}
Finally, note that this space has a gamma of approx 2.2, but RT will ignore that, as all working spaces are restricted to be liner (i.e. “gamma 1.0”).
HTH
3 Likes
Thank You for taking so much trouble to get to the bottom of this…appreciated…Are you using Catalina ?
I ask because RT posted on line recently that there was a problem with Catalina 10.15.4 handing over files to RT through normal HD searching methods.possibly a directory issue…?.workaround was to use Pictures Gallery in Mac Os 10.15.4 to load files into for use by RT … as the only known way to transfer files to RT…a fix was needed…RT works now using Mac OS 10.15.4 Pictures Gallery but otherwise I cannot ask RT to retrieve a file directly from MacOs 10.15.4…probably no connection to the issue that i am having?
My point is that I am still unable to load DCam3 J.Holmes.icc into the RT Working Profile Bay.
I did successfully install renamed DCam 3.icc in a ACESp1 shell ( DCam 3,J.Holmes.icc…(you remember advising using that name simplificaton for workingspaces.json )
You will see below that i got the Holmes profile tucked in to ACES1.json and with a couple of “bold boy” warnings as well.The ACESp2.json warning was just a demo to show you now…I didn’t copy the warning last night for ACESp1 but the warning was the same as the ACESp2 attempt …
To summarise
a I am writing (i hope) the .Json as you outlined
b the Displays folder on Mac OS is “good to go” for profiles, as it seemed to work last night and definitely did work last weekend for INPUT and Output testing.
c In RT preferences, Color Management, I named the Displays folder as the Directory for Color Profiles
d I checked permissions on the profiles and they are good for RT to use and with open permission for everyone.
This is the makeup of the current workingspaces.json profile
I am sorry that I cannot seem to fix this …I would like to
thank you for all your work…
matt
1 Like
Thanks for that information…
I will back off to an older mac os and see can i replicate Alberto’s work,
After that I guess Pro Photo with floating point is a simpler way to utilise RT for a working space. I have been told that resizing up in RT is the best there is and I’m determined to use it !
You have all been extremely helpful. Mac Os has been difficult for the last 6 month s or so…
matt
Matt
Not at all, thanks for the info, interesting. (I used to print but haven’t for quite a while now)
@Keentolearn Hey Matt and everyone else who are still interested, sorry for being late to the party. Please take note from Rawpedia’s notes the following, emphasis mine:
Working Profile
The default working profile is ProPhoto and should not be changed for normal use.
The working profile specifies the working color space, which is the color space used for internal calculations, for instance for calculating saturation, RGB brightness/contrast and tone curve adjustments, chrominance, etc.
When RawTherapee was based on integer math it was wise to not use working space larger than absolutely needed to get the best precision in the calculations. However, RawTherapee had switched to floating-point processing in 2011, and since version 4.0.12 the default working profile is ProPhoto, which has a very large gamut.
The choice of working profile has an influence on the effect of the curves in all modes except for perceptual - in that mode, changing the working profile will not alter the effect of the curve. If you have trouble fitting colors within the output gamut you can experiment with changing the working profile when using curves in any mode but perceptual.
Note that the working profile will only specify the red, green and blue primaries, gamma will not change as RawTherapee’s processing pipeline is floating point with no gamma encoding (that is gamma = 1.0). Some tools (like curves and histograms) will still display with a gamma (usually sRGB gamma) which is hard-coded for the tool and stays the same regardless of working profile.
My suggestion is to
- Leave the working space profile as Prophoto as per default
- Use Joe Holmes’ Dcam spaces/whatever preferred RGB color space you prefer as the output space (this is how ACR works, the list of spaces at the bottom of the window refer to output spaces, it doesn’t change it’s internal working space either)
- Do not use any RGB “working” colour spaces as the input space (of course)
Do beware of clipping issues when outputting to RGB spaces smaller than Prophoto RGB, and use whatever preferred method of gamut mapping if needed (such as Dcamprof/Lumariver’s gamut mapping function for this purpose).
