Request/Suggestion: Fix to avoid being caught-out if NOT using Soft Proofing (with new Working Color Space)

I’m just curious: Is soft-proofing these days still necessary? My understanding of doing it was to have to deliver data for magazines printed on various print machines and various profiled papers on this machines, so it made sense to not do hard proofs but soft proofs. But whoever wants to print his/her images will almost certainly buy a decent printer at least in A3+ or bigger. There are profiles for paper and printers, monitors are calibrated (again, in the workplace of serious amateurs or professionals). I have to join the question of @Joanna: What has to change?

Edit: Just saw, there’s PL 6.0.1 which should solve some problems with Soft-Proofing.

6.01 only for Mac, not for Windows.

Hi Joachim,

soft proofing for me is a essential part of my workflow.
I did not own a printer for the formats I need for posters I place on the walls of my flat.
I have made the experience that without softproofing the posters don’t look like I’ve seen it on my calibrated monitors. The most of them was a little bit too dark, or some colours was “washed out”.
So over the times I have approached my current workflow.
Monitors are calibrated and set to the brightness of 90cd, soft proofing with LR because DXO hasn’t this function, and asking my manufacturer for the printer/paper profile. Since them I was happy with the results.

best regards

Guenter

1 Like

I understand. When you let others print your images (like I also would do for bigger sizes than A3+) it can become essential to see the paper underneath the image before it’s printed. Thanks for clarifying.

about new pics in PL6 …

For now, in PL6 the user can decide to stay with the so far default Wide Gamut WCS or to change it and go with Classic-Legacy – on a per picture basis.

DxO – why not let the user decide, how to work and keep that setting in the preferences to start with, what he/she wants?

There are multiple reasons for to be in Classic-Legacy WCS, ranging from the often used uncalibrated screens (“profile? – what profile?”), missing knowledge (“so far not necessary”), convenience (“let the program take care for the colour management”) … right to those, who just want to keep working in sRGB for ease of use or to get the job done.


Working with wider gamut than the screen is capable of, implicates problems.

The handling of out-of-gamut colours, when converting from a wide(r) to a small(er) colour space is done by so-called rendering intents. In softproof, the user can choose a perceptual or relative colorimetric rendering intent, while in export DxO decided for an ‘automated’ process.
[ If the latter is related to the several times mentioned “Protect Colour Saturation”, I don’t know. ]

When in softproof, one can temporarily simulate the outcome of a pic before to export & apply a certain
profile (ProPhotoRGB, AdobeRGB, sRGB IEC61966-2.1, a dedicated printer/paper profile and the likes).
[ note – softproofing for another monitor is something else … ]

It doesn’t help – working in a colour space beyond the screen’s capability ( → visible with Monitor gamut warning on ) is ‘like a blind date’.

… which is available in PL5 only → different “…(gamma 2.2)” in PL6

For printing, I would export to ProPhoto RGB and print the resulting TIFF file from Preview, ColorSync or Canon’s utility, using a custom ICC profile for my printer/paper/ink.

The results are wonderful. I never knew things could be any better and everyone who used my printing services were totally satisfied.

just continue without any headache and don’t bother, that your PL5 export to ProPhotoRGB doesn’t contain colours beyond AdobeRGB. – You always have been happy with the results.


Supplementary Question (post #8)

1.I want to use the WGCS
2. My camera uses AdobeRGB
3. My monitors are Apple (which I believe are P3) but custom profiled.
4.I export for web/email/etc using sRGB
5.I export for printing to ProPhoto RGB
6.I do not need to soft proof for printing as I use ICC profiles.

7.I also need to teach this stuff to beginners at our club photo,
so I don’t want to have to teach anything more than absolutely necessary.

Just do as you like. – Now in PL6 and depending on the pic, your export to ProPhoto RGB can contain out-of-gamut colours, which your screen most probably doesn’t show, but your print should do.

That’s not really terrible, as I had tried not so long ago → Yet more colour space confusion - #24 by Wolfgang (and also w/ your infamous crabs). The prints then showed some more saturated colours and possibly a ‘better’ gradation, but nothing earth shattering, nothing what I couldn’t do without. And these were straight out prints from PS, no softproof, no correction.

It’s a big difference, if one is on a sRGB monitor and tries to get around w/ wide gamut colour space,
or on one like your’s. A ‘good’ monitor, covering P3, AdobeRGB …, shows most of the colours and you can check with softproofing to sRGB IEC61966-2.1, what / how users see on a sRGB screen.

[ When I was in a photo club (run their homepage and also had put out instructions about CM, beamer,
printing and such → monitor & beamer were in sRGB), I always told them to ‘stay in sRGB, when they don’t know better’ … difficult for some. ]

Now the big question, why ‘should’ you softproof?

  • As long you don’t change colour space, there is no need for – you see on your screen, what’s happening. If you are happy with, what you get out of your prints – no need for …

  • Things get different, when e.g. using matte paper, which cannot reproduce deep blacks and you have to lift the shadows … better in softproof (and w/ the ‘coming soon’ paper& ink simulation)
    [ wherever you might do that … I’ve been using PS so far ].

