Question for Those Who Use DxO Photolab As RAW Post Processor

[Joseph T Lewis III]

I have a perfectly good M4/3 kit, but continue to be intrigued by Fuji (especially the rumored X-H2 40 mp camera).

For years I've used DxO PhotoLab and its predecessors to post process my Olympus and Panasonic RAW files. I'm certainly not an expert, but I can muddle along well enough to get my pics the way I like them. Basically, I choose the setting / parameter I want to change (pre-set, WB, lighting, shadows, hue, saturation, contrast, sharpening, geometry corrections, etc etc etc.) and play with them one by one, until I finally arrive at the look I want. I then run the images through Deep Prime, and export them as JPG files.

If I finally break down and buy a Fuji, I'd like to continue to use PhotoLab to process my RAW files. Assuming DxO provides support for whatever I buy, would my approach of "select a parameter and adjust it until I am happy" still work? Would I still be using DxO the way I'm used to and more or less proficient with, or does the X-Trans filter require me to use PhotoLab in a totally different way?

Sorry if this is a dumb question, but I Googled around and was unable to find the answer to what seems a pretty straightforward question.

Thanks in advance to anyone who is kind enough to share their insight with me.

 

[Coldpaw]

No, it doesn't require you to completely rethink your processing in my opinion. Of course every sensor is different and requires some adjustments. I've been using DXO since years with my Micro Four Thirds kit, when the software was still calles Optics Pro. They introduced X-Trans support as beta last year when they released PhotoLab 5 and finalized the X-Trans support just a couple of weeks ago with the release of 5.2 - they also introduced more Fujifilm cameras to be supported, like X-Pro1 and X-E1.

The major difference I noticed is mostly to additional software they offer in form of FilmPack: I've noticed quite a number of their color film profiles seem to look rather off on my X-Pro3 files when compared to my Olympus files. Their new "Digital Film" profiles, which essentially mimic Fujifilm's Film Simulations, are fine though.

DXO fixed quite a few bugs since the initial release when it comes to X-Trans support and I can't complain about it. Ever since I've got my first Fujifilm cameras as of last year, I was relying on Capture One Express and was tempted in getting the full Capture One version, I completely moved away from that thought since DXO introduced X-Trans support.

 

[Raymond L]

Tom, suggest you visit the dpreview gallery and download some sample Fuji raw files and try them out for yourself

 

[yayatosorus]

I have a perfectly good M4/3 kit, but continue to be intrigued by Fuji (especially the rumored X-H2 40 mp camera).

Just FYI, there will be two versions of the X-H2 - one 26MP "speed demon" with a stacked sensor and one 40MP high res version with a standard sensor.

For years I've used DxO PhotoLab and its predecessors to post process my Olympus and Panasonic RAW files. I'm certainly not an expert, but I can muddle along well enough to get my pics the way I like them. Basically, I choose the setting / parameter I want to change (pre-set, WB, lighting, shadows, hue, saturation, contrast, sharpening, geometry corrections, etc etc etc.) and play with them one by one, until I finally arrive at the look I want. I then run the images through Deep Prime, and export them as JPG files.

If I finally break down and buy a Fuji, I'd like to continue to use PhotoLab to process my RAW files. Assuming DxO provides support for whatever I buy, would my approach of "select a parameter and adjust it until I am happy" still work? Would I still be using DxO the way I'm used to and more or less proficient with, or does the X-Trans filter require me to use PhotoLab in a totally different way?

As far as I know PhotoLab offers a unified interface for all supported cameras, regardless of brand, so there shouldn't be any problem with processing Fuji files.

One difference may however be output, i.e. your Olympus presets won't necessarily be optimized for Fuji, as you'll be using a different camera/lens combination.

As @RaymondL suggested in his comment, you should download some Fuji RAWs from DPReview's vast sample gallery and see how you like the output.

Sorry if this is a dumb question, but I Googled around and was unable to find the answer to what seems a pretty straightforward question.

No worries, there are no dumb questions!

Thanks in advance to anyone who is kind enough to share their insight with me.

 

[Phil1]

DxO now works (and very effectively) with Fuji files.

You may need to change your work flow slightly as Deep Prime works best by demosaicing and applying noise reduction to RAW files before any further processing.

All subsequent processing will benefit from working with files with no noise (or greatly reduced noise in the case of very high ISOs) and this will reduce artefacts. For example, it is possible for some sharpening S/W to interpret noise as part of the image and so produce artefacts.

