Why is Neat video the best video denoiser?

[rrats]

SMDegrain is worse, they guy had only a Neat video 5 trial version, but even the Trial proved to be better than the above mentioned free solutions...

Imo neat video looks worse

[Truthler]

I already said, the sensor of DSLRs and mirrorless are not fit for video. .

See also: "Digital will never replace film! (or tape)". How'd that work out?

That's why their sensor are not able to fulfill the 4K IMA tests in sufficent degree. IMA tests tell everything.

You're a measurebater. You'll gladly spend all day shooting test patterns, walls, etc. But not content. Some of us are busy actually using gear, not testing it 24/7. Nor having orgasms when it passes "tests" (nevermind that the test may be skewed, seeking a certain result with limited objectivity, or sometimes complete objectivity without the necessary artistic subjectivity).

Aren't the individual pixels of a still camera too small to provide low noise signal?

If you think that, you're not very informed.
Look at the still ISO performance here: https://photographylife.com/reviews/nikon-z6-ii/4
Video is comparable.
It's amazing how high ISOs have come. I still remember having to push 800 film to reach grainy 3200. Now we have 3200 that looks like low ISO slide film. Something shot at a whopping ISO 12800 would only need minimal denoise.

Universal Classic Monsters are now out on 4K remastered beautifully. I wonder which denoiser/cleaning tools they used to clean these classics up?

I can assure you it wasn't Neat Video.

Yes, it was.

... and we're back to batshit crazy again.

their sensor suck in 4K video ... Videocameras rulez.

It's ironic (to me) how we have audiences that nitpick amazing video shooting tech, and in the same breathe will capture VHS with an Easycap and no TBC (and make unviewable crap). Or instead of Easycap, butcher it with HDMI adapters. Complete lack of video perspective these days. You can't have it both ways. The real issue is that some people just like to complain (or defend their purchases, especially cheapskates), and cannot simply value these as tools for tasks. So emotional, I didn't realize the world had so many teenage girls.

Some tools are good, some not:
Avisynth = excellent, NeatVideo = mediocre to bad.
Mirrorless vs. camcorder is model specific, hard to draw generalities.
Easycaps, HDMI adapters = bad; lack of TBC = bad.

Why is this so hard?
Rhetorical. It's not. You just want to make it hard, for whatever ulterior (and trollish) motive.

It is astonishing that you are older than me, but your own experience with digital video started later than mine. However it can not explain that you are one of the least knowledgeable person of this forum. After so many years... It is amazing in itself...

For the measurement of the real visible resulotion for image sensors IMA test was ISO charts were created. There is no other way to measure the horisontal and vertical resolution in the scientific world. You know, these were created by real scientists with academic rank, and who have more knowledge in their little finger than half of the forum members combined. You don't know even what is IMA ISO charts, you heard about it from me. Of course you look after it in the Google.

I don't care about the IMA TESTS of photographs, I interested only about the video resolution of the sensor in 4K or above. In that regard, the results of the still cameras are bad.

Cam Link 4K doesn't need TBC, TBC is for mechanical devices, even a teenager knows that. https://en.wikipedia.org/wiki/Time_base_correction
Camlink allows Time code (from cameras) to pass trough during capture. It is not only for multiple cameras, but helps a lot during capture too.

[Truthler]

SMDegrain is worse, they guy had only a Neat video 5 trial version, but even the Trial proved to be better than the above mentioned free solutions...

Imo neat video looks worse

Very beliveable. My posted pictures show the reality: It is better than SM Degrain by a long shot.

[rrats]

Have you even installed avisynth and vdub2 on your computer?

[rrats]

Cam Link 4K doesn't need TBC, TBC is for mechanical devices, even a teenager knows that.

Bruv he was talking about VHS

[lordsmurf]

It is astonishing that you are older than me, but your own experience with digital video started later than mine.

You keep repeating this, but it's false.

However it can not explain that you are one of the least knowledgeable person of this forum.

Amusing. This is essentially a child telling an adult that he doesn't know anything.

You know, these were created by real scientists with academic rank, and who have more knowledge in their little finger than half of the forum members combined.

