[Wingincamera]

Most of you have probably read about the latest update to Adobe Lightroom with the new Denoise tool. I have been trying it out with DNG files from my KP and am a little disappointed in the results. I have been using photos with iso's from 1600 to 6400 and the results are usually no different from before. Some were slightly better, but not that much. I also have the Topazlabs Denoise and its results are far better than the new Lightroom Denoise. The Lightroom Denoise and Topazlabs Denoise both produce new files about three times the size of the original, Topazlabs a TIF file and Lightroom a new DNG file.
Has anyone else noticed this? Maybe it's just the DNG files made by the KP, but I would be interested to know from others with KP, K-3 or K-1 cameras have better results. At this time I will continue to use Topazlabs Denoise for high iso photos.

[vonBaloney]

Most of you have probably read about the latest update to Adobe Lightroom with the new Denoise tool. I have been trying it out with DNG files from my KP and am a little disappointed in the results. I have been using photos with iso's from 1600 to 6400 and the results are usually no different from before. Some were slightly better, but not that much. I also have the Topazlabs Denoise and its results are far better than the new Lightroom Denoise. The Lightroom Denoise and Topazlabs Denoise both produce new files about three times the size of the original, Topazlabs a TIF file and Lightroom a new DNG file.
Has anyone else noticed this? Maybe it's just the DNG files made by the KP, but I would be interested to know from others with KP, K-3 or K-1 cameras have better results. At this time I will continue to use Topazlabs Denoise for high iso photos.

It seems ok at getting rid of color noise, but not the actual grain of the noise. It definitely doesn't compare to Topaz.

[nixxo2002]

Most of you have probably read about the latest update to Adobe Lightroom with the new Denoise tool. I have been trying it out with DNG files from my KP and am a little disappointed in the results. I have been using photos with iso's from 1600 to 6400 and the results are usually no different from before. Some were slightly better, but not that much. I also have the Topazlabs Denoise and its results are far better than the new Lightroom Denoise. The Lightroom Denoise and Topazlabs Denoise both produce new files about three times the size of the original, Topazlabs a TIF file and Lightroom a new DNG file.
Has anyone else noticed this? Maybe it's just the DNG files made by the KP, but I would be interested to know from others with KP, K-3 or K-1 cameras have better results. At this time I will continue to use Topazlabs Denoise for high iso photos.

well , i think adobe in right direction ,with couple updates and few adjustments , soon it will be in same level with topaz . topaz denoise if you are using as a plug in in photoshop not creating a new file , it's just a new layer.
i used adobe denoise in ACR , and it does create a huge file with poor denoise compare to topaz . my files were iso 1600 -3200-6400 . the quality wasn't close to topaz . but it's just a begging .............

[vonBaloney]

I think it will take Adobe a long time to get there, but with this tech it is hard to say -- it may become so easy anybody can do it. But for adobe it is just another feature, for Topaz it is all they do, and they've gotten very good at it.

[Paul the Sunman]

As it happens, I just cancelled my LR subscription a couple of days ago, so I can't try it. However, this video from Matt Kloskowski seems to show it's even more effective than the Topaz Denoise.

You Tube

You Tube

However, it saves an additional DNG RAW file, which is unfortunate from my point of view. More clutter. You'll have two files: with and without noise reduction. I guess it may speed things up though compared to the non-destructive one-file strategy where the algorithms have to be reapplied every time you open the file (as in ON1 for example).

[ramseybuckeye]

When I first read that LR had "Denoise" I wondered if Topaz and Adobe were cooperating, strangely having the same name. I have tried the LR Denoise on a few K-3III files and really didn't see much change, but I can't say that I had any really noisy files to try, I'll have to look back to find some. In don't have Topaz to compare. There is also the "Raw Details" and "Super Resolution" check boxes. When I feel like messing with it I'll spend some time, just haven't had that time lately to deviate from the normal workflow.

[rpjallan]

Just remember that this is Adobe's first go at this. It will get better. I wouldn't use their noise reduction before. I generally use DxO PhotoLab, which I'm very happy with but I tried out some images from my old *istD & I was pretty impressed.

