Bonjour,
Merci pour votre réponse.
Je viens de refaire de nouveaux tests avec l’objectif Canon EFS 10-18mm f/4.5-5.6 IS STM.
Le module DXO est pris en compte.
Mes tests précédents ont été faits avant l’arrivée des nouveaux modules optiques.
L’EOS R7 ne pouvait pas reconnaître les nouveaux modules optiques inexistants à la date de mes tests.
J’attends la prise en compte des nouveaux modules DXO pour les nouveaux objectifs Canon RFS.
Cordialement,
Gérard
I am a big fan of DXO Photolab 5, which I used with Canon’s APSC M6 II for birds and wildlife (it’s the mirrorless equivalent to the D90). However, I recently upgraded to an R7, and while waiting for DXO to introduce R7 RAW support, I was using Canon’s Digital Photo Professional, which is too painfully slow for regular use. So I was excited when R7 support was added to DXO PL5. But to my surprise, the image quality from the R7 files doesn’t seem to match what I’m used to with the M6 II, consistently underperforming compared to DPP. What I’m mostly seeing is lack of sharpness, with artifacts creeping in while the images are still soft. I initially wondered if I was imagining it, since I would have guessed that the RAW formats for the the two 32 MP sensors were similar. But after several days tweaking DXO settings and experimenting, it seems real. I was wondering if anyone else had experienced similar, and more importantly, if they had found a solution. Given that DXO 6 is now out, I don’t expect to see further improvement in R7 support in version 5.
Hi and welcome,
Can you upload a RAW file from each camera along with the DOP files created by PL5 illustrating the difference in sharpness in a zipped folder to a file sharing service such a Dropbox, Wetransfer, pCloud, etc. and then post the link to it here. This way we will be able to see the problem and be better able to assist you.
I use the R7 on DxO 6 and haven’t had any problems with sharpness etc. Unfortunately, I can’t say for DxO 5 as I bought the camera once I knew DxO 6 supported it (and the lenses of course)
The artefact issue (had a couple myself) is common to a number of body and lens combo’s and is not just restricted to the R7
Photolab 6 works well for me also with my R7
The sensors might be the same, but maybe you used different lenses? Not all of these are created equal, therefore you might see the effect of a not-so-good lens…
Thanks for your prompt reply. You will find a number of files in the following Dropbox
In subfolder IMG_1499, you will find
1K4A1499.CR3 (R7 CRAW file)
1K4A1499.CR3.dop
1K4A1499_DXO.jpg (rendered by DXO)
1K4A1499_DPP.JPG (rendered by Canon Digital Photo Professional)
And in subfolder IMG_9598, you will find
IMG_9598.CR3 (M6 II CRAW file)
IMG_9598.CR3.dop
IMG_9598.CR3_DXO.jpg (rendered by DXO)
IMG_9598.CR3_DPP.JPG (rendered by Canon Digital Professional)
There’s a subject at the centre of each jpeg image that is probably over-sharpened in the DPP output, but I wanted to illustrate the difference.
Thanks again.
Thanks for the suggestion. According to Canon, the sensors are not quite the same. Though details are sparse, formal tests of dynamic range confirm the differences. It’s not a bad thought, but I’m seeing problems with my new RF 100-500 zoom, which is Canon L (pro) glass. In general, the image quality of the new lens is breathtaking compared to the adapted EF glass I was using with the M6 II. In both cases, I’m using DXO’s lens specific correction modules, though I don’t know if DXO had access to proprietary lens data.
Well, I compared the image from the R7 in DPL5, DPL6, LrC and DPP. Here is what I got:
DPL6 vs DPP
DPL6 vs LrC
DPL5 vs DPL6
Apart from slight colour differences that can be adjusted with the HSL tool in DPL, I see that DPL lifts the shadows just a little bit compared to DPP (I’ve adjusted DPP, so you don’t see a big difference) and LrC. The images look equally sharp in all apps when I disable all adjustments. This means that any difference we get is caused by whatever setting is changed in the app.
Summary: As of the time of this writing all apps can be used to develop images from Canon R7. Tools are different in the apps though, as are defaults, but with proper care and effort (and possibly experience), results can be fine.
I can’t tell you how much I appreciate the time and effort you put into this. I agree with your assessment that the all of your examples look the same, which is reassuring. But I think my DPP rendering (in the DropBox folder) is slightly sharper, particularly around the eyes and plumicorns (“horns”). I’m still not sure what I’m missing in DXO, but you are probably right on the experience question. Thanks again
This might well be so because DPP and DPL use different approaches for sharpening. In comparison, all apps seem to render similarly sharp when sharpening is disabled - like I did for the screenshots above. This means that DPP defaults to more aggressive sharpening than DPL. This is easily corrected in DPL by increasing “Lens Sharpness” or adding some “Unsharp mask”. Once you’ve found your recipe in DPL, you can save it as a new preset and use it as default, which you’ll set in DPL’s preferences/settings.
Remember: DPL default settings are meant to provide a starting point, they are by no means the way to heaven, unless you find the one preset that suits your intentions.
Your in-camera sharpening is set to 6 which is very high. DPP defaults to your in-camera settings. To match this level of sharpness in PL6 you would need to set Lens Sharpness to about 1.5. Turn off the Unsharp Mask, you don’t need it. It is only there for use with lenses that have no DxO profile and is far inferior to Lens Sharpness.
Yes, I was trying to exaggerate the difference to illustrate. But once I saw that comparable sharpness could be achieved with all the apps, I experimented enough to discover the equivalence with lens sharpness of 1.5 or so, which is further than I had previously pushed it. Thanks for explaining the difference between this setting and the unsharpen mask, which would not have been obvious to me. Since DPP also uses the lens-specific information, I wasn’t doing much lens sharpening in either, but clearly that was a mistake. Thanks again.