This reminds me of Big Bang Theory.
Sheldon: "I know everything."
Penny: "Who's Radiohead?"
Sheldon: "Uhhh...??"

I'll all about science, but I'm not naive about methodology. Simply the scope of a test can skew results to not reflect real-world usage.

I don't care about the IMA TESTS of photographs, I interested only about the video resolution of the sensor in 4K or above. In that regard, the results of the still cameras are bad.

Same sensor.

Cam Link 4K doesn't need TBC,

Depends on workflow.

TBC is for mechanical devices,

No.

even a teenager knows that.

Right.

Camlink allows Time code

Let me guess. You think "time code" is somehow related to TBC? Because of the word "time". Amusing.

from cameras) to pass trough during capture. It is not only for multiple cameras, but helps a lot during capture too.

Why are you capturing from a digital camera?

[Truthler]

It is astonishing that you are older than me, but your own experience with digital video started later than mine.

You keep repeating this, but it's false.

However it can not explain that you are one of the least knowledgeable person of this forum.

Amusing. This is essentially a child telling an adult that he doesn't know anything.

You know, these were created by real scientists with academic rank, and who have more knowledge in their little finger than half of the forum members combined.

This reminds me of Big Bang Theory.
Sheldon: "I know everything."
Penny: "Who's Radiohead?"
Sheldon: "Uhhh...??"

I'll all about science, but I'm not naive about methodology. Simply the scope of a test can skew results to not reflect real-world usage.

I don't care about the IMA TESTS of photographs, I interested only about the video resolution of the sensor in 4K or above. In that regard, the results of the still cameras are bad.

Same sensor.

Cam Link 4K doesn't need TBC,

Depends on workflow.

TBC is for mechanical devices,

No.

even a teenager knows that.

Right.

Camlink allows Time code

Let me guess. You think "time code" is somehow related to TBC? Because of the word "time". Amusing.

from cameras) to pass trough during capture. It is not only for multiple cameras, but helps a lot during capture too.

Why are you capturing from a digital camera?

Again, you confused video resolution tests and photograph resolution tests, They are not the same!

IMA test charts had no competitor. It is the only scientific method to measure real resolution, your layman method based on bias personal opinion, thus it is fully irrational.

Still cameras have little individual pixels , which generates noise, (you can not change the laws of physics) what the camera try to remove, so hard denoising process run in the background, which makes their video image less detailed.

I proved with Wiki link that TBC is a name only for mechanical devices. Time Code is the correct word for modern cameras (which do not use tapes), it help to use multiple cameras, and for single cameras it is good for even more proper maitenance of FPS timing and audio timing and sync too.

[DB83]

How about answering the last Q. (and, again, you are ignorantly quoting full replies)


I see two scenarios.


1. You take your laptop with you to record direct to that with camlink. Not that practical since your claim of unlimited recording is only as good as the battery in the laptop.
2. You transfer from the camera with camlink at home.


And what is the capture software ? Let me guess that. The letters SOB come to mind but not in that order.


And neither are surely necc or does not the camera come with any means to copy a recording straight to PC with a codec good enough to allow you do even attempt the de-noising should it even be required ?.

[rrats]

This reminds me of Big Bang Theory.
Sheldon: "I know everything."
Penny: "Who's Radiohead?"
Sheldon: "Uhhh...??"

Man, Hannah Montana knows more than Sheldon..

[rrats]

Truthler, I don't f-bloody care which camera I am using. I will be satisfied as long as lower the contrast and tune down the sharpening. I don't care about it's ISO performance and other rubbish.


Let's not forget you used to tell us that you record in highest ISO just to denoise it. The way you started shilling your camcorder by pointing out minimal differences is just funny

[Truthler]

How about answering the last Q. (and, again, you are ignorantly quoting full replies)


I see two scenarios.


1. You take your laptop with you to record direct to that with camlink. Not that practical since your claim of unlimited recording is only as good as the battery in the laptop.
2. You transfer from the camera with camlink at home.


And what is the capture software ? Let me guess that. The letters SOB come to mind but not in that order.