With PL5 you can tune the Deep Prime to taste of course to cater for various ISOs. As always, expose (carefully) to the right to minimise shadow noise.

Have fun trying – you may be pleasantly surprised!

Phil

 

[Homo erectus]

I have a perfectly good M4/3 kit, but continue to be intrigued by Fuji (especially the rumored X-H2 40 mp camera).

For years I've used DxO PhotoLab and its predecessors to post process my Olympus and Panasonic RAW files. I'm certainly not an expert, but I can muddle along well enough to get my pics the way I like them. Basically, I choose the setting / parameter I want to change (pre-set, WB, lighting, shadows, hue, saturation, contrast, sharpening, geometry corrections, etc etc etc.) and play with them one by one, until I finally arrive at the look I want. I then run the images through Deep Prime, and export them as JPG files.

If I finally break down and buy a Fuji, I'd like to continue to use PhotoLab to process my RAW files. Assuming DxO provides support for whatever I buy, would my approach of "select a parameter and adjust it until I am happy" still work? Would I still be using DxO the way I'm used to and more or less proficient with, or does the X-Trans filter require me to use PhotoLab in a totally different way?

Sorry if this is a dumb question, but I Googled around and was unable to find the answer to what seems a pretty straightforward question.

Thanks in advance to anyone who is kind enough to share their insight with me.

I've used Capture One Pro since 2016 but I switched to DxO Photolab a few months ago. I gave PL a look when they added X-Trans support and it really impressed me.

You won't have any problems processing X-Trans raw files in Photolab.

The only issue I can see that might arise is if you really want to use Fuji Film Sims. Photolab seems to only have available the sim that was set in camera when an image was made. I haven't found a way to change to a different film sim in PL. bFilm sims aren't a part of my workflow and I just noticed this but didn't spend any time to see if it's possible to change the film sim in PL.

You need the current version of Photolab for X-Trans support.

Deep prime does wonders for high iso shots. It's very impressive. I find their lens profiles to be over-zealous in correcting vignetting though.

Hope that helps!

 

[Joseph T Lewis III]

No, it doesn't require you to completely rethink your processing in my opinion. Of course every sensor is different and requires some adjustments. I've been using DXO since years with my Micro Four Thirds kit, when the software was still calles Optics Pro. They introduced X-Trans support as beta last year when they released PhotoLab 5 and finalized the X-Trans support just a couple of weeks ago with the release of 5.2 - they also introduced more Fujifilm cameras to be supported, like X-Pro1 and X-E1.

The major difference I noticed is mostly to additional software they offer in form of FilmPack: I've noticed quite a number of their color film profiles seem to look rather off on my X-Pro3 files when compared to my Olympus files. Their new "Digital Film" profiles, which essentially mimic Fujifilm's Film Simulations, are fine though.

DXO fixed quite a few bugs since the initial release when it comes to X-Trans support and I can't complain about it. Ever since I've got my first Fujifilm cameras as of last year, I was relying on Capture One Express and was tempted in getting the full Capture One version, I completely moved away from that thought since DXO introduced X-Trans support.

Thanks for the informative and encouraging post, Coldpaw. I appreciate it. Now, "all" that has to happen is for the 40 mp X-H2 to be announced / produced / distributed / turn out to be a good camera, and for me to get off my butt and buy one!

 

[Joseph T Lewis III]

Tom, suggest you visit the dpreview gallery and download some sample Fuji raw files and try them out for yourself

Good idea Raymond, I will definitely do that!

--
Tom

 

[Joseph T Lewis III]

I have a perfectly good M4/3 kit, but continue to be intrigued by Fuji (especially the rumored X-H2 40 mp camera).

Just FYI, there will be two versions of the X-H2 - one 26MP "speed demon" with a stacked sensor and one 40MP high res version with a standard sensor.

Yes, I am aware of that. I tend to shoot static / very slowly moving subjects, so the 40 mp version (especially if it is cheaper than the fast one, as the rumors suggest) would probably be my choice.

For years I've used DxO PhotoLab and its predecessors to post process my Olympus and Panasonic RAW files. I'm certainly not an expert, but I can muddle along well enough to get my pics the way I like them. Basically, I choose the setting / parameter I want to change (pre-set, WB, lighting, shadows, hue, saturation, contrast, sharpening, geometry corrections, etc etc etc.) and play with them one by one, until I finally arrive at the look I want. I then run the images through Deep Prime, and export them as JPG files.