  • Exporting a pic w/ strong out-of-gamut colours from Wide Gamut WCS to sRGB IEC61966-2.1,
    most probably will look different than when done from Classic-Legacy (PL5 style).
    The reason is, PL6 WG (then) handles colours differently and you should see it on your screen.
    .
    In spite of export, you can compare the masterfile (WG) and a virtual copy (CL)
    – or more interesting softproof VCs (set to sRGB) from each of them …
    and compare them to both exports (WG → sRGB and CL → sRGB).
    If the equivalent results from WG resp. CL look identical, save your time & judge from your screen.

The biggest practical problems I see i is using virtual copies since the current stupid ones were introduced it takes a lot of messing about. You have to copy and paste them then delete the unwanted ones. OK with one or two but 20 or more. Some users have large numbers and it’s not practical, if DxO have introduced this as a problem when using wide gamut they haven’t done so with a solution to solve it and made it the way to mitigate export problems difficult with the existing virtual copy’s. The original virtual copies were so straight forward, just delete the unwanted and the let one was the master one.

Hi John,
I have no problems w/ VCs – except after a while I don’t remember, what they were meant for
and try to find out with
grafik

This is where you need the Windows version to get the Mac feature of being able to rename virtual copies. It is so useful

4 Likes

Currently, with PLv6, if you expect what you see within-PL in your Step#2 (Edit Image) to be always, consistently & reliably exactly what you see in your Step#3 (Look at the result and share with others) - then you will need to have Soft Proofing = ON. … … … Why? - - - See above.

And that’s pretty much the point of my proposal, Joanna

  • For those who don’t want to even think about Soft Proofing … including those who don’t understand why they should need to - I reckon PL should always work with WYSIWYG behaviour, by default.

  • Instead, with PLv6’s current implementation, sometimes (but not with all images) that will not be the case; it will be mystifying to the average user, and that will impact confidence in using PLv6 … :-1:

If my proposal were to be implemented (or one of the variations of it, as suggested by @maderafunk and @KeithRJ) then we could all go back to ignoring Soft Proofing until we really needed it for some specific purpose … such as printing or export to different color space(s).

That is the result I’m aiming for.


Yes - If you’re expecting the WYSIWYG behaviour we enjoyed pre-PLv6, in all cases.

  • It’s incorrect to assume that SP is needed ONLY for printing.
  • This is the precise issue that this proposal is addressing - - I reckon PLv6 should “fail safe” … instead of setting-up unsuspecting users for a (probably mystifying) surprise.

Why? - - - See above.


And now, with PLv6, SP=ON is advisable in all cases - including for simple export to digital files … NOT only for printing. Why? - - - See above.


They’re all good points, Wolfgang - but, that’s not the issue under discussion here. My concern is that PLv6 sets-up unsuspecting users - who, in good faith, adopt WCS=WG - will eventually get caught-out if they do not apply Soft Proofing for all images they’re processing — NOT ONLY for printing.

Unfortunately, that’s no longer the case … Why? - - - See above.


Keep the questions coming … I feel we’re “getting there” !

(Apologies for my repeated referral to the background info above; but I reckon it’s best that I don’t confuse matters by continually repeating the explanation).

Sounding good, Keith - as seems it would address both our concerns (Win/Win :sparkle: )

But, I’m still not quite sure what you mean by; {Make PL6 behave exactly like PL5 …}

  • Could you elaborate a bit more (again), please Keith … In what way "Same as PLv5 " ?

John

Quite simple actually:

  1. If SP=Off then make PL6 fully WYSIWYG as with PL5 no matter which WCS is used. This includes the ability set your monitor profile that was removed from PL6
  2. If SP=On then make PL6 behave as it currently does. This must include adding compare against the original image WITH the SP profile applied but no other corrections other than currently selectable options (with/without geometry) to the compare button.

Hope that clears things up.

OK - I reckon I’ve got it now - finally !! … (Apols for my slow take-up - In retrospect, it was obvious).

But, let me paraphrase to be sure I’m truly on-the-same-page with you …

  1. With SP=ON - no change is required to PLv6 bahaviour.

  2. With SP=OFF - then the following changes are required to PLv6

  • Re-instate ability (in preferences) to specify the ICC Profile of one’s monitor
    image - - as it was with PLv5
    – with “Current profile of the display device/monitor” as the default.

  • When SP=OFF, the additional Protect Saturated Colors algorithm (that’s currently applied only when SP=ON and also for the Export to Disk process) is applied for the ICC Profile specified in Preferences.
    This ensures the “holy grail” of consistent WYSIWYG :+1:

Plus some additional, essential, refinements;

  • When using the Compare option, regardless of whether that’s with Geography or without, the same behaviour as above (1. & 2.) must also be applied.

  • In the Export to Disk dialogue, the ICC Profile option should default to “Same as Soft Proofing” (rather than “As Shot”) … as another “fail safe” step in protecting unsuspecting users who may have their camera settings = Adobe RGB, without appreciating the unintentional impact).