And neither are surely necc or does not the camera come with any means to copy a recording straight to PC with a codec good enough to allow you do even attempt the de-noising should it even be required ?.

I always knew, that you aren’t the sharpest knife in the kitchen cabinet drawer.


Again, I spoke about normal video which is recorded in memeory cards , like SD cards. recording to SD cards in camcorders like Ax700 is unlimited. However recording with still cameras is limited, it generates newer and newer files , and they like to overheating.

[Truthler]

Truthler, I don't f-bloody care which camera I am using. I will be satisfied as long as lower the contrast and tune down the sharpening. I don't care about it's ISO performance and other rubbish.


Let's not forget you used to tell us that you record in highest ISO just to denoise it. The way you started shilling your camcorder by pointing out minimal differences is just funny

Jesus!



It is even ridiculous! You even confused the camera ISO with the organization ISO. LEARN: https://en.wikipedia.org/wiki/International_Organization_for_Standardization IMA TEST resolution CHARTS are registered ISO standard products.

[rrats]

truthler

[DB83]

Well define 'unlimited' since in my dictionary a recording to a SD card is restricted ie 'limited' to the size of the card.


But again you shift the goalposts. In one breathe you write of SSDs and now you write of SDs. Make your friggin' mind up of what is what.


So, again, confirm your transfer process from the SD card since in my experience you really do not need that convoluted recapture.


And by all means insult me as well. Just as you have done with most who have dared to challenge your warped thinking. But keep that up and I will have no hesitation to report you for it.

[rrats]

I think he's talking about some old DSLR limit in which you can record only for 30 minutes.


Also @truthler, try changing the lens on your camcorder, I dare you

[Truthler]

Well define 'unlimited' since in my dictionary a recording to a SD card is restricted ie 'limited' to the size of the card.


But again you shift the goalposts. In one breathe you write of SSDs and now you write of SDs. Make your friggin' mind up of what is what.


So, again, confirm your transfer process from the SD card since in my experience you really do not need that convoluted recapture.


And by all means insult me as well. Just as you have done with most who have dared to challenge your warped thinking. But keep that up and I will have no hesitation to report you for it.

Unlimited, swappable cards with two slots. Imagine 2X1TB card, you can record days continously. The still camera split the files into 10-15-20 min slices, however camcorders can record even 48h SINGLE files too... Don't forget the overheating of still cameras. either their small sized individual pixels which must have strong denoising process..

[Truthler]

I think he's talking about some old DSLR limit in which you can record only for 30 minutes.


Also @truthler, try changing the lens on your camcorder, I dare you

Why shld I cahnge? You can not attach true professional lenses to your DSLR....

Try to connet that to your still camera: https://www.3dbroadcastsales.com/canon-cj18ex7-6b-kase-4k-uhdgc-portable-eng-broadcast-lenses

Its cover comes off from the weigth. Still cameras are NOT ideal for outdoor filing, due to their shit form factor. Let's don't forget the small shit individial pixel sizes of its sensor and the heavy denoising processes...hahaha

Simply they were not designed for outdoor.

[rrats]

You are probably butthurt 'cause you spent your money on overpriced camcorder instead of a good mirrorless.

[Truthler]

You are probably butthurt 'cause you spent your money on overpriced camcorder instead of a good mirrorless.

Mirrorless is not for video, it is for still pictures. I would wonder If you can sgot better qualiry video than me.

Buy your mirrorless camera, and try to shot in the evening...hahahaha

[rrats]

Ax700 has 13,2 x 8,8 mm sensor, my mirrorless has 22.3 x 14.9 mm

Ax700 has 14MP, my mirrorless has 24.

I can change lenses on my mirrorless, you cannot on your camcorder.

My mirrorless costed 650€, yours camcorder costed you 1400€.

Please explain me how your camcorder is "superior in every way".

[rrats]

Buy your mirrorless camera, and try to shot in the evening...hahahaha

I made some pretty good shots in the middle of the night, with the only light source being streetlight.

