HSL in Develop Personna

[malogrand]

Hi,

I would love to see a HSL tool in the Develop personna tab in Affinity Photo 1.9. The reason is the following. When you develop a raw file, it become a Pixel layer, and so it seems like the way the colors are managed is not good, or at least not perfect compare to editing a raw file in Develop Persona. I think putting an HSL tab in Develop Persona could help use get better results and also make the overall affinity photo experience way better.

 

Thanks

Malo GRANDBOUCHE (@malogrand)

[Pšenda]

Sorry, but what is the reason why you can't use HSL editing after developing a file in Photo Persona? Any modifications (including HSL) are not applied directly to the RAW data, but only after they have been developed and converted to image data - and it doesn't matter what Persona is called.

[malogrand]

@Pšenda, I am able to use HSL after developping it, just sometimes I would like to use HSL tools while working on my raw file in Developp Personna

[Pšenda]

41 minutes ago, malogrand said:

@Pšenda, I am able to use HSL after developping it, just sometimes I would like to use HSL tools while working on my raw file in Developp Personna

I understand, but I don't see any difference - adjustments are still made after the RAW data is developed. So what is the significance of these adjustments in the Develop Persona and not only in the Photo Persona, which is intended for all adjustments and manipulations with image data?

[malogrand]

Because

1 I find the HSL tool a little weird and really hard to use propely (which I think is due to the photo being converted to pixel from raw)

2 I want/need to make HSL adjustment while I am in the Developp Tab working on shadows and hightlights and more

[AlejandroJ]

On 10/31/2020 at 8:36 AM, Pšenda said:

Sorry, but what is the reason why you can't use HSL editing after developing a file in Photo Persona? Any modifications (including HSL) are not applied directly to the RAW data, but only after they have been developed and converted to image data - and it doesn't matter what Persona is called.

If applying an HSL variation to a Raw file or to a tiff file affects differently the result you can obtain for an image is something that I would want to know about (as I don’t).

When for example changing a colour or the saturation of a colour or its brightness when working with a raw file (with a program that would allow to do so), you are not affecting the colour of an image because there is no image (pixel image) yet, you are just altering the output that you’ll get for the pixel image that will be the result of the development process. (correct me if I’ wrong).

But when you apply an HSL variation to a riff file: are you reassigning values or are you changing the colour (pixel values) themselves?

Or if asked differently: Is it the same and with equal effects regarding the quality of the image will end up with, if one uses an HLS tool while working with a raw file (with a program that allows to do so) than if working with the HSL tool with a 16 bits tiff file?

[Pšenda]

20 minutes ago, AlejandroJ said:

correct me if I’ wrong

https://affinityspotlight.com/article/raw-actually/

[AlejandroJ]

58 minutes ago, Pšenda said:

https://affinityspotlight.com/article/raw-actually/

Thank you very much. Very interesting and enlightening article. Although most of the time it is comparing how would or would not affect an image if working with a raw file versus editing a jpg file (not versus editing a 16 bits tiff file after having processed the raw file).

Nevertheless, this three paragraphs from the article that I quote below, I think might be providing an answer or a clue to the answer to my question (are they in regards to for example changing the hue of a colour when working with a 16 bits tiff file or when already being in the Photo Persona after having developed the raw file?):

“What this all means for the end user is flexibility. A JPEG has already been mapped to a colour space and its white point has already been defined. Any colour values outside of the colour space range (e.g. sRGB/Adobe RGB) will be clipped and discarded. Add to this JPEG quantisation and compression, which further discards detail, and it becomes clear you’re working with a very limited subset of image information. You can only push the tonal information so far, and major alterations such as white balance prove very difficult to achieve. With RAW, you can push colour intensity values much further before clipping them, and changing the white balance produces a more natural, accurate result.”

“Having more values -regarding the ones from a raw file- translates to smoother gradients and variations in colour. You’ll notice this with blue skies, especially when you try and push them tonally: 8-bit JPEGs ‘fall apart’ very quickly, with banding becoming quite problematic, whereas 16-bit images maintain fine detail and can withstand heavier tonal work.”

“A big factor in shooting RAW is control. You retain ultimate control over your final image; its tone, colours, sharpness and noise profile. You get to decide how cool or warm an image is, bring back or crush shadow detail, saturate or drain the colour and find the best balance between sharpness and noise.”

[Pšenda]

1 hour ago, AlejandroJ said:

working with a raw file versus editing a jpg file (not versus editing a 16 bits tiff

Keep in mind, that Affinity doesn't work "in" any JPEG or TIFF !!!
JPEG/TIFF/.... are only file formats in which Affinity exports its working lossless and multi-bit data, which saves in a *.afphoto file.

