PureRAW vs DeNoise AI vs NoNoise AI - Which Is BEST?
Вставка
- Опубліковано 16 сер 2021
- In this tutorial, I send the same high iso Raw file into Dxo PureRAW, On1 NoNoise AI, and Topaz Labs Denoise AI then compare the results.
Computer Specs:
iMac (Retina 5K, 27-inch, 2020)
Processor: 3.6 GHz 10-Core Intel Core i9
Memory: 64 GB 2667 MHz DDR4
Graphics: AMD Radeon Pro 5700 XT 16 GB
2TB SSD Drive
macOS Big Sur ver 11.5
If you're interested in DxO PureRAW, click on the link below:
tidd.ly/2S1WB8H
I'm sorry, I do not have a discount code for PureRAW.
For more info about NoNoise AI, visit On1 Software here:
bit.ly/2zJmA9I
Save 20% with Promo Code: AM17
Promo Code may not work on sale products.
If you're interested in Denoise AI by Topaz Labs, click the link below and scroll down to the Denoise AI. Bundles are also available toward the bottom:
bit.ly/3cDqa5J
To purchase, click on the link above then use my discount code to save an additional 15% off the price and on everything at the Topaz Labs:
AMDISC15
*Note that my 15% discount code may not work on sale products.
** I am an affiliate for all of the companies listed EXCEPT Affinity Photo.
Please read my Code of Ethics Statement:
onlinephotographytraining.com...
Anthony Morganti’s MUST HAVE applications:
At least one Non-Destructive RAW Editor
Lightroom - bit.ly/2zwQ0nW
Capture One - captureone.38d4qb.net/y1Djy
Luminar - bit.ly/2JUJxKw (Save with the Promo Code AM16)
On1 Photo RAW - bit.ly/2zJmA9I (Save 20% with Promo Code: AM17)
Exposure X6 - bit.ly/2U8UxrK (Save 10% with Promo Code: AnthonyMorganti)
PhotoLab 4: tidd.ly/2HhiN9X
At least one FULL Editing App:
Photoshop - bit.ly/2zwQ0nW
Affinity Photo - affinity.serif.com/en-us/photo/
My MUST-HAVE Plugins:
Topaz Denoise AI - bit.ly/3cDqa5J
Topaz Gigapixel AI - bit.ly/3cDqa5J
Topaz Sharpen AI - bit.ly/3cDqa5J
*Save 15% on all Topaz Labs apps - use the Promo Code: AMDISC15
Nik Silver Efex Pro 2 - tidd.ly/3dc4gYm
My MUST-HAVE Mac Utility:
CleanMyMac X - Info: bit.ly/3jtVpU3
To save 10% on your purchase of CleanMyMac X:
1. Visit the CleanMyMac X store: bit.ly/3bEMuMx
2. Scroll down to the very bottom of the page and click on "Enter it now" under "Already have a coupon code?"
3. Paste the promo code TONY10 and hit "Verify"
4. The prices will be reduced, and you’ll be able to select the license and proceed with the payment
** Note that all of the promo codes listed above may not work on sale products.
*** I am an affiliate for all of the companies listed EXCEPT Affinity Photo. Please read my Code of Ethics Statement:
onlinephotographytraining.com...
Please follow me on Instagram: / anthonymorganti
View my recommended gear list: www.amazon.com/shop/anthonymo...
As an Amazon Associate, I earn from qualifying purchases.
I use this software to record my screen and make parts of my screen enlarge and zoom out so you can see it:
Screenflow: telestream.pxf.io/DMrW2
Unsure of how to price your photography? Check the 2019 Guide to Pricing Your Photography:
amzn.to/2S1CiU7
I am an affiliate for all of the companies listed EXCEPT Affinity Photo. Please read my Code of Ethics Statement here:
onlinephotographytraining.com...
Thank you!
In this tutorial, I send the same high iso Raw file into Dxo PureRAW, On1 NoNoise AI, and Topaz Labs Denoise AI then compare the results.
Computer Specs:
iMac (Retina 5K, 27-inch, 2020)
Processor: 3.6 GHz 10-Core Intel Core i9
Memory: 64 GB 2667 MHz DDR4
Graphics: AMD Radeon Pro 5700 XT 16 GB
2TB SSD Drive
macOS Big Sur ver 11.5
If you're interested in DxO PureRAW, click on the link below:
tidd.ly/2S1WB8H
I'm sorry, I do not have a discount code for PureRAW.
