DxO or Topaz for removing noise from RAW files? You be the judge!

[@ashwathivalsaraj8964]

You are doing a wonderful job by giving Knowledge many thanks

[@cameron7889]

I didn’t even realize there was better software for denoising photos. I’ve always assumed Lightroom worked the best. Thanks for the video.

[@MatthewRaifman]

Definitely try the free trials of these (I’d recommend DxO). They are way better than Lightroom in my opinion. Thanks for checking out the video!

[@steveherman1283]

Thanks for putting the comparison together. My experience is similar, mostly I have been running RAWs through DXO before importing to lightroom. I am able to use a stop or two higher ISO. On occasion I still use Topaz.

[@davidnaylor6582]

Great video Matthew, thank you. I use Topaz at the moment and find the artifacts (out of the box settings) can be worse on certain colours, for example light brown plumage gets blurred.

[@MatthewRaifman]

Interesting! These software noise reduction packages are really exceptional these days, but I'm not surprised that there are some specific quirks based on what the machine learning algorithms can properly detect and address. I think it's worth doing side-by-side comparisons like this on your own images and evaluating if the DxO product is appealing to you. Happy shooting!

[@damonjalali8669]

one and only easy and working method

[@WernerBirdNature]

Hi Matthew, I fully agree with the conclusions of your interesting review, even when I got there via a different path. I'm using DxO PhotoLab 5 for most of my processing as alternative to the Adobe stuff. But I also have Topaz DeNoise and Gigapixel because my R6 was rather low in resolution for small distant birds. I'm always using DxO DeepPrime while processing, and only optionally I'm adding DeNoise or Gigapixel as an integrated final step of processing. However both Topaz apps pose a risk for including artifacts. I found out keeping the great DxO lens correction, but disabling DxO Lens sharpening when intending to later use Topaz is a key to reduce the risk of Topaz adding artifacts. Depending what you want to do in Photoshop, turning off lens sharpening in DxO PureRaw may also improve the end results. Regarding the colors: youtubers like Jan Wegener, Duade Paton and others say Adobe doesn't show the colors from Canon R3/5/6/7 correctly, which is one extra main reason they include DxO PureRaw in their Adobe workflow.

[@MatthewRaifman]

Thanks for the comment! Glad I’m not the only one. I’m curious if you’ve tried adobe Enhance Super Resolution and how it compared to Gigapixel. Seems like a similar concept but not sure if it’s the same quality. Even with the R5, I sometimes need to crop considerably making an enlargement option attractive.

[@WernerBirdNature]

@@MatthewRaifman I haven't tried those Adobe tools, but before purchasing Gigapixel I have watched several reviews on YT, and gigapixel at the time seemed far superior to Enhance Super Resolution. And within the first year, I'm getting free updates from Topaz and the quality of gigapixel has clearly improved with those updates. Just last week, I've upgraded my R6 to an R5 .. so you can imagine I needed enlargement options way too often ;-)

[@BenelliMr]

I have found that your workflow is the best also for my Sony A7RIV; with a fast PC all these corrections are almost instantaneous with 62 MP RAW files. I do as much as possible in DXO. before exporting as TIFF. Optical corrections are always welcome and often useful, despite the fact that I use prime lenses only. A prime Zeiss lens needs corrections too. So I strongly recommend Photolab where you can parameter everything

[@MatthewRaifman]

@@BenelliMr now I have to buy photo lab too?! Haha, thanks for the tip. I’ll check out a trial as I’ve only used PureRAW.

[@nordic5490]

Great review. Thanks

[@MatthewRaifman]

Thanks for watching!

[@jlawr4427]

After processing either completed image in PS, the saved jpeg is only 1/5 the size of the DxO or Topaz file, so not an issue for me. The huge file sizes turn out to be temporary.

[@MatthewRaifman]

That's really interesting. I don't personally mind the file sizes in general, as external storage is cheap and I'd prefer the photo quality. Also, doesn't hurt to be a bit more selective up front anyway. I suppose that has to do with the efficiency of jpeg compression? Thanks for the comment

[@davidcollazos8985]

It looks like the output from DXO were with Adobe color profile whereas the other with topaz SRGb color profile...according to the colors difference...