If I finally break down and buy a Fuji, I'd like to continue to use PhotoLab to process my RAW files. Assuming DxO provides support for whatever I buy, would my approach of "select a parameter and adjust it until I am happy" still work? Would I still be using DxO the way I'm used to and more or less proficient with, or does the X-Trans filter require me to use PhotoLab in a totally different way?

As far as I know PhotoLab offers a unified interface for all supported cameras, regardless of brand, so there shouldn't be any problem with processing Fuji files.

One difference may however be output, i.e. your Olympus presets won't necessarily be optimized for Fuji, as you'll be using a different camera/lens combination.

Yes, I know about the presets. I create new ones for each new camera I buy, using the Standard DxO preset as a base. I will do the same if I buy a Fuji.

As @RaymondL suggested in his comment, you should download some Fuji RAWs from DPReview's vast sample gallery and see how you like the output.

Sorry if this is a dumb question, but I Googled around and was unable to find the answer to what seems a pretty straightforward question.

No worries, there are no dumb questions!

Thanks in advance to anyone who is kind enough to share their insight with me.

Thanks again for your help; I appreciate it.

--
Tom

 

[Joseph T Lewis III]

I have a perfectly good M4/3 kit, but continue to be intrigued by Fuji (especially the rumored X-H2 40 mp camera).

For years I've used DxO PhotoLab and its predecessors to post process my Olympus and Panasonic RAW files. I'm certainly not an expert, but I can muddle along well enough to get my pics the way I like them. Basically, I choose the setting / parameter I want to change (pre-set, WB, lighting, shadows, hue, saturation, contrast, sharpening, geometry corrections, etc etc etc.) and play with them one by one, until I finally arrive at the look I want. I then run the images through Deep Prime, and export them as JPG files.

If I finally break down and buy a Fuji, I'd like to continue to use PhotoLab to process my RAW files. Assuming DxO provides support for whatever I buy, would my approach of "select a parameter and adjust it until I am happy" still work? Would I still be using DxO the way I'm used to and more or less proficient with, or does the X-Trans filter require me to use PhotoLab in a totally different way?

Sorry if this is a dumb question, but I Googled around and was unable to find the answer to what seems a pretty straightforward question.

Thanks in advance to anyone who is kind enough to share their insight with me.

I've used Capture One Pro since 2016 but I switched to DxO Photolab a few months ago. I gave PL a look when they added X-Trans support and it really impressed me.

You won't have any problems processing X-Trans raw files in Photolab.

The only issue I can see that might arise is if you really want to use Fuji Film Sims. Photolab seems to only have available the sim that was set in camera when an image was made. I haven't found a way to change to a different film sim in PL. bFilm sims aren't a part of my workflow and I just noticed this but didn't spend any time to see if it's possible to change the film sim in PL.

You need the current version of Photolab for X-Trans support.

Deep prime does wonders for high iso shots. It's very impressive. I find their lens profiles to be over-zealous in correcting vignetting though.

Hope that helps!

Thank you for your post, Homo Erectus, I appreciate the information / confirmation I can keep on using the processing software I've used for years and am comfortable with.

 

[Joseph T Lewis III]

DxO now works (and very effectively) with Fuji files.

You may need to change your work flow slightly as Deep Prime works best by demosaicing and applying noise reduction to RAW files before any further processing.

My understanding is that Deep Prime processing isn't performed until the subject file is exported. So, if I understand you correctly, I would have to apply Deep Prime to the RAW files and export them as DNG or TIFF files, and then loop back and do my normal editing on the Deeply Primed DNG / TIFF files?? Or, am I missing something really obvious?

All subsequent processing will benefit from working with files with no noise (or greatly reduced noise in the case of very high ISOs) and this will reduce artefacts. For example, it is possible for some sharpening S/W to interpret noise as part of the image and so produce artefacts.

With PL5 you can tune the Deep Prime to taste of course to cater for various ISOs. As always, expose (carefully) to the right to minimise shadow noise.

Have fun trying – you may be pleasantly surprised!

Phil

Thanks again, I appreciate it!

 

[Coldpaw]

Thanks for the informative and encouraging post, Coldpaw. I appreciate it. Now, "all" that has to happen is for the 40 mp X-H2 to be announced / produced / distributed / turn out to be a good camera, and for me to get off my butt and buy one!