For more info about NoNoise AI, visit On1 Software here:
bit.ly/2zJmA9I
Save 20% with Promo Code: AM17
Promo Code may not work on sale products.
If you're interested in Denoise AI by Topaz Labs, click the link below and scroll down to the Denoise AI. Bundles are also available toward the bottom:
bit.ly/3cDqa5J
To purchase, click on the link above then use my discount code to save an additional 15% off the price and on everything at the Topaz Labs:
AMDISC15
*Note that my 15% discount code may not work on sale products.
** I am an affiliate for all of the companies listed EXCEPT Affinity Photo.
Please read my Code of Ethics Statement:
onlinephotographytraining.com/code-of-ethics/
Anthony Morganti’s MUST HAVE applications:
At least one Non-Destructive RAW Editor
Lightroom - bit.ly/2zwQ0nW
Capture One - captureone.38d4qb.net/y1Djy
Luminar - bit.ly/2JUJxKw (Save with the Promo Code AM16)
On1 Photo RAW - bit.ly/2zJmA9I (Save 20% with Promo Code: AM17)
Exposure X6 - bit.ly/2U8UxrK (Save 10% with Promo Code: AnthonyMorganti)
PhotoLab 4: tidd.ly/2HhiN9X
At least one FULL Editing App:
Photoshop - bit.ly/2zwQ0nW
Affinity Photo - affinity.serif.com/en-us/photo/
My MUST-HAVE Plugins:
Topaz Denoise AI - bit.ly/3cDqa5J
Topaz Gigapixel AI - bit.ly/3cDqa5J
Topaz Sharpen AI - bit.ly/3cDqa5J
*Save 15% on all Topaz Labs apps - use the Promo Code: AMDISC15
Nik Silver Efex Pro 2 - tidd.ly/3dc4gYm
My MUST-HAVE Mac Utility:
CleanMyMac X - Info: bit.ly/3jtVpU3
To save 10% on your purchase of CleanMyMac X:
1. Visit the CleanMyMac X store: bit.ly/3bEMuMx
2. Scroll down to the very bottom of the page and click on "Enter it now" under "Already have a coupon code?"
3. Paste the promo code TONY10 and hit "Verify"
4. The prices will be reduced, and you’ll be able to select the license and proceed with the payment
** Note that all of the promo codes listed above may not work on sale products.
*** I am an affiliate for all of the companies listed EXCEPT Affinity Photo. Please read my Code of Ethics Statement:
onlinephotographytraining.com/code-of-ethics/
Please follow me on Instagram: instagram.com/anthonymorganti/
View my recommended gear list: www.amazon.com/shop/anthonymorganti
As an Amazon Associate, I earn from qualifying purchases.
I use this software to record my screen and make parts of my screen enlarge and zoom out so you can see it:
Screenflow: telestream.pxf.io/DMrW2
Unsure of how to price your photography? Check the 2019 Guide to Pricing Your Photography:
amzn.to/2S1CiU7
I am an affiliate for all of the companies listed EXCEPT Affinity Photo. Please read my Code of Ethics Statement here:
onlinephotographytraining.com/code-of-ethics/
Thank you!
Computer specs - That, yours is called a "beast". Operating a 64GB (very low latency) i7 with a 1080ti (11GB video RAM) - the configuration gave me 98th percentile in SPECmark last year - based on my experience I would try to optimize one thing in your configuration: I/O. Many tasks are I/O dependent. LrC (or CR) and Ps use cache disk space, the operating system does that, etc. If you run the computer in base configuration, then likely the 2TB SSD is on the so-called "South Bridge" chip. That is a switch that connects to 4 PCIe lanes with the CPU and has a bunch of I/O lanes on the other end. These 4 lanes do not compete with expansion boards like the GPU, but the connections on it compete with each other. But I only have one SSD on it so there is no competition, you may say. Well, yes and no. The operating system and your applications have all sorts of I/O tasks and when the files that belong to these tasks are all on the same "drive" then these I/O streams compete in that they cannot be parallelized - they are executed "concurrently" through time slicing.