[rrats]

Also it's ironic how truthler started with "I shoot in high ISO and high shutter speed" and ended up with "look at ISO charts"

You can not attach true professional lenses to your DSLR....

First, it's mirrorless, second, you cannot attach anything to your camcorder, except for maybe fish eye lens add-on

[Cornucopia]

I think he's talking about some old DSLR limit in which you can record only for 30 minutes.


Also @truthler, try changing the lens on your camcorder, I dare you

Why shld I cahnge? You can not attach true professional lenses to your DSLR....

Try to connet that to your still camera: https://www.3dbroadcastsales.com/canon-cj18ex7-6b-kase-4k-uhdgc-portable-eng-broadcast-lenses

Its cover comes off from the weigth. Still cameras are NOT ideal for outdoor filing, due to their shit form factor. Let's don't forget the small shit individial pixel sizes of its sensor and the heavy denoising processes...hahaha

Simply they were not designed for outdoor.

That is wrong. Quite the opposite in fact. Yes, you can attach pro lenses to dslr. And there is a whole line of converter/adapters built and sold to allow for it and all its brand & size variability.

https://www.google.com/amp/s/www.bhphotovideo.com/explora/amp/photography/tips-and-sol...-lens-adapters

Weight can be an issue. But you need to think outside of the box and realize that the dslr does not need to be the object that is solely doing the supporting. There are support sleds for those ridiculously large and heavy lenses that hold the adapter rings, and those do the job of holding up BOTH the lens and the camera.

But how about you get back on topic for this thread and provide those capture steps, intermediate steps, AND SETTINGS, in DETAIL, that we have requested from you multiple times?

Scott

[Cornucopia]

You are probably butthurt 'cause you spent your money on overpriced camcorder instead of a good mirrorless.

Mirrorless is not for video, it is for still pictures. I would wonder If you can sgot better qualiry video than me.

Buy your mirrorless camera, and try to shot in the evening...hahahaha

That is also wrong.

https://www.creativelive.com/blog/using-best-cameras-for-night-photography-settings/

These are cameras they RECOMMEND for evening/night shooting. Notice that down on the list are a number of mirrorless cams as well as DSLRs.

And why are you even saying something like "mirrorless is not for video"? ALL VIDEO CAMERAS USE MIRRORLESS style of tech. Maybe you aren't familiar with how they are built?

Also, and more importantly, one area that still cams excel that video cams cannot do is long exposure shots in low light. Like 2sec-8sec, or longer astrophotography, or long range night shots of skylines and traffic. Video and cinema cams are limited in their exposure times by the shutter speed - they can go shorter (using "shutter angle" adjustments), but they cannot go longer than 1/24th or 1/24th or 1/30th of a second.

Scott

[poisondeathray]

It is not for eye pleasuring, but the reproduction of reality, as much as possible. I have a lot of details geometrical forms on the building, which is covered by noise on your version.
The original building has no noise....

So the goal is "reality"...

Yes - original building has no noise - but oversharpening halos, blending artifacts, ghosting do not occur in real life either - all these are "noise" since they are unwanted signal. Those are not representative of "reality". Sharpening residual noise accentuates noise artifacts - so you're adding and amplifying noise in a sense

This is supposed to be about DEnoising, not "ADD-noising"

Read up on "oversharpening" and "halos" on photography forums , books, web articles - because this is a common theme you seem to not be getting. Everyone is guilty of oversharpening when they start out. If they say otherwise, they are lying.

If I want to upload videos for youtube, sharpening is necessary, since the low bitrate will reduce the sharpness and remove many side-effects of sharpening. If I want to recomress the original lossless video into lossless HEVC, the compression will remove a lot of side effects. (Not all side effects, but most of them)

Sharpening actually impairs compression efficiency. If you sharpen a PNG, the sharpened output takes more bitrate if you use the same PNG compression settings. Try it. At low bitrates such as YT uses, a heavily sharpened version usually causes more problems. Think of it as wasting bitrate on those artifacts. IF you had unlimited bitrate - sure you could reproduce those artifacts exactly how you want to - but since you don't - the entire image, and especially edges, degrade more than if you didn't sharpen. One of the YT "optimization" strategies is temporal NR and blurring. Yes - blurring, not sharpening. It makes it easier to compress. Of course these are generalized comments - but these are fundamental basics for low bitrate lossy video compression, not just YT