 

[Phil1]

I'm afraid I can't answer your subsequent question with absolute authority. However, I would be very surprised if you couldn't do everything on PL5 i.e. it would be very unusual to have to have to export in say TIF and then re--import into PL5 to get the benefits of Deep Prime.

The reason I wrote as I did is that I use Lightroom for most of my processing, but even with that you can use PL5 as a plug-in which allows you to transfer the Deep-primed image back to LR.

In your case, if PL5 is your main processing S/W, I would expect all of this (de-mosaicing, Deep Prime and all subsequent processing) all to be done within PL5.

The best option would be to try it for yourself and for confirmation send DxO a query. They are pretty good at replying.

Finally, sorry if my lack of clarity misled you.

Good luck with it all!

Phil

 

[Joseph T Lewis III]

I'm afraid I can't answer your subsequent question with absolute authority. However, I would be very surprised if you couldn't do everything on PL5 i.e. it would be very unusual to have to have to export in say TIF and then re--import into PL5 to get the benefits of Deep Prime.

The reason I wrote as I did is that I use Lightroom for most of my processing, but even with that you can use PL5 as a plug-in which allows you to transfer the Deep-primed image back to LR.

In your case, if PL5 is your main processing S/W, I would expect all of this (de-mosaicing, Deep Prime and all subsequent processing) all to be done within PL5.

The best option would be to try it for yourself and for confirmation send DxO a query. They are pretty good at replying.

Finally, sorry if my lack of clarity misled you.

Good luck with it all!

Phil

Thanks for the comeback, Phil.

From my experience with Optics Pro and then PhotoLab, the Prime / Deep Prime noise reduction is done when exporting the files. A bit of poking around with Google just now increases my belief that it is not possible to do the de-noise processing without exporting the files.

One of the discussions I found was the following, which talks about exporting the files to DNG, and then doing the other corrections to the denoised / demosaiced DNG files.

Deep Prime Export to DNG

Thanks again

--
Tom

 

[Tim van der Leeuw]

I'm afraid I can't answer your subsequent question with absolute authority. However, I would be very surprised if you couldn't do everything on PL5 i.e. it would be very unusual to have to have to export in say TIF and then re--import into PL5 to get the benefits of Deep Prime.

The reason I wrote as I did is that I use Lightroom for most of my processing, but even with that you can use PL5 as a plug-in which allows you to transfer the Deep-primed image back to LR.

In your case, if PL5 is your main processing S/W, I would expect all of this (de-mosaicing, Deep Prime and all subsequent processing) all to be done within PL5.

The best option would be to try it for yourself and for confirmation send DxO a query. They are pretty good at replying.

Finally, sorry if my lack of clarity misled you.

Good luck with it all!

Phil

Thanks for the comeback, Phil.

From my experience with Optics Pro and then PhotoLab, the Prime / Deep Prime noise reduction is done when exporting the files. A bit of poking around with Google just now increases my belief that it is not possible to do the de-noise processing without exporting the files.

One of the discussions I found was the following, which talks about exporting the files to DNG, and then doing the other corrections to the denoised / demosaiced DNG files.

Deep Prime Export to DNG

Thanks again

Well, the denoise runs when generating output from the RAW file. It can be previewed on a small portion of the image.

At what point would you _want_ to run the noise reduction??

You can do all editing in DxO that you want until you are happy with it. You will see a preview of most changes directly, but the noise reduction only on a small portion, because it is very computationally expensive.

All raw file editors work the same basic way though: they show you a preview of your edits, and generate the final output when exporting.

Do you want to do all your editing in DxO? Or do you want to use it as a plugin for LightRoom / CaptureOne / other?

In the latter case, "exporting" from DxO is just the way to bring it back into LightRoom / etc. and is not a step that can be avoided.

In the former case, doing everything in DxO, don't bother first doing noise reduction, exporting, then doing more edits. You will have much better results and more flexibility doing everything all on the RAF file and exporting at the end, to generate your final TIFF / JPEG output.

 

[Joseph T Lewis III]

I'm afraid I can't answer your subsequent question with absolute authority. However, I would be very surprised if you couldn't do everything on PL5 i.e. it would be very unusual to have to have to export in say TIF and then re--import into PL5 to get the benefits of Deep Prime.

The reason I wrote as I did is that I use Lightroom for most of my processing, but even with that you can use PL5 as a plug-in which allows you to transfer the Deep-primed image back to LR.