To defeat this, I have my raw images on RAID 1 volumes (reads fastest) and catalogs and cache on RAID 0 (writes fastest). These are on separate PCIe channels - separate from the lanes to the South Bridge. If you, like I, have only one video card on a 16 "lanes" connector, you should have plenty of spare lanes that you and your CPU are not using. A simple low latency add in card can speed performance up. In my case significantly. Just make sure that the PCIe connector you use on the motherboard (mobo) does not compete with the one with the GPU card. The mobo may have 2 16 lane PCIe connectors, but if you populate both the mobo may decide to run both at 8 lanes to the CPU. That will be only a matter of choosing the right socket out of, say, 4 empty ones. My RAID 0 card is a "software RAID" only simple and basic thing and sports 4 drives and I may turn that into 8 one day. Note that SSD have extremely fast cache and that is limited. RAID 0 multiplies that cache and if you have enough members in the RAID, you can get sustained writes at cache I/O speed.
Thanks for a great overview!! I already have Topaz DeNoise, don't see any need to switch!!!
Thank you for a great comparison. I believe you are correct they had different programs have different outcomes depending on the image being processed. That being said, I'm a software junkie and have three of these programs. So far I've been very impressed with on one, but after this review, I may want to do more testing. Thanks for the great videos, keep up the great work.
Thanks Anthony, I love all your videos. I have both NoNoise and DeNoise and with some tweaking I find I get very similar results. As primarily an indoor sports photographer, I tend to use NoNoise because I can do my cropping there too unlike the others. This saves me a lot of disk space (which I know is cheap but it does add up). A Canon raw file of ~30MB comes back as a linear DNG after noise processing of ~140MB, so doing my cropping at the same time means I get a smaller DNG file back. Also, I find NoNoise does work well directly back and forth with Lightroom, the workaround to send files via the Plug in extras menu option is fine, you send CR2 out and get DNG back and still get the stacking you want etc.
Thanks again, you're my go-to UA-cam tutor!
P.S. I'm insanely jealous of your mac specs.....
topaz labs also allows cropping now and compatible with photoshop
Thanks for sharing this Anthony. It was very helpful. I own the DxO product and use it for images that have higher ISOs. It can do some miraculous things with noisy images. I really like the way you can integrate PureRaw into your workflow - the minimalist set of controls works well for batch operations. The downside for all three applications is that they create DNG files with RGB values (as opposed to raw files that store one value per pixel and then rely on the application of the bayer filter in conversion). As a result, the files are 3-4 times larger that the original raw file. This means that you'd better have some serious disk space for the output files. I only put noisy images through PureRaw because of the file size problem. It means that I have to have two workflows which is a nuisance.
you could not have explained better...sometimes I use topaz Denoise and Sharpen AI when I go see the size of the file it can get up to 120g sometimes easily lol
in my opinion Dxo is the clear winner. I already own topaz and it is great, but I did some iso 25600 milky way shots from a boat (therefore only 5 sec exp) and tried also DXO for comparison. same result as in your test, topaz maybe reduces the noise even a bit more, but the overall result is much better in dxo, the image looks far more natural, in the night sky espcecially also contrast and tones are much better. my workflow is to always start from DXO, because it has much better raw engine then adobe, then edit in LR and PS, and as a finishing touch, I make a run with topaz denoise, because it adds also nice and natural sharpening to the images
Many thanks for your, as always, very clear presentation. Any chance that you could re-run this test one day on a Milky Way shot with a dark background taken at ISO 6400 or 12800.
Perhaps there are horses for courses.
Good video - many thanks. I own none of these apps - I suppose I may at sometime in the future - they all looked pretty similar to me - I certainly couldn't see that the Topaz and Dxo apps "were considerably better" as you said. I guess a choice, for me anyway, may really boil down to comparative prices and how you get on with the different interfaces and tools. Still useful to see them in action, side-by-side (Rocky looks very smart in all them, btw!)
Study more carefully the improvements of Topaz Denoise AI over the others. Even with the auto settings it beat out the others and, as Anthony pointed out, he could have tweaked it and gotten even more out of it.