Look at the railing , there is a ring of "white" around the black railing. Or black jeans have glowing ring of white around it. Does that happen in real life ? Where can I get black jeans that glow white - very cool effect for the kids on halloween, but unrealistic. By sharpening so much, you're enhancing the noise on some sections such as the wall. If it was denoised properly, then oversharpening wouldn't be as bad. I'm not saying don't sharpen... just do not sharpen so much.


[Attachment 61198 - Click to enlarge]

"The original building has no noise...." - then what is this ? Walls can have noise but buildings can't ? Is that what it looked like in real life ? Was a cloud of grain following the guy walking down the stairs? It's like "pigpen" in Charlie Brown .


[Attachment 61199 - Click to enlarge]



The black line artifacts around edge of frame borders and the artifacts selur pointed out are from temporal stabilization edge artifacts, and/or too high temporal denoising settings. It's a bit distracting, but in real usage, you'd probably keep some stabilization and zoom in a bit instead of reverting the stabilization. It's introducing more problems with artifacts, and probably not worth it in that section. You don't need to stabilize the input, neat video does good job "out of the box". You can read the wall "visitor center" by default just sampling that area. Pre stabilizing is probably not a trade off most people would make if it generated ghosting and edge artifacts like that


The oversharpening wouldn't be as bad if it was cleaned properly. Now there are "splotchy" noise patterns - it's not consistent. The ghosting artifacts, splotchy noise do not occur in real life, do they ?

[Attachment 61200 - Click to enlarge]


The avs version of SMDegrain does not exhibit as bad ghosting frame border issues, when the stabilization method uses the "poster frame" lock method I mentioned earlier. So even though frames +/- x might have "black borders", smdegrain is smart enough to reduce the issues even when using large radii (even 20 or 30). The havs vpy version is limited to 6, the G41Fun version is limited to 24 and is "cleaner" than avs version at equivalent TR but exhibits more ghosting. I'll post more on this later, but smdegrain+pre/post is a viable method of clean border fill that you can use with stacking (there are other more complicated ones , that involve multiple tracking layers and 3D camera projection)

If you use SMDegrain , default settings and don't adjust, or appropriate settings or filtering, you're not going to get good results. But the same thing can be said about Neat Video or any filter



Compare to neat video using basically default settings, no sharpening; and use the original as reference.

original

[Attachment 61202 - Click to enlarge]

truthler

[Attachment 61201 - Click to enlarge]

neat video "default", no sharpening

[Attachment 61203 - Click to enlarge]



Look at the ghosting. The black pole in front of the lady is partially transparent. There are overharpened artifacts, noise enhancement on the road, splotchy grain. The uneven-ness of denoising makes it look bad on some frames - the "splotchy" patterns of partially clean, some large noise, some small, and then sharpened. In real life you don't see patches of smooth and patches of noise on the same road at the same time. Does that look "realistic" to you when you were there? Uniform noise/grain, or smooth but less detail would be more preferable to splotchy pattern . The background wall has sharpened noise artifacts - they are not real details - and you're not supposed to see clear details on objects farther away in real life, unless you see even more clear details in those objects which are closer in proximity, such as the road more clearly - and that contributes to the unnatural look - it looks "wrong". There should be depth aware sharpening, BG sharpen less than the FG, because that's how vision works in real life. Also notice the white halo around the chain link - clear signs of oversharpening. Again, I'm not saying don't sharpen, I'm saying just don't sharpen so much, because it looks terrible, is enhancing noise and causing problems


There are other various temporal artifacts, missing objects eg. such as poles, blurred feet on people, guy walking in park talking on phone is missing leg - but any denoiser using strong temporal settings will have those sorts of artifacts, unless you use some sort of protection like motion masks, or weaker temporal settings. As good as Neat Video is (when used properly) - the temporal denoising tends to be not as good as some other temporal denoisers (or you have to be very careful about the strength when using NV) . NV has improved a LOT since v2, but temporal denoising is still a relative weakness (which you can improve for any denoiser by pre- stabilizing). On the other hand, because NV can completely overdenoise (noise and detail) easily, you can add back noise and detail in layers - so if you use NV in more advanced manner it can still produce very good results.



more to come later...