If that’s what you mean then I’m totally on-board … this would solve all current issues. :white_check_mark: … Thanks for your help in getting this across, Keith … Hopefully, my proposal will now attract many more votes :thinking:

John M

Hi @John-M,

that looks very good and more precise than my wording :slight_smile:

1 Like

I have not yet upgraded to PL6.
But having worked in the printing industry for decades - soft proofing is a simulated view on how the current page/image/photo will look like after it’s been printed or processed.

It’s temporary preview which one can toggle o or off for a quick view on how the final resume will look like.
It’s nothing that should be permanent, included in the files or otherwise affect the production.

If that actually occurs in PL6, then DxO is either using the term “soft proofing” wrong or their color management is processing the profile simulation incorrect.

That’s all completely true of PLv6 too. There’s nothing about PL’s implementation of SP (in and of itself) that’s permanent.

However, for the new Working Color Space = Wide Gamut, when PL exports one’s image to disk (via Export to Disk) …

  • it applies a “secret sauce” algorithm to ensure that colours from the WG color space fit “appropriately” into the target color space (such as when exporting to sRGB - but equally to any Color Space inferior to Wide Gamut) … Ditto when exporting for printing.

  • this additional algorithm is applied in the cases above - and when SP=ON … to ensure that what-you-see (WYS) in Soft Proofing mode is what-you-get (WYG) with the exported image.

  • however, this additional algorithm is (currently) NOT applied when SP=OFF … – and this is the issue I’m aiming to resolve - because, in this case, WYS (is not necessarily) WYG !!
    – as explained in a LOT (!) more detail here.

John

Hi @John-M

Sorry to be so persistent but, despite all your excellent explanations, I find myself entrenching more firmly in the camp for not upgrading to PL6.

When I used to use Photoshop (CS3) all I did was setup the default working space for all images to be converted to on first reading - I used ProPhoto RGB.

This meant that all images were converted to that colour space and I didn’t have to care any more.

When I printed directly from PS, all I did was tell PS to manage the output profile and selected the appropriate profile for the printer/paper/ink I was targeting.

No soft proofing, no nothing, but every print came out as expected. I can only assume that the default viewing mechanism took care of mapping because I never had to dow anything explicit to get my prints right.

Now - all of a sudden - PL6 dictates that I have to complicate my workflow and virtually insists that I do soft proofing, even if all I want to do is export to an sRGB JPG file.

Yes, I am a total techie nerd, a computer programmer who has written extraordinarily complex code and an LF film and digital photographer who has held exhibitions.

You are also equally experienced and techie minded, otherwise you wouldn’t have participated in the beta test to try and help DxO find bugs.

But, what about those people who just want to take the photos of their family and friends or holidays and distribute them on the web? Are DxO really saying that they have to get into all this colour space “stuff” just for that?

I know perfectly well I have enough trouble teaching PL5 to members of our club photo, without having to now sit them down and explain about having to use soft proofing, just for web sharing or emails.

All most folks want to do is use a transparent process to open, edit and export their photos, as they see them whilst editing. Why should they have to go through the contortions of making a virtual copy, turning on soft proofing, fiddling around to eliminate OOG warnings, etc, etc, just for that use?

If it were my proposal, it would be to simply be able to preset the default workflow and, if that implied soft proofing, then just allow all editing on the soft proofed version by default.

A.A.Milne, writing for Winnie the Pooh said

“When you are a Bear of Very Little Brain, and you Think of Things, you find sometimes that a Thing which seemed very Thingish inside you is quite different when it gets out into the open and has other people looking at it.”

There are a lot of us who are “bears of little brain” who really don’t want to have things more complicated than they already are.

5 Likes

That is not possible on a Mac. As far as I can tell, the default is always AdobeRGB.

Seems that DxO have mixed something up.

Soft proofing should never affect anything being output.

When going from a larger color space to a smaller one - like aRGB to sRGB - there’s two way of doing this.

The hard transformation way.
You convert/compress the larger color space to the smaller which causes the final file to be within the smaller space. This without any need on embedding any color profile or force the system on which it’s viewed on to display it a certain way.

The soft transformation alternative.
You do not compress or convert the larger color space to a smaller one. You simply embed the profile of the larger space into the exported image and let the receiving system do the transformation. The receiving system will then transform the larger color space defined by the embedded profile into it own or targeted system defined or chosen smaller color space.

This soft way ensures and enables the image to be output to different target systems with different quality or color spaces.
But it put some demands on the system to have a color management and force the users to manage it properly.

So according to your description DxO is doing a hard transformation from wide gamut to smaller one. That’s perfectly fine.
But they seems to be mess it up when the soft proofing profile is enabled and used.

There’s the bug as I see it.

2 Likes

They have created a problem and not a solution that most are willing to mess about with. Some users will have just a few images to output others large number. I usually have between 10 and 30 and to work on with the cumbersome process we are expected to fallow isn’t workable. So poorly thought out that DxO haven’t even had soft proofing opened in a virtual copy but leave you to do that. As said above the poor work flow of the current virtual copy adds even more to doing all this. I have 6 and am thinking of going back to 5 and if DxO don’t get it sorted I fear it would be the last version of PL I will be paying for.

2 Likes