I own the latest versions of DxO Pure Raw and Topaz DeNoise AI. I've been using them on an M1 MacBook Pro for over 6 months. I only shoot raw and they are from a Nikon D850. After many many comparisons I can say definitively that DxO Pure Raw is much better for me. It doesn't eliminate all the noise but I can easily do the rest in Affinity Photo. Additionally, sometimes Topaz DeNoise AI yields strange color casts which then increases my workflow. Bottom line is that I've given up on Topaz and only use DxO Pure Raw.
Same color cast issue with me going from DeNoise AI to Affinity on a Mac IF I save as dng. Not a problem so far if save as tiff.
My experience and conclusion as well. Dx0 doesn't ruin my files with a lot of artifacts and HQ is more than sufficient for most of my files (< iso 3200 FF).
DxO has the best noise reduction on the market. There is nothing to add or take away. The rest are still smoking nervously on the sidelines. 😄 Imho.
I would have gotten a lot more out of this with a different subject than fur. It's a fine and random continuum (fur, more fur, blurry fur) that doesn't expose a lot. How about trying sculpted polished pure white marble against a neutral flat background, shot so the hard edge of the marble is dominant. Or finely finished wood like a guitar against a plain wall. Also, arrange the shot so the subject lays horizontal. That way when you use the before/after slider there are a variety of features presented at the same time.
Anyone contemplating DXO PureRaw; the company periodically runs limited time pricing down from $129 to $89 but check your eligibility depending on where you live.
i've owned nonoise myself but i have to agree with you, the first one of denoise is much more pleasing to the eyes. even though there's still a little bit of noise left, but i think this might be eliminated when it's out physically (if that make sense). i've found that noise on print is less detected than what you see on screen.
that said i'm very happy with my nonoise, so far i'm really pleased and quite blown away by the results. i haven't seen any weird artefacts as yet..at least i haven't noticed them.
Thanks!
Thank you John!!
Fascinating comparison. It is worth noting that there is all this competition relatively suddenly in the area of noise reduction. Topaz has released frequent updates to DeNoise over the last year. DxO took longer to upgrade from their PRIME and DeepPRIME noise reduction, but it was a big change. I think that ON1 benefited from coming in late, and I suspect they will work hard at keeping up. Ultimately, we all benefit from the competition, because all three are pretty amazing. I now use DeNoise almost daily and am used it, so when I evaluated ON1 I decided I liked Topaz’s approach better. But I’m sure each will nose ahead over the coming years to claim they are Number One.
May I ask whether all informations like lens, sensor etc are included in the output file? While LR or C1 do there usual lens corrections either the files coming from Denoise? Thx
So, I'm worried about the conversion inside LR. You indicated it converts the raw file to TIF. What am I losing if I do that? Dynamic range? What?
I use Topaz. Will I see a difference?
Just out of curiousity. PureRAW uses DxO lens adjustments. Do either of the other two use lens adjustment? Aslo when you loaded it into Adobe, did this then do a lens adjustment? If so, was the PureRAW dng file re-adjusted?
For Topaz vs On1 (both of which I use) I agree with your comment that the results are image specific. Also both products get frequent updates so comparisons become out of date very quickly. The spottiness that you demonstrated in On1 seems to have disappeared in my images (Fuji X-T3) with one of the 2022 updates.
Because of the batch processing capabilities of On1, it is my go-to solution. I unload all the images from a shoot into a single folder, batch process them all and then import both the RAW and On1 processed images into Lightroom; I've got both versions should I need to reprocess an image in Topaz.
On1 side comment, batch processing never seems to harm an image in any way. Low ISO images come thru just fine. No need to separate out the high ISO images.
For the X-T3 with On1 auto processing, I get what are to me, excellent results in all cases up to ISO 3200 and frequently excellent at 6400 but never above that. Considering that the sensor in this camera is 4+ year old technology, the results are amazing. Without On1 or Topaz, images above ISO 1600 are poor when just processed in Lightroom.
Processing (On1) some Canon R5 and R6 images downloaded from DPReview gave stunning results with the R6 1EV better than the R5. In both cases, 1 EV below the camera's top ISO gave excellent results.
Thanks for the comparison Anthony. I have NoNoise AI, but when I export a DNG and then open it in capture one, the output is always very saturated.
