Are the Deep Prime XD algorithms the same in both programs? A photo buddy and I are getting far different processing times using comparable computers. We are trying to track down the reason(s). The corrections applied are just “Optical” and Deep Prime XD.
@Mangurian1 I use DxPL and not PhotoRaw but when you say comparable computers, does that include the graphics card? The GPU is arguably the most important element when it comes to exporting DP XD images!
Saying two computers are comparable doesn’t tell us much. If they are not identical including identical versions of the identical operating system and loaded software as well as an identical hardware configurations than there is the potential for performance differences.
I have been using the same Windows 10 I7 computer with 24GB of ram since 2016 and purchased PhotoLab 1 in Novenber 2017. I installed a GTX 1050 ti card in late 2020 for DeepPRIME support. Prior to the release of DeepPRIME, and after the installation of my upgraded graphics card once DeepPRIME was released, I have had no performance issues with PhotoLab…ever!
I know of many other long term Windows users here who also have not had significant performance issues. Clearly others are having significant performance issues. Suggesting that two comparable machines should yield similar results is meaningless. We have seen many cases here of two users with similar spec’d computers with very difference performance results.
Mark
I suppose that DxO uses the same algorithms in both PhotoLab and PureRaw. Nevertheless, there might be differences in how those algorithms are implemented, called and embedded.
In order to make your test results more telling, you should install DPL and DPR on both computers and run the same images with the same customising and export settings through both apps on both computers. This should give figuresyou can compare for the tests - and more than one test is needed if you want to find out what the impact of DPXD is.
I would say yes, they likely are. But there’s an important difference between DPR and DPL : in DPR, you can only specify the technology that should be used. You don’t have access to the sliders allowing to fine tune how the technology is applied. This probably doesn’t affect the processing time but this has a great impact on the generation of unwanted artifacts with high ISO, especially when using DeepPrime XD. By the way, enabling or disabling the optical corrections also affects the generation of these artifacts. The problem is still there in DPL 7. For most images, it doesn’t appear but it is indeed present with some images, especially portraits but not only. This issue has been already discussed in a few other threads.
This is why some of my colleagues are a little bit disappointed with DPR because there are artifact problems appearing with DPR when using DeepPrime XD that I’m able to fix in DPL by adjusting the Luminance and Noise Model sliders. So, DPR is providing only a part of the denoising functionality and IMHO this is another bad marketing decision.