[@MatthewRaifman]

That's a very interesting observation. I will try to check if the information is embedded in the files and if I can confirm.

[@wildpatagoniafilms16]

Nice video... yeah agree about 'crispy" look, but that is why one shouldn't use the 'auto settings" because Topaz adds "sharpening" a lots of it! Or use auto, but lower the "sharpening " slider completely to the negative.... cheers!

[@MatthewRaifman]

Thanks! Yeah, there are a lot of permutations of the settings to try! After adjusting settings, have you found t Topaz to be better than DxO?

[@stefansandmeier2080]

Hey Matthew, nice video and comparison of the two tools. I prefer the DXO file too. My one criticism of DXO is that users cannot influence the settings in more detail, especially the amount of lens sharpening that is added in the process. I sometimes feel it almost too much for my taste. DXO's noise reduction is outstanding though as are the optical corrections. Your file was high ISO but shot in fairly good lighting conditions. Where DXO really shines is with files shot under bad lighting conditions. Although it cannot do magic it can turn an originally unusable file into one that you would be able to get away with if not printed too big.

[@MatthewRaifman]

Thanks! Hoping to do more things like this. I'm wish you. I also wish there was more flexibility in the way we toggled the options as well. Thanks for the comment

[@kazarian619]

After you purchase DXO, do you have to pay for any future updates?

[@MatthewRaifman]

Good question. You get free updates to the software under the perpetual license but then when they release new version you have to pay to upgrade (eg from DxO PureRAW 1 to 2).

[@kazarian619]

@@MatthewRaifman thank you. I’ve been looking for a good noise removal software and your video was very helpful. 👍🏽👍🏽

[@dannyhollis484]

I'm on an older ver of DXO PhotoLab, but the camera body selection under color rendering has a significant impact on exposure and saturation to greens. just thinking about the foliage in your image... wondering about the difference in presentation.

[@MatthewRaifman]

Interesting! I have my calibrating correctly (ie Canon R5 + whatever lens I'm using). I wonder if that's even something to play with in DxO Pure RAW as it isn't so much an adjustment, as a configuration. Would you suggest trying other camera's other than the R5 to adjust to color? I do find it interesting how different the greens are.

[@giottodiotto1]

i use NEAT IMAGE a older but allso very capable denoiser, i compared it to Topaz AI and certainly for the price difference very good, test that one out aganst these two "AI" programs

[@FrankLinssenPhotography]

Hi Matthew, I am testing DXO Pure Raw 2. I have processed DNG files that I exported out of: 1) LR 2) Capture One. The LR DNG's do get processed using the Deepprime option. The Capture One DNG's keep giving an error while using the Deepprime option?! Any idea why the LR-ones don't have issues while the Capture One files do give errors?

[@MatthewRaifman]

Hi Frank: that's interesting. What error are you receiving? I wonder if Capture One is applying some unique formatting to the file that DxO cannot read. Have you tired reversing your process? Opening the SOOC RAW files in DxO first, denoising, and then loading that DNG into Capture One? That might work better.

[@FrankLinssenPhotography]

@@MatthewRaifman i found out that DXO is only allowing DNG files created by LR or Adobe DNG converter

[@rleung2001]

I always ask myself the following two questions. 1. Who are the viewers of the pictures. If the viewers are using a mobile phone or a web browsing. Both image would look the similar. 2. Once you printed both pictures, does the bird and the background look different?

[@MatthewRaifman]

Makes sense. Thanks for your perspective.

[@Torsten8392]

R5 is a mirrorless camera not a DSLR

[@MatthewRaifman]

Agreed. Apologies if I misspoke.

[@mortenthorpe]

you do realize that high iso has far worse side effects than noise? the dynamic range goes way down in higher iso!

[@MatthewRaifman]

Yup! Absolutely. Often find myself shooting at ISO 12800 when doing wildlife with the R5 + RF 100-500 combo as that is a pretty slow lens. On the other hand, never had such good dynamic range at 12800 as I see with the R5. For this video, it’s the same shot compared so not particularly relevant though.