I really think the shallow dof makes the comparison stutter since no details are suffering from the nr in the out of focus, darker, areas. Personally I use dx0 to produce a dng when needed, gives me almost one stop starting point in C1 with files from D500. With D750, D850 the difference is more like 1/3 - 1-2 stop, but the optical correction have a lot of merits with dx0. Finally, in dx0 I can tweak the balance noice/sharpness to taste, a process of trial and error for different photos. Nice upload, thank you.
Hi, thx for your comment. May I ask whether C1 applies additional corrections when importing a ‚corrected DNG file‘ after passing through DxO PureRaw? Is the raw converter engine of dxo better than of C1? Thx!
@@michaelchrist4297 good question! I'll answer after investigating more thoroughly. And no, I don't think dx0 engine is better than c1's, it's the ability to handle noice that is the main reason I bought it. Both has its merits, but if I have to choose one of them I'd choose c1for many reasons. Workflow is one crucial
Great review Anthony as usual. Thank you for sharing. Just wondering for future reviews if you could include a comparison with Topaz Shaperning AI v3.x for its denoise capability as well? Thanks
I had performed the same comparison but I used 256000 in Low Light night scene images and no noise was terrible, denoise of and DXO did FAR better. Coul you run your comparison using high ISO in a low light setting? would really like to see your results as most of the time high ISOs are used in very low light settings
Dxo is the better software. No doubts
Interesting but how well would lightroom perform on its own? Then I would be able to assess if it was worth spending money.
I have On1 NoNoise, so I'm kind of "married to it".
In some images, I have seen odd artifacts that have shown up from time to time, especially in low light situations with deep shadows. Not as severe as your example, but I would like to see if this would be there with a different camera lens perhaps as well.
My wife and I have D500 bodies, and we both use 2 different lenses to shoot wildlife, and there is a difference in how On1 NoNoise works with the 2 lenses, even when we shoot the same exif values.
Same, I got it for 35 dollars, it was too good a price to pass up. I did want pure raw, but 90 dollars is the most expensive of the three.
@@erik3205 I don't blame you. I think I paid $60 for my copy. A little more than I wanted to spend, but the nice thing, is I'm not "renting" it each month. It's mine, unlike Lightroom or some other subscription based platform.
ON1's NoNoise will be integrated in PhotoRaw 2022 which is coming soon. I use 2021 at the moment and Topaz deNoiseAI (the last version), which works really well. I still think Topaz has the edge but given their upgrade price isn't particularly generous I may just wait for the inevitable ON1 upgrade offer (which last year was very cheap) and go with an all in 1 solution. While Topaz charge so much for upgrades after 12 months, they may just end up losing more and more custom.
For On1 NoNoise you can use as lightroom plugin and edit a copy from there as DNG. So possible to do that as raw wit NoNoise AI.
You said it is a comparison of raw files but it is not . The image was first in LR . A true comparison would be with the original Nef file from camera ?
Great comparison and video, thank you. My conclusion watching at 1440, taking DXO out of the mix (the poorest performer_ it is six of one half dozen of the other. But having more control is a plus. Suppose it is relative to the cost of the programs and what time editors are prepared to spend on an image.
Have you really worked with dx0? Aside from Nikons own software I don't know any better raw converter quality wise. Dx0 user for 10 years. Prefer C1 because of the workflow, but dx0 is fallback and backbone with suspicious files (read bad light/exposure).
@@805atnorafertsera6 Hi 805. My comment is based on the comparisons made in the video, can't recall if I implied I used DXO. I do have the other two programs and this evening I did a stand alone comparison between ON1 and Topaz Denoise AI, in this instance ON1 triumphed, but finished sharpening in Topaz Sharpen A1. I would like to make the DXO comparison.
@@masterwizard2881 ok, great info, thx. I recommend trying out Pure Raw for free, export some Dng's and compare with On1. I'm curious about your findings, last time I did it myself I didn't see any reason not to use dx0/pure raw. I do a lot of wildlife/birds and my high ISO files (above 3200) are often crunched I dx0 1st. The optical correction is fabulous.
@@805atnorafertsera6 Thanks for the thumbs up 805, I certainly will give it a whirl.