[Anonymous5394]

So Poisondeathray's 'neat' reality check has kicked in, showing that Neat Video isn't the best video denoiser after all!😂

Maybe one should stick with Vague Denoiser, because everything up until now has been 'vague as hell!'

[Truthler]

It is not for eye pleasuring, but the reproduction of reality, as much as possible. I have a lot of details geometrical forms on the building, which is covered by noise on your version.
The original building has no noise....

So the goal is "reality"...

Yes - original building has no noise - but oversharpening halos, blending artifacts, ghosting do not occur in real life either - all these are "noise" since they are unwanted signal. Those are not representative of "reality". Sharpening residual noise accentuates noise artifacts - so you're adding and amplifying noise in a sense

This is supposed to be about DEnoising, not "ADD-noising"

Read up on "oversharpening" and "halos" on photography forums , books, web articles - because this is a common theme you seem to not be getting. Everyone is guilty of oversharpening when they start out. If they say otherwise, they are lying.

If I want to upload videos for youtube, sharpening is necessary, since the low bitrate will reduce the sharpness and remove many side-effects of sharpening. If I want to recomress the original lossless video into lossless HEVC, the compression will remove a lot of side effects. (Not all side effects, but most of them)

Sharpening actually impairs compression efficiency. If you sharpen a PNG, the sharpened output takes more bitrate if you use the same PNG compression settings. Try it. At low bitrates such as YT uses, a heavily sharpened version usually causes more problems. Think of it as wasting bitrate on those artifacts. IF you had unlimited bitrate - sure you could reproduce those artifacts exactly how you want to - but since you don't - the entire image, and especially edges, degrade more than if you didn't sharpen. One of the YT "optimization" strategies is temporal NR and blurring. Yes - blurring, not sharpening. It makes it easier to compress. Of course these are generalized comments - but these are fundamental basics for low bitrate lossy video compression, not just YT



Look at the railing , there is a ring of "white" around the black railing. Or black jeans have glowing ring of white around it. Does that happen in real life ? Where can I get black jeans that glow white - very cool effect for the kids on halloween, but unrealistic. By sharpening so much, you're enhancing the noise on some sections such as the wall. If it was denoised properly, then oversharpening wouldn't be as bad. I'm not saying don't sharpen... just do not sharpen so much.


[Attachment 61198 - Click to enlarge]

"The original building has no noise...." - then what is this ? Walls can have noise but buildings can't ? Is that what it looked like in real life ? Was a cloud of grain following the guy walking down the stairs? It's like "pigpen" in Charlie Brown .


[Attachment 61199 - Click to enlarge]



The black line artifacts around edge of frame borders and the artifacts selur pointed out are from temporal stabilization edge artifacts, and/or too high temporal denoising settings. It's a bit distracting, but in real usage, you'd probably keep some stabilization and zoom in a bit instead of reverting the stabilization. It's introducing more problems with artifacts, and probably not worth it in that section. You don't need to stabilize the input, neat video does good job "out of the box". You can read the wall "visitor center" by default just sampling that area. Pre stabilizing is probably not a trade off most people would make if it generated ghosting and edge artifacts like that


The oversharpening wouldn't be as bad if it was cleaned properly. Now there are "splotchy" noise patterns - it's not consistent. The ghosting artifacts, splotchy noise do not occur in real life, do they ?