1 Like
Just to add, any RGB “working” color space can have its RGB coordinates modified such that assigning this new profile will change the chroma of the image. Dcam3 in itself is not somehow special that it is “capable of producing more saturation than the space itself”. Further, Joe Holmes has chroma variants of the standard AdobeRGB and ProphotoRGB as well, which achieve the same chroma modification effect. Their main advantage used to be superior increase/decrease of chroma without causing luminance or hue shift, and its effect is uniform across all hues. Saturation sliders tend to exaggerate the effect in warm colours while “Vibrance” sliders tend to exaggerate the cool colours.
Interestingly RawTherapee has the ability to achieve virtually the same chromaticity change effect, (not possible in ACR/LR) and has even more sophisticated functions such as boosting chroma in relation to luminance/hue/chroma. But Joe’s profiles and their related variants retain one very important advantage, which is the non-destructive manner in which they are supposed to be applied (and the use of masking in Photoshop whilst the chroma variant is applied to another image layer and copied over (requires conversion into the document space then) allows for more selective adjustments).
As the five Dcam spaces are designed as input-centric spaces, it is irrelevant what the output intent is for your images. You can benefit from them if you do not print. These profiles are designed to be as efficient as possible in containing all the incoming colours from your raw images. The intent is for one to select the smallest possible Dcam space for output to minimise quantization error. This is a more serious problem with 8 bit images than 16 bit. The spaces’ RGB coordinates are also optimized to encompass in their gamuts the most useful and commonly occurring real world colours. Admittedly they are not so useful anymore in 15/16/32 bit floating point workflows.
To clarify further, Dcam3 is not larger than Adobe RGB all over, it has a green coordinate that is shifted towards yellow while Adobe RGB is shifted towards cyan. In practice, useful real-world colours like foliage is encompassed better in Dcam3, while there are not so many super-saturated cyan-green colours (so that part of the gamut can be considered as wasted in Adobe RGB). Dcam 4 and 5 encompass those extreme colours if so required for your needs, while some no-so-colourful images can use Dcam 2 or even 1.
Hi Samuel
Great to see you joining in the discussion and thank you for some sound advice and lots of knowledge too! .
Over the weekend , I had made up my mind , to INPUT /OUTPUT , (in Raw Therapee) with Holmes Profiles (with a floating point Pro Photo WORKING SPACE in the middle …and then back to Photoshop CC … then print from Photoshop CC with Imageprint printing Profiles .
I have used Holmes DCam3 plus Variants for a decade or more ( I would guess) and i have found them , for me personally, unbeatable really when it comes to printing when combined with Imageprint. Everyone has their own best way and I don’t imagine for a moment that I have learnt all there is to know (especially on this blog) ! The file I am bringing in to RT started its life as Photoshop CC Raw ( J.Holmes ) and moved on to finish its initial journey in CC processed , as a tiff, with 16 Bit DCAM3 and some Chroma Variants for colour touchup. My reason for accepting Pro Photo as a WORKING SPACE in RT was that I have never left 16 Bit and Pro Photo is floating point and there is no interference to DCam 3’s Gamma of 1.96 coming through RT INPUT and on to RT WORKING SPACE. Joseph Holmes is very proud of his “Perceptually and Proprietary” based Tone Curves and has little time for what he calls “Gammas” that are found in contemporary working spaces.(see below)…(I personally haven’t a notion of the finer points)
To be honest, my initial intention is to head, once I am in RT , to RESIZE UP ( which I am told is the best there is on the market today) and I don’t see any Gamma churning from Pro Photo and as I won’t actually be using RT WORKING SPACE just now…just yet…(I am brand new to RT and captivated by whats in store but its for another day). I will use Neutral in Preferences to be sure.