@@805atnorafertsera6 Hi 805. Finally got to test DXO Pure Raw. Does a good job, but was very slow on my sys compared to Topaz and On1. Also found it clumsy (perhaps with use I could improve). The total lack of control is also not for me. I could not get it to recognize my old Sony R1 raw files, not even the option to import the "drivers". I have been a supporter and active user of Nik since its first versions and will continue to do so. But DXO Pure Raw, will give it a pass.
I have Denoise and PureRaw and I tried NoNoise. I have to agree with you, Denoise is still the best in 99% of cases! Great comparison!
My files won’t appear in dxo pure raw for some reason
I've head from another photographer that Topaz Denoise leaves artifacts. Have you had this experience Here, Denoise AI softens the eye quite a bit. Noise is gone, but so is resolution. Pure Raw seems to keep the most resolution.
I own all the softwares and yes sir topaz is the best but does leaves artifacts if you don't know how to use or tweak incorrect , Pure Raw is my favorite , DXO deep Prime technology IMO is the same in Lightroom its based on Luminance and makes your image looking like a painting if you push to hard, the ON1 is. very fast very impressive but also can works towards luminance and I don't like because topaz success was when they found the formula to create a software that can remove noise without going towards the same route all the others softwares that was coping Lightroom with the noise reduction process with ,I use them all it depends on the picture on my mood lol
@@wesleybrilhante5989 Good. I had issues with ON1 and don't use it anymore. I really like Capture One, and I use Photoshop for things Capture One doesn't do. Gigapixel can leave artifacts if you try enlarging an image even more than Gp. did. But so far I haven't had any issues with any of the tree. Limits do exist, of course, as to what can be done. But I believe that Topaz has the best plugin aps out there.
Tried Topaz Denoise and won't open.
So found On1 NoNoise AI for €39 and works perfect! Loved it because of price, its also AI, and works as plugin or standalone. And it process true RAW files! And tried it on dark very noisy raw files and did a perfect job. I think its a clear winner. And it has this layer mask think I still need to figure out.
This review was done in August 2021 and while I don't use DXO I have both Topaz Denoise and ON1 NoNoise. To date (3-21-2022) the Topaz Denoise is superior. However, ON1 is coming out with a new version of NoNoise in April 2022 which could improve it substantially. It is not only reducing noise but keeping or improving sharpness that makes it a good choice. Old methods can reduce noise but they blur the photo too. Topaz publishes a new update every month or so. Now the version is 3.6.0
love your videos so thank you - I would disagree with you the NoNoise was the worst. I have both it and Topaz DeNoise. In my landscape images Topaz DeNoise is too soft even in the latest version. NoNoise on the other had keeps details better in auto and has the ability to modify that as needed. I think the big issue is I am watching your use of Topaz DeNoise as a plugin for Lightroom which forces it to be a tiff or psd file copy. I wondered on your other videos why you didn’t use Topaz DeNoise as a plugin , process the raw file and put the results out as a dng then work with it in Lightroom? Because I am a On1 Raw 2022 users with NoNoise part of the raw developing I have the full features of On1’s raw development mode (including camera specific setting) and NoNoise being part of the raw development. I seem to get much better results that way than using Topaz DeNoise or NoNoise as a plugin or even standalone. It makes for a much simpler workflow for me.
Thanks, but to get the most out of any program requires manual tweaking as the "AI" isn't always that "I"--it often requires human judgement to get the results you want. I have DxO PhotoLab 6 Elite (which includes DeepPRIME XD) and the Topaz IQ bundle (DeNoise, Sharpen, and Gigapixel) plus Photo AI. I generally prefer the results I can get from DxO for RAWs but it only works on RAWs. I also have some "vintage" low-res JPGs from the early days of digital as well as scans from prints/negatives/slides and Topaz can handle any filetype I have, so that's a plus, and it does a great job most of the time. As you point out, there are many user-adjustable variables and it can take some time (including a learning curve) to get what you want. A comparison with the latest iterations of these (and possible other) programs would be great, as the tech's constantly evolving. It is really amazing what you can do these days--ISO has become practically meaningless...
All of them are very effective and equivalent, however the simplest to work with is DxO. Only viewing at 200 or 300% will find small differences, but this is beyond the intended reality.
the latest version of topaz labs is the best
In your video I noticed that DXO makes lighter (highlights?) areas even lighter compare to DeNoise. I've got the same result on my test file. Anthony, could you please comment on this effect?