In your case, if PL5 is your main processing S/W, I would expect all of this (de-mosaicing, Deep Prime and all subsequent processing) all to be done within PL5.

The best option would be to try it for yourself and for confirmation send DxO a query. They are pretty good at replying.

Finally, sorry if my lack of clarity misled you.

Good luck with it all!

Phil

Thanks for the comeback, Phil.

From my experience with Optics Pro and then PhotoLab, the Prime / Deep Prime noise reduction is done when exporting the files. A bit of poking around with Google just now increases my belief that it is not possible to do the de-noise processing without exporting the files.

One of the discussions I found was the following, which talks about exporting the files to DNG, and then doing the other corrections to the denoised / demosaiced DNG files.

Deep Prime Export to DNG

Thanks again

Well, the denoise runs when generating output from the RAW file. It can be previewed on a small portion of the image.

At what point would you _want_ to run the noise reduction??

You can do all editing in DxO that you want until you are happy with it. You will see a preview of most changes directly, but the noise reduction only on a small portion, because it is very computationally expensive.

All raw file editors work the same basic way though: they show you a preview of your edits, and generate the final output when exporting.

Do you want to do all your editing in DxO? Or do you want to use it as a plugin for LightRoom / CaptureOne / other?

In the latter case, "exporting" from DxO is just the way to bring it back into LightRoom / etc. and is not a step that can be avoided.

In the former case, doing everything in DxO, don't bother first doing noise reduction, exporting, then doing more edits. You will have much better results and more flexibility doing everything all on the RAF file and exporting at the end, to generate your final TIFF / JPEG output.

To hopefully answer your questions...

I was following up on Phil1's post above, which said the following:

"You may need to change your work flow slightly as Deep Prime works best by demosaicing and applying noise reduction to RAW files before any further processing.

All subsequent processing will benefit from working with files with no noise (or greatly reduced noise in the case of very high ISOs) and this will reduce artefacts. For example, it is possible for some sharpening S/W to interpret noise as part of the image and so produce artefacts."

I interpret Phil's comments as saying Deep Prime should be invoked first, before performing any other edits or corrections to the photo files. So, I was just trying to figure out how to do that. In my processing of non-Fuji files, I always do the Prime / Deep Prime processing as the last step before exporting the modified files into JPGs. (I use DxO as a standalone processor, having abandoned Lightroom for it a few years ago.)

Thanks

 

[Tim van der Leeuw]

Hm, to my knowledge, DxO performs the DEEP Prime NR as part of the demosaicing process, along with a number of other functions. Then it applies various edits on the result in a 2nd step I believe. But that is all internal to the way DxO functions and doesn't mean one has to separately invoke DEEP Prime NR, produce an intermediary output, and then do further processing.

I'll re-read again the post you mention.

Either way, you plan to use DxO PhotoLabs standalone on its own? And not as plugin for LightRoom?

 

[Tim van der Leeuw]

I think what it all comes down to is:

1. DxO noise reduction is done when exporting. It takes a lot of time, a real-time preview is not feasible.

2. Thus you can not preview before exporting how noisy the final photo will look like

3. You can only preview this on small areas of your image. You can drag around the viewport over the image to preview the effect of NR (and the rest of your edits) on different parts of the image.

Any editing that you do may (re)introduce noise into your image. So the best approach, in my opinion, is to do all editing on the original file and build up experience on how to get the best output.

I suggest that you download DxO for a 1 month free trial.

I don't know btw how much of an issue you now have with noise in your images?

 

[Joseph T Lewis III]

Hm, to my knowledge, DxO performs the DEEP Prime NR as part of the demosaicing process, along with a number of other functions. Then it applies various edits on the result in a 2nd step I believe. But that is all internal to the way DxO functions and doesn't mean one has to separately invoke DEEP Prime NR, produce an intermediary output, and then do further processing.

I'll re-read again the post you mention.

Either way, you plan to use DxO PhotoLabs standalone on its own? And not as plugin for LightRoom?

As I said in my previous reply to you, I use PhotoLab in a standalone mode.

--
Tom

 

[Phil1]

Hi Tom

I think you are right to say DP works on export but my (limited!) understanding is that it does this first before applying other changes e.g. exposure. If you are staying in PL5 there is no need to export a file (as DNG or otherwise) unless you want to say save a jpg or make a print.

Phil