Unfortunately, I do not understand how this affects the need/necessity of HSL adjustments in the Develop Persona, not in the Photo Persona.

[AlejandroJ]

Thank you again for your feedback.

In my case, I am not asking if this “affects the need/necessity of HSL adjustments in the Develop Persona, not in the Photo Persona”, Though I thought that what I am asking might be related to the subject.

My question is if there is any difference for the final quality that one can get for an image, if for example one changes the hue of a colour while working with a raw file, than if one does it in Photo Persona after having processed a raw file or if loading in Photo Persona a 16 bits tiff file. And I am asking because I don’t know for certain.

The thing is that I process my raw files in DxO, convert to tiff and then edit with Affinity. I try to do most of what I can while working with the raw file, even changing the hue of a colour if I need to. I have always thought it is better to fiddle with colours (in every respect), while working with the raw file. But I am not certain if it wouldn’t make any difference, specially if for example I want to change the hue of a colour, if doing so while editing a 16 bits tiff file. I suppose it is not the same, but I am not sure (though I know it is certainly not the same if editing a jpg file).

If it is the same, if wanting to change the hue of a colour after I have my tiff file, I could do it directly in Affinity Photo Persona with no fear of loosing quality for the image. If it is not the same, I would rather go back to my raw file and do the whole process all over again (go back to the raw file and process it again).

Would it be the same (indistintict) for the final quality of the image  if instead of loading a tiff file to Affinity Photo (one developed with DxO), I develop the raw file in Affinity (in Develop Persona instead than in DxO) and then develop the raw directly to Photo Persona and change the hue of a colur there? Would it make any difference in comparison to what I normally do? Would there be a benefit regarding the editing process in Photo Persona?

Sorry if I am not making myself clear. Sorry for my ignorance and sorry if I should have asked this in a different thread (is this the case?)

Thanks again.

[Pšenda]

37 minutes ago, AlejandroJ said:

I have always thought it is better to fiddle with colours (in every respect), while working with the raw file.

I thought that from what I wrote and what is described in the referenced affinityspotlight article, is obvious the RAW file/data is not image data, and therefore it is not possible to modify his colors (their H or S components). This data from sensor must be "everytime" recalculated and converted into image data.

But if I understand your query correctly - yes, the TIFF file/format is very suitable for storing "image" data, because it is lossless. Thus, developping RAW data in DxO, transferring it via TIFF to Affinity, is the correct/appropriate procedure. Doing it using JPEG (if it is not used lossless type) would be a mistake.
Just for interest, Olympus stores its RAW data (*.ORF) using the TIFF format, although it is not directly viewable/editable image data.

 

54 minutes ago, AlejandroJ said:

developed with DxO

If you are satisfied with the quality of processing/development in DxO, I would definitely not change the procedure.
As far as I know, the processing of RAW data in Develop Persona still has certain shortcomings, and the developers will hopefully improve it over time.

[malogrand]

I mean I did not really undertsand what happened here but Lightroom Classic does it and I would love to see it in Affinity Photo to edit my photo from develop Persona only. Plus the way the HSL tool work in Affinity Photo is not very intersting to me because it is very HARSH when I just want to make my green a little bit deeper for exemple (thats really hard to do with the HSL tool )

[malogrand]

It's really hard because most of the time it create artefacts.....

[AlejandroJ]

33 minutes ago, Pšenda said:

If you are satisfied with the quality of processing/development in DxO, I would definitely not change the procedure.
As far as I know, the processing of RAW data in Develop Persona still has certain shortcomings, and the developers will hopefully improve it over time.

Thank you very much for your answer. And I bet the developers of the program will improve the Develop Persona over time.

[Fixx]

Applying colour and tone correction in raw developing state supposedly applies those corrections to original raw data and outputs image with corrections applied during raw-to-image process. This is touted to be the merit of raw processing. 

What I have read in this forum that may not be the whole truth. It seems in AP raw data is predeveloped to very high fidelity image data to which colour corrections are applied. In that case it would make no difference if you do no adjustments in developing stage but develop image directly to 32-bit mode and apply corrections there. Am I right?

[malogrand]

honestly i don't care if there is no difference i just want HSL while i'm in Develop Persona please....

[AlejandroJ]

10 hours ago, Fixx said:

Applying colour and tone correction in raw developing state supposedly applies those corrections to original raw data and outputs image with corrections applied during raw-to-image process. This is touted to be the merit of raw processing. 

What I have read in this forum that may not be the whole truth. It seems in AP raw data is predeveloped to very high fidelity image data to which colour corrections are applied. In that case it would make no difference if you do no adjustments in developing stage but develop image directly to 32-bit mode and apply corrections there. Am I right?