[Medex]

As already discussed on Pentaxians FB group, some Pentax cameras including KP, K-70, K-3III, KF, K-1ii use accelerator chip to reduce noise. RAW files already have reduced noise. And probably Pentax RAW files were not used for A.I. learning process. Therefore denoise results for some Pentax cameras are not good

[btnapa]

However, this video from Matt Kloskowski seems to show it's even more effective than the Topaz Denoise.

I do not have Topaz but I have ON1 and Luminar Neo. I processed a 400 ISO and 1600 ISO images shot with my Sony which is not that good when it comes to noise. I am beyond impressed with what PS did. Very effective indeed.

[StigVidar]

I have tried the LR Denoise on a few K-3III files and really didn't see much change

As already discussed on Pentaxians FB group, some Pentax cameras including KP, K-70, K-3III, KF, K-1ii use accelerator chip to reduce noise. RAW files already have reduced noise. And probably Pentax RAW files were not used for A.I. learning process. Therefore denoise results for some Pentax cameras are not good

My experience is that Lightroom denoise function works great on older files from Q7, K-5 and K-3. But it does not work any good at K-3 III files.

---------- Post added 26-04-23 at 08:54 ----------

And probably Pentax RAW files were not used for A.I. learning process.

According to a blog post from Adobe, they have not used any raw file for this. They have made an algorithm to add "standard looking" noise to copies of
clean images and then feed those images to the learning process.

[Medex]

My experience is that Lightroom denoise function works great on older files from Q7, K-5 and K-3. But it does not work any good at K-3 III files.

---------- Post added 26-04-23 at 08:54 ----------



According to a blog post from Adobe, they have not used any raw file for this. They have made an algorithm to add "standard looking" noise to copies of
clean images and then feed those images to the learning process.

At the moment I am ok with noise in my pictures. I use standard slider if I need. All that fuss about denoise AI or Topaz Denoise is irrelevant to me. Sometimes I even add noise and the post of Digitalis about grain in film is more interesting opportunity

[James O'Neill]

I have tried this with a range of pictures and found the results vary.
I went back to some pictures of New York form the *ist-D pictures and they cleaned up pretty nicely with the default value of 50% and even a little higher.
People shots with the K1 beyond ISO 6400 needed much lower values otherwise the result was large areas with no detail, but a stray hair (a crease under an eye) would be super sharp. turning the value right down -reducing, not eliminating the noise - helped a lot. But I'm finding some parts of a picture can clean up well and other parts not so well, which is frustrating - I looked making the pictures into layers in photoshop and blending different NR / denoise settings together but I didn't get results I was happy with.

[Wingincamera]

I did some more experimenting with both the new LR Denoise and Topazlabs Denoise. Before I had done some processing to the raw files before running them through the different denoise programs. This time I started with zero processing and obtained different results. I used the suggested settings in both Topazlabs Denoise and LR Denoise (50%). This time the results were much better for both programs. After processing both files were processed in the LR Develop, with sharping added to the LR file. Since the LR file is a DNG I was able to add Lens profile settings, which you can't do with the Topaz TIFF files. Results, both were pretty much identical, some photos looked slightly better in LR, and others in the Topaz tiff file looked slightly better, neither one enough for me to worry about. In both files, some parts looked slightly better/worst in different parts of the picture, but only noticeable if both are side by side in the compared view. When taking the different photos I took DNG plus Jpeg, with the in-camera anti-noise settings turned on for auto settings. I processed the jpegs the same as the dng files, plus adding a little sharpening and noise reduction in the LR develop section. The jpegs actually looked pretty good after a little processing, but you do lose color correction if the light changes. I will still continue using the DNG files for the ability to correct color.
For me and what I expect I will continue to use LR Denoise when needed, but for light stuff just use LR manual mode for noise reduction.

[kramedt]

My experience is that Lightroom denoise function works great on older files from Q7, K-5 and K-3. But it does not work any good at K-3 III files.

---------- Post added 26-04-23 at 08:54 ----------

I see the same thing. For DNG files out of my old k50 it seems to work great. For those from my k70 and k3 iii, it doesn't remove any noise. If anything, it may even increase what is already there.