1 In your posting above, included is information from the RT manual (p334) as follows ; (emphasis mine)
“you can experiment with changing the working profile”
RT sees no problem in adding an additional working space if required …Alberto kindly demonstrated that he could include Holmes working spaces in the working spaces section of his own RT…Unfortunately for myself, Mac Os 10.15.4 Catalina is simply unable to function adequately with RT 5.8 and it is not possible to include any additional RT working space profiles for the moment with Catalina.I will keep on trying !
2
Do beware of clipping issues when outputting to RGB spaces smaller than Prophoto RGB
Hopefully if I can get DCam 3 into RT working spaces, the journey from a raw image in Photoshop CC to a printer profile (on the return journey back to Photoshop CC) will have been UPSIZED in RT at the highest possible standard and without any potential clipping issues on that journey .
Two points made by Joseph Holmes, in parenthesis, have stuck in my brain over the years as follows:
1 " All five DCam spaces utilize the ICC standard D50 white point definitions and they share my proprietary tone curve which is an excellent match to perceptual linearity and is superior to any gamma curve for preserving image levels when converting into the profiles of well behaved printers".
2 “If a color doesn’t fit into DCam3 because its moderately saturated but too light i.e because there isn’t enough headroom in DCam3, then it wont fit into any printer or display gamut either , so this discipline tends to be useful”
Matt
Hi Matt,
Thanks too! This is fun stuff 
Just to be sure, did you say you are using Dcam3 as the input profile in RawTherapee? For working on raw images, the input profile should be a proper camera profile, not a working space profile like Dcam3.
Great to hear that, as long as you get the results you are after and pleased with your workflow, who’s to say otherwise? I’ve been familiar with Joe’s Dcam profiles for over a decade too.
Sorry, I don’t understand what you mean by this. Is this some sort of raw file captured on a digital camera? What is a Photoshop CC Raw? Why is Joe’s name in parenthesis here, what has he got to do with your raw file?
As you rightly pointed out, Joe isn’t using a gamma TRC in his profiles, so Dcam3 does not have a gamma of 1.96.
You might want to look into Topaz Gigapixel and Photozoom Pro. Out of curiosity, why are you resampling your image at this earliest stage in your workflow?
Yes of course one can use any working profile you like, RT is flexible that way. It doesn’t necessarily mean that this is the best approach for everyone, or in every situation.
This would depend on the way your raw file’s colours are rendered, so clipping may or may not occur in any given instance. Dcam3 is indeed a good general space for most real world images, though there certainly will be instances where a larger space might be required to avoid unnecessary clipping. Otherwise Joe would not have made Dcam4 and 5. Likewise sometimes Dcam 2 or even 1 could be better, if one is maniacal about avoiding quantisation.
This is probably right when Joe was working on 8 bit images back in the day when he created the Dcam spaces, but with 15/16 bits, it doesn’t matter in a meaningful way any longer.
Unfortunately then you may eventually encounter clipping which cannot be undone moving forward. Not to mention that printer gamuts have grown quite a bit since the Dcam spaces were invented; my printer is capable of rendering some greens and cyans outside of Dcam3. Anyway, it is irrelevant whether your working space is larger than your current printer or display gamut. It is best to think of it as a container for your data. You would like to not clip any data (throw it away), even if you are going to “pour” it into a smaller container later. Good gamut mapping will help to squeeze out-of-gamut colours into printer space without losing detail the way hard clipping will. And you will have this extra data if one day you can get another larger container (better output devices with better technology). It is best to think of them in an input-centric way as Joe recommends, as opposed to an output centric way which is what you are doing currently.
Hi Samuel
Fun continues 
did you say you are using Dcam3 as the input profile in RawTherapee? For working on raw images, the input profile should be a proper camera profile, not a working space profile like Dcam3.
Sorry ! I should have been much clearer about what I was doing before entering RT.