Good video Anthony. Thank you. See my comment below.
This is not a good example because high iso is used in low light and dxo is far better in that situation. You just need to disable the detail section in lightroom when you import it back and the photo does not look overprocessed anymore
DeNoise didn't seem to have a 'processed' look, it was a more natural-looking image. Of course, we aren't seeing it clean like you (compression).
I agree, Topaz DeNoise did the best job of all of them.
I haven't had access to PureRaw until just a few days ago because I shoot mostly Fuji CX-Trans. As a result I started using DeNoise AI both on JPEGs and converted DNGs. I like that I have several different starting points AND that I can fine tune the model that I like best. One thing that frustrated me with you video was that I never saw the in-focus areas (your cat's right eye). I get the point of concentrating on the OOF areas, but more critical to me is how the NR's AI treats the more important parts of the image. I've fiddled in the background of a photo only to find that the overall look of the photo wasn't what I look. Having started shooing with Tri-X and pushed Tri-X, I don't get into a lather about noise, but I really enjoy pulling details out of the noise, and DeNoise AI has allowed me to take another look at files going back ten or more years, I can't imagine using an NR plugin/App the didn't allow me the flexibility of DeNoise AI.
Your Must-Have plug-ins (the three Topaz and Silver Efex) are mine too. (Oddly, years back Topaz took the approach that simulating B&W was just a matter of adding grain, while there are significant differences in Silver Efex's approach to the chromatic sensitivities of different B&W emulsions)
I"m going to say it. Nothing beats DXO Deep Prime so why keep making these videos. I'm not trolling but its true. DXO technology is way ahead of the other companies.
It appears DxO don't pay influencers, so usually they skip PureRAW when comparing. Nice to see ANthony has bitten the bullet.
In testing I found Deep Prime oversharpens and there is no adjustment for sharpening.
The other thing I like about DXO is that you can get it to remove distortion etc at the same time and you can batch process - just send it all 50 of your photos and come back to nice clean distortion free images ready for the next step in your workflow
Unless it doesn' support your camera's RAW images, then you've just wasted your money.
@@Vargol Yep it dosn't support Fuji apparently. With less than 5% of the market, and falling, Fuji are investing in their healthcare business and not so much in cameras. If Fuji don't support themselves then it isn't surprising if software companies start to do the same. Weirdly, last time I checked Infinity it failed to support my Tamron 90mm Macro (F017). And an email asking about was ignored......
Is it better to put « cpu performance » or « gpu performance » in Topaz software (denoise, Sharp en, Gigapixel) ?
I have an 8700k (6c/12t) and an rtx 2080 (8go vram) for my desktop and, i have an 9750H (6c/12t) and an rtx 2060 max s (6go vram) for my laptop
Thanks a lot ! :)
If the object was to get rid of noise then your last part answered all you need to know. On1 noise got rid of all the noise whilst the other two didn't.
hasn't been the case for me, completely disappointed
#1 - Topaz DeNoise AI, #2 - DxO Pure RAW, #3 - On1 NoNoise
FIRST!!!!
First I think you are splitting cat hairs. You said the other 2 did not remove all the noise and that On1 removed all the noise but created an artifact. But you also said On1 had 3 other windows which used different "automatic" algorithm that did not remove all the noise either. Why didn't you use one of them which would have made them closer in output, choosing one of the outcomes from On1 that also left some noise in it. Severe noise choice may be for a different application that you are using it for, and the cat image did not seem to me to be a severe noise candidate. But in the end you are splitting hairs. They are close enough for me that workflow becomes the most important factor and because it is now built into On1 2022 which I am using, there is no reason to use anything other than on1 in my workflow. If I found that On1 did not give me acceptable results that I could see. I would use Topaz which I also have. I do not have DXO so that isn't even an option for me. But your video confirmed something for me. Even though you ranked on1 as the poorest outcome, you did not use the other algorithms that still contained noise. So I am going to say, close enough and continue to use On1. I don't think I will updating Topaz plugins any longer. It also makes me wonder if any of these companies actually code their own noise reduction software or if they are using canned coding from another company, perhaps the same company and then tweaking it when they put it in their application and we are just seeing different variations of the same base codeing.
Thanks!
Thank you so much Joel!!!