For what I know (or I think I knew), corrections can’t be applied to the original raw data (raw data can’t be altered nor edited), but corrections of values (reassignment of values?) done during the development of the raw file will be applied to the image / file that will result from the development process (or something like that). The programs that I know that are specific or specialized in developing raw files, either add information to the raw file to let the program know what “corrections” (ok, they are not corrections because a raw file is not an image, but I don’t know how to put it in other words) have been applied, or others create a subsidiary file (I think it is called a “sidecar”) containing the information of what have been done regarding how the raw file ought to be developed, and this information is kept (separately, nor affecting the information of the raw file itself) so one can go back and forth applying “corrections” to the raw file or even close the program and open it again and go to the same raw file and the program remembers the “corrections” done, but the data of the raw file has not been nor will be altered. What you are seeing when you are opening a raw file is an interpretation of the data, not an image. (Did I get it right?).

But once you have developed the raw file, the result is a pixel image, being it a tiff or jpg or whichever file format you chose the exportation to be done to (an image), or a pixel layer in Photo Persona in Affinity after developing the file (again, an image to which you can apply non-destructive edits before exporting it as edited, but though it being a 16 bits per channel or whichever value, an image nevertheless, not the data of the shot as perceived by the sensor and as registered by the camera in a raw format file). (Is this correct or sort of?).

Therefore, if what I have said is correct (and I have always thought it is, but I am opened to accept a different explanation), developing a raw file is obviously not the same as editing an image file. Which doesn’t mean that maybe, if one fiddles with colour (hue, saturation and/or luminosity) while editing a 16 bits per channel image file or pixel layer, there might be not a lost or not that much of a lost of image quality (as could be noise being added, or banding, or whatever) or maybe it is better for the resulting quality of an image to fiddle with colour while processing the raw file or maybe it is indistinct to do it one way or the other. And I don’t know because I have never tried not to do as much as I could in this respect while processing a raw file and everything I have been able to find on the internet refers to the differences of processing a raw file against editing a jpg 8 bits image file, but not which are the differences between the former, and editing a 16 bits per channel image file (although some are obvious, as the fact that all the information, as could be the details in the shadows or in the bright areas, that is not taken advantage of or is not “rescued” when developing the raw file, can’t be used once it has been developed simply because it won’t be there, that much I know). And I don’t have enough knowledge about the subject so as to be able to deduce it myself,  nor to deduce which are all the differences in every respect (so, therefore, the reason for my original question in this thread).

10 hours ago, malogrand said:

honestly i don't care if there is no difference i just want HSL while i'm in Develop Persona please....

I think you have made your point. I myself would also like to have HSL included to the Develop Persona some time in the future, so I support your requirement. Maybe it would be added or maybe not.

But I think (and as for everything, I could be wrong) that if there is or if there is no difference, makes a difference. If there is no difference, maybe HSL might be included in Develop Persona anyway some time, but it won’t be a priority for many, and not for Serif. Other improvements might be added earlier (not that it won’t be done, but who knows). If there is a difference or a substantial difference, perhaps it might become more of a priority.

Meanwhile and for the time being, I will stick to Psenda’s recommendation regarding my particular case (which is not yours of course, so I understand and support your wish).

PS., And sorry if I might not be sufficiently clear when trying to express what I intend to say, English is not my native language.

[Mweber]

Hi
I've only just recently changed to Affinity photo - and like it a lot so far! - but I am dearly missing HSL correct in the Develop Persona too.

Seconding malogrand, it doesn't really matter if it makes sense technically, but for the workflow it makes heaps of sense.
Alone to be able to save all adjustments in one preset in one environment saves plenty of time if you work on a larger set of images.
Additionally if you adjust curves and white-balance then you would want to work on specific colors in the same session. Not having to go forth and back between Photo and Develop Persona.

In any case, thumbs up for the great piece of software!

Edited June 30, 2021 by Mweber
misspell

[Gallego]

I echo the desire to have it in the develop persona as well, while working on your RAW file. Once you are happy with the way the RAW file has been edited there may be little (or nothing) to do before exporting. It definitely makes the workflow so much easier and is perhaps the ONLY thing I miss from ACR. Basically the sliders are there in AP (just like in ACR), but only available if you choose to convert to gray scale - in ACR there is a checkbox to decide whether you want to use those sliders in color or in gray scale. C'mon y'all.... just upgrade this feature in the next version update, I would even pay for this.  

 

[NotMyFault]

Requesting to having HSL and other adjustments in Develop persona is a kind of "workflow compatibility" to other RAW development apps like Lightroom, which actually would cause harm to Photo.