Round 1
Nikon D300 and Nikkor 2.8 lens
Images NEF
Operating System :MacOS 10.15.4 Catalina
Applications :ACR with Joseph Holmes as the raw converter color space nominated to process the nef files ,relative colorometric, 16 bit
Nikkor lens also profiled in ACR as below:
Then across to Adobe Photoshop 2020 (Joseph Holmes is the working space in Photoshop)
Then whatever in Photoshop … smart objects back and forth to ACR, layer adjustments ,contrast, luminosity layers etc everything except sharpen or sizing…
flattened copy file saved as tiff (with holmes working space profile included) with the idea of entering RT next to avail of the UPSIZING in RT. (ps will look at Topaz and Photozoom as you suggest!)
Round 2
RT
The INPUT …( tiff this time) Holmes
Working Space (Hopefully) Holmes
OUTPUT space Holmes
saved again as tiff after Resizing in RT with neutral set in RT Preferences and then back to Photoshop (Stage 3) to sharpen and print with Imageprint printer profiles and Epson k3 Inks and Epson 7800 printer…old stuff!! but I can do either Colour or B/W with the Phatte Black system on the one set of inks.
Sorry, I don’t understand what you mean by this. Is this some sort of raw file captured on a digital camera? What is a Photoshop CC Raw? Why is Joe’s name in parenthesis here, what has he got to do with your raw file?
Sorry again!! NEF files are being processed in ACR with DCam3JHolmes Color Space doing the converting from NEF into RGB …first snapshot above shows me choosing DCam3 JHolmes in ACR
As you rightly pointed out, Joe isn’t using a gamma TRC in his profiles, so Dcam3 does not have a gamma of 1.96.
On opening the DCam3 profile a Gamma of 1.96 is stated (along with the primaries ,white point etcetera). Can you tell me what is the relationship between a 1.96 gamma as stated in the profile and a perceptually linear TRC? I guess that a TRC is redistribution of the bits back towards the darker areas so as to match better the way that our eyes are more sensitive to darker information rather than brighter areas .As you know, Its a patented TRC that Joseph Holmes greatly prides himself as being the inventor thereof. You know I can’t help but feel that there is a likeness in the way Holmes thinks about printing as being eerily similar to the way Ansel Adams thought about his zone system as applied to the print itself. My knowledge is limited in this area and I would be grateful if you can take me a stage further on.
why are you resampling your image at this earliest stage in your workflow?
I hope the above explanations will clear up the confusion here. Basically i am heading towards the middle stage of the workflow when i enter stage 2 with the entering into RT
**can use any working profile you like, RT is flexible that way. **
Catalina is the problem. I have an older computer with Mac Os 10.6.8 …but guess what? …is a 32 bit system and RT 5.8 is 64 bit
to be solved…
there certainly will be instances where a larger space might be required to avoid unnecessary clipping
You are so right. Joseph advised me to get going with the DCam3 at the time and as you say there are 4 other spaces to suit all needs. DCam3 (and all the spaces) can expand its chroma capacity by 99% just by using its variants. I was impressed to learn that DCam 3 is 61% bigger than Adobe RGB ! I also must confess that i have the Pro Photo set which I had to have when ACR excluded Holmes spaces from its menu. Yes I will use Pro Photo with Holmes variants but first i want to have DCam 3 installed in RT and try that.
This is probably right when Joe was working on 8 bit images back in the day when he created the Dcam spaces, but with 15/16 bits, it doesn’t matter in a meaningful way any longer.
Again you are right. In fairness to Joe he states that fact over and over again on his blog. What he advised me to do way back was go 16 bit in ACR and stay 16 bit for as long as possible .