[Attachment 61200 - Click to enlarge]


The avs version of SMDegrain does not exhibit as bad ghosting frame border issues, when the stabilization method uses the "poster frame" lock method I mentioned earlier. So even though frames +/- x might have "black borders", smdegrain is smart enough to reduce the issues even when using large radii (even 20 or 30). The havs vpy version is limited to 6, the G41Fun version is limited to 24 and is "cleaner" than avs version at equivalent TR but exhibits more ghosting. I'll post more on this later, but smdegrain+pre/post is a viable method of clean border fill that you can use with stacking (there are other more complicated ones , that involve multiple tracking layers and 3D camera projection)

If you use SMDegrain , default settings and don't adjust, or appropriate settings or filtering, you're not going to get good results. But the same thing can be said about Neat Video or any filter



Compare to neat video using basically default settings, no sharpening; and use the original as reference.

original

[Attachment 61202 - Click to enlarge]

truthler

[Attachment 61201 - Click to enlarge]

neat video "default", no sharpening

[Attachment 61203 - Click to enlarge]



Look at the ghosting. The black pole in front of the lady is partially transparent. There are overharpened artifacts, noise enhancement on the road, splotchy grain. The uneven-ness of denoising makes it look bad on some frames - the "splotchy" patterns of partially clean, some large noise, some small, and then sharpened. In real life you don't see patches of smooth and patches of noise on the same road at the same time. Does that look "realistic" to you when you were there? Uniform noise/grain, or smooth but less detail would be more preferable to splotchy pattern . The background wall has sharpened noise artifacts - they are not real details - and you're not supposed to see clear details on objects farther away in real life, unless you see even more clear details in those objects which are closer in proximity, such as the road more clearly - and that contributes to the unnatural look - it looks "wrong". There should be depth aware sharpening, BG sharpen less than the FG, because that's how vision works in real life. Also notice the white halo around the chain link - clear signs of oversharpening. Again, I'm not saying don't sharpen, I'm saying just don't sharpen so much, because it looks terrible, is enhancing noise and causing problems


There are other various temporal artifacts, missing objects eg. such as poles, blurred feet on people, guy walking in park talking on phone is missing leg - but any denoiser using strong temporal settings will have those sorts of artifacts, unless you use some sort of protection like motion masks, or weaker temporal settings. As good as Neat Video is (when used properly) - the temporal denoising tends to be not as good as some other temporal denoisers (or you have to be very careful about the strength when using NV) . NV has improved a LOT since v2, but temporal denoising is still a relative weakness (which you can improve for any denoiser by pre- stabilizing). On the other hand, because NV can completely overdenoise (noise and detail) easily, you can add back noise and detail in layers - so if you use NV in more advanced manner it can still produce very good results.



more to come later...

It would be more interesting to post a response video with SM-Degrain (I guarantee that it will be worse), and you can write chriticism after that.

[rrats]

Can you please stop quoting entire posts, it's hurting my brain

[Truthler]

You are probably butthurt 'cause you spent your money on overpriced camcorder instead of a good mirrorless.

Mirrorless is not for video, it is for still pictures. I would wonder If you can sgot better qualiry video than me.

Buy your mirrorless camera, and try to shot in the evening...hahahaha

That is also wrong.

https://www.creativelive.com/blog/using-best-cameras-for-night-photography-settings/

These are cameras they RECOMMEND for evening/night shooting. Notice that down on the list are a number of mirrorless cams as well as DSLRs.

And why are you even saying something like "mirrorless is not for video"? ALL VIDEO CAMERAS USE MIRRORLESS style of tech. Maybe you aren't familiar with how they are built?

Also, and more importantly, one area that still cams excel that video cams cannot do is long exposure shots in low light. Like 2sec-8sec, or longer astrophotography, or long range night shots of skylines and traffic. Video and cinema cams are limited in their exposure times by the shutter speed - they can go shorter (using "shutter angle" adjustments), but they cannot go longer than 1/24th or 1/24th or 1/30th of a second.

Scott

But I did not speak about astro-photgraphy, but about average quality (often no name without old brand) consumer still cameras what these guys want to buy. Their consumer grade still cameras were unable to compete even with an Iphone 12 in video quality ))

[Truthler]

A simple Iphone can beat a 6000$ flagship DSLR easily.

The test:
https://www.youtube.com/watch?v=L9TeYua8bpA

On the test, the DSLR makes better photos, but the DSLR video is worse than Iphone's video.