Affinity Photo differs structurally and principally from those apps, and it doesn't really make sense as it would break some of the principal benefits of Photo.

In Raw development, these steps are essential:

1. Convert camera sensor specific RAW image data into a standard format suitable for further editing and interchange between apps
2. Do corrections like CA, defringe, Lens corrections which depend or benefit from having access to RAW image data
3. Do initial white balance, exposure and shadow / highlight adjustments to ensure the actually used dynamic range will fall into the range of the selected bit depth (32, 16, 8 bits per channel).
4. Choose the suitable format for later editing in Photo Persona.
5. Do everything else outside Develop Persona.

Why?

• All edits in Develop Persona are "destructive": The result gets baked into the image data. This can make a difference if you re-edit the same image again in Develop Persona. If you iterate and adjust e.g. brightness several times (double, half, ...) you will loose quality. In Photo Persona, you can (and should) use live filters, where the settings can be re-edited any time later. This helps to preserve image quality.
• Adjustments and filters from Photo Persona cannot be simply reused in Develop Persona, as the image data format is different. If you take a look at the functions which are available both in Develop and in Photo, you can see that the UI is actually totally different, and there are slight differences. This means every adjustment in Develop Persona was and would need to be re-programmed, causing huge effort without real customer benefit.
• Users of Affinity Photo should adjust their workflow / style of editing documents to the way Affinity Photo works best. It doesn't make sense on the long run to "work against the system", and keep all habits from a legacy products. I know this sounds harsh and against user interest at first, but adopting to current standards is unavoidable.
• Photo is no one-to-one clone of Photoshop or Lightroom. It is an App providing almost the same functionality for Photo editing, but offering a different UI, and requiring users to adapt in certain parts of their workflow. Forget about the legacy style of doing all in a RAW editor.
• To be totally honest: If you want an app working 100% like Lightroom or Photoshop, you might better stay with that legacy products.

What i said above only applies to the context of this (or similar) feature requests to double functions from Photo Persona in Develop Persona. Develop Persona still is lacking certain functions, has lots of unsolved issues, and could be improved in many aspects. But please keep it clean from Photo-Persona-doublets watering its purpose.

 

 

[RichardMH]

If you want that degree of colour control have a look at 3D LUT Creator.

[Largiewski]

As malogrand wrote. Even if the action will be the same. HSL panel in Develop will be a very convenient addition to working with RAW files.

[ultrainfra]

Forget HSL, what Affinity Photo really ought to do is implement the HCLab or HCLuv color models based on CIE Lab or CIE Luv respectively. HSL and HSV are kludgy hacks from times when  CIE color spaces were too slow to calculate. HSL and HSV have horrible perceptual uniformity! If any devs who work on the color adjustments module happen to see this post, check out this page for some information

[NotMyFault]

2 hours ago, ultrainfra said:

Forget HSL, what Affinity Photo really ought to do is implement the HCLab or HCLuv color models based on CIE Lab or CIE Luv respectively. HSL and HSV are kludgy hacks from times when  CIE color spaces were too slow to calculate. HSL and HSV have horrible perceptual uniformity! If any devs who work on the color adjustments module happen to see this post, check out this page for some information

Can you elaborate in more detail what you are expecting? Photo has LAB/16 Color Model, and can use Curves and Levels Adjustment in LAB/16 mode even if document stays in RGB mode.
 

Colors can be defined as LAB/16, too, but only on combined 2 La / Lb axis (not as 4 separate axis).

What is missing?

[ultrainfra]

27 minutes ago, NotMyFault said:

Can you elaborate in more detail what you are expecting? Photo has LAB/16 Color Model, and can use Curves and Levels Adjustment in LAB/16 mode even if document stays in RGB mode.
 

Colors can be defined as LAB/16, too, but only on combined 2 La / Lb axis (not as 4 separate axis).

What is missing?

What I mean is replacing HSL with CIE Lch , which can be based either on CIELAB or CIELUV. It would operate just as the HSL panel currently does, but the advantage is perceptual lightness and saturation are more stable when doing hue rotations.

To see this, go to this HSL slider online, leave saturation and lightness at a fixed value. Slide the hue. Notice how as the hue changes, the apparent lightness and saturation alters despite the L and S values remaining fixed. Now go here to the CIELch slider and do the same.  Notice how much more stable the brightness and saturation are when leaving the values fixed and adjusting only hue

When using the CIELch model, it is easy to see that colors with the same lightness value are equally perceptually light, and any colors with the same chroma are equally perceptually saturated. So, when using an Lch color panel adjustments can be made to an image with much smoother gradients.

So yes LCh can be just a transformation of the lab color model but it gives an interface with the ability to focus on particular hues etc just like HSL