Joe ,(as I said earlier) wasn’t included in ACR back then so ProPhoto was his ACR recommendation for that time and then convert to DCam3 in Photoshop. As i quoted Joe in an earlier post …"if you see clipping in DCam 3 then don’t expect to print that clipped colour(s)…its a good discipline to work in DCAm 3 " NOT HIS EXACT WORDS BUT CLOSE
Thing is …why would I jump into a bigger space when i have a space that not only suits my printing needs right now but that I can expand its chromacity(99%) without moving up to a higher space? Pro Photo doesn’t do variants and compression might have to happen to jump back down into my printer space. If DCam 3 works for me combined with its 30 (+_) Variants and Holmes Perceptual TRC , is it possible that a Pro Photo floating curve system might be better again?. The pressing need right now though , it seems to me, is to get Dam 3 working in RT as a working space and then look very carefully at the options that your conversation is creating!.
It is best to think of them in an input-centric way as Joe recommends
yes point taken …I do have Dcam 3 and Pro Photo Holmes Variants and also have the master files themselves on a couple of external drives as well. What you are advising though does make great sense…plan for the future and don’t throw away information that might be very valuable and usable in a few years!
Thank You Samuel for sharing so much of your knowledge and time as always.
Matt😊
Oh I see! Thanks for clearing that up. So the raw processing is not occurring inside RawTherapee (RT) but ACR. I used to use ACR all the time, but have since switched to RT for all my raw processing. I find the quality of processing for many adjustments in RT much better with more control than ACR. Initially there were a few tools where ACR seemed to be superior, and at the time when I switched RT didn’t handle DCP camera profiles properly but this has all been fixed and done beautifully, with some additional useful functionality over ACR. Since you are using RT for upsampling now, perhaps consider using it for your raw processing too.
If you are processing a TIFF in RT, you can set the input profile to “use embedded”. This is by default anyway, saves you an additional step. It is also better in the case for whatever reason, you open a file that isn’t in Dcam3 space and you forget to change the input profile setting.
I’m assuming that if the working space is left to its default of Prophoto, there will be a conversion to Prophoto, and then conversion again back out to Dcam3 as your output profile. Ideally, we would like to avoid that, but I think the damage will be minimal but someone else might chime in here who knows better. Do TIFFs enjoy 32 bit floating point precision for editing in RT if they are 15 bits to start with (coming out of Photoshop)?
I think it’s just a behavior to express the profile’s TRC as an approximation in terms of gamma, similar to sRGB being displayed as gamma 2.2, in whatever app you are using to open the profile. It may not be an accurate reflection of the true TRC of the profile.
Got it. Though there are better apps for upsampling for print…
Yes, but don’t get too excited on just this point without the full picture It is important to know where it is bigger and whether that extra gamut is useful (in the case of Dcam3, it is, because it was carefully designed).
Sounds like a contradiction? Joe did make chroma variants for Prophoto…
OK, just to be clear, I don’t mean to say you must use Prophoto or Dcam3 or any specific space. Use what works and what suits you. Just beware that Dcam3 is not a one-size-fits-all space but if your imagery’s incoming colours all happen to happily fit into Dcam3, you are just fine.
Not if you don’t require the additional gamut of Prophoto. I’ve only stuck to Prophoto since I’m using that as the working space inside RT and it is convenient to have the output space be the same.
That’s it! Have fun
@Keentolearn @samuelchia
I don’t know this Joseph Holmes guy, or if he’s a photography hotshot, but after having read the introduction to his profiles (About My Profiles - Joseph Holmes), I call bullshit on his claims.
- He’s using authoritative arguments to convince you his profiles are somehow better than the industry standard ProPhoto (“I’ve been using them for years”).
- He apparently needs testimonials to help you be convinced to pay a ridiculous amount of money for his profiles. Not even Adobe does that…
- His profiles intend to determine a particular look and feel for your images. He claims some vague argumentation about better chroma and a “proprietary TRC”. But this is exactly the opposite of what a working profile should do. A working profile should simply give you the room to edit to your photos so that you can accurately determine their look.
- He gives only two tiny zip files with examples on why his working profiles are useful. The rest are walls of text and pictures of gamuts. I find that a hardly convincing argument. That may be just me, but since working profiles are a very technical matter, I prefer technical explanations. Take for example the work of Elle Stone (Elle Stone's well-behaved ICC profiles and code).
Bottom line: as far as I have understood, the simple truth is that in a floating point workflow you want a gamut that is (much) larger than your device ‘color space’ and your output color space. It may matter a little which RGB primaries you choose, but with a large enough gamut, then under normal circumstances you will never have to fuss over banding and clipping. And you will have full control over the look and feel of your photos. ProPhoto fits that description perfectly.
Please @Keentolearn, can you explain again why you are still keen in using something else? Unfortunately I couldn’t make heads or tails of your previous answer to the same question by @RawConvert.
Please call me out if I’ve written bullshit myself!
2 Likes
@afre asked me my point of view
I do not know the works of J.Holmes, surely the profiles are of good quality, otherwise he would not sell them.
I only look to see technical informations and I don’t found many things…
I’m going to remember what we can do with RT (free):
- of course you can use all the output profile furnished in version ICCv2 or ICCv4 (RTv2…, RTv4…)
- and of course you can generated your own profiles with “ICC profiles creator”
You can choose:
-
The primaries :
a) default : ACES AP0, ACES AP1, AdobeRGB, Prophoto, Rec 2020, sRGB, Wide Gamut, BestRGB, BetaRGB, Bruce RGB
b) Custom : you can choose the values for
RedX - range 0.63 0.735 and RedY - range 0.265 0.335
GreenX - range 0.0 0.31 and GreenY - range 0.59 1.0
BlueX - range 0.0001 0.16 and blueY - range -0.08 0.07 -
the TRC : Tone response curve
a) default - BT709 g=2.2 slope=4.5, sRGB g=2.4 slope=12.92, linear, standard 2.2, standard 1.8, High g=1.3 slope=3.35, Low g=2.8 s=6.9, Lab g=3.0 s=9.03296
b) custom ; any combination with gamma between 1.0 and 3.5 and slope between 0.0 and 15.0 -
illuminant
a) Default
b) D41, D50, D55, D60, D65, D80, StdA 2856K -
ICC version: ICCv2 or ICCv4
And of course Add a “description”
Jacques
1 Like
Hi again:
Sorry to say this, but I’m not really sure you have fixed your problem, nor if you really know much about color management.
Anyway, let’s just try again, step by step:
- you already have a good workflow for your use case and just wish to use RT as an intermediate tool in your workflow, so there will be no need (and no wish) to change the color profile of the input image
-
you already have access to your Picture folder within macOS Catalina. Let’s say that folder is located in:
/Macintosh HD/YOUR_USER_NAME/Pictures(of course you have to change «YOUR_USER_NAME» with your actual user name in macOS) - copy your DCam profile to «/Macintosh HD/YOUR_USER_NAME/Pictures»
- create a «workingspaces.json» file next to your DCam profile, with this contents (remember you have to change «YOUR_USER_NAME» with your actual user name in macOS):
{"working_spaces": [
{
"name" : "DCam3",
"file" : "/Macintosh HD/YOUR_USER_NAME/Pictures/DCam3.icc"
}
]}
- open RT and go to Preferences>Color Management>Directory containing color profiles. Change that directory to «/Macintosh HD/YOUR_USER_NAME/Pictures» (most probably you will only see «Pictures»)
- restart RT
You should have now your working profile ready to be selected. And at the same time you will be able to select DCam as your input and output profiles.
Hope this helps
Hi Xavier
Outline as suggested by you…(and thank you for your genuine effort to help me)
1 RT Color Mgt setup for Pictures in Mac OS
2
3 Workingspaces file being created
4 Combobox without the DCam 3 working space included in it
I obviously am still doing something wrong.
Matt
Correct me if I’m wrong, but it seems that your file name has spaces between words, and in your workingspaces.json file it seems that the spaces have been removed?
If that’s the case, then RT is looking for a file that doesn’t exist. If macOS accepts spaces then copy/paste the file name into your workinspaces.json file, so they both share the same filename