madVR - high quality video renderer (GPU assisted)

[huhn]

on the mario picture. yes it doesn't look good.

[Georgel]

In my video examples, the higher the algorithm is set, the better it looks.

I would probably not use medium either - which is puzzling since on the Mario image I wasn't bothered at all by it.

[ryrynz]

So why remove "low"? Is it really *that* terrible? If I compare it to NNEDI-16, IMHO NGU pixart "low" competes just fine. Or try comparing "low" to the regular (non-pixart) NGU variant!

Actually when I first tested this I don't think that NGU pix low actually activated and I was looking at Lanczos. I noticed this same thing occurring when I played with it on my GTX 960.
NGU pix low isn't terrible, but NNEDI3 16 handles thick low res lines more elegantly.

So I had a look vs NNEDI3 on the HTPC, the high preset is the closest in terms of visual appearance to NNEDI3 with how it enhances lines. Very high is more aggressive with parts of the image which is thinks are not lines and sometimes removes them (yes they're blurry but NNEDI3 does not do this) So in this way it seems to operate very much like NGU. Basically NGU just comes off as a tad aggressive with how it operates on lines to me, even with this version of it.

When I compare the original image upscaled with Lanczos to NNEDI3 64 neurons + my Avisynth profile it looks similar in appearance just more refined, lines are similar in shape and appearance. With NGU pix in places there's shrinking of lines, some areas it works nicely with line edges curving them naturally and in other places it just looks a bit off. NNEDI3 is more natural with lines and in places it's sharper too which just gives it a more drawn feel to it where as NGU just looks rather bland almost artifical in appearance by comparison, this is more important to me than anything else.

NGU pix might be a better fit for me with upscaling 720p content due to the extra resolution (will have to take a look) but for DVD level stuff I'm sticking with NNEDI3. Basically you'd have to improve on what I almost consider to be basically perfect for my use case. I guess I'll have to upload pictures to show you, my guess it will be hard to make me replace NNEDI3, it's a great alternative only IMO because obviously in certain situations it shines, just not for me where I need it most.

[Arm3nian]

1) Do you think NNEDI3 looks better? Or the new NGU pixart variant?
3) Can I replace NNEDI3 with NGU pixart in the next official build (pretty please)?
4) Can I replace super-xbr (for luma doubling, only) with NGU pixart in the next official build?

1) Even on low, I think NGU pixart does a better job of making a crap source look better, compared to max nnedi3 settings. It looks the cleanest, which nnedi3 was the king of.

3) Yes because NGU pixart looks better for low quality sources and NGU already looks better for high quality sources.

4) I've liked super-xbr because it looks good and requires little power to run. But NGU pixart on low runs just as fast and looks better than super-xbr for low quality sources. Regular NGU on low looks pretty much the same as super-xbr for high quality sources, and runs at the same speed. So I don't see the need for super-xbr anymore.

[mogli]

4) i've liked super-xbr because it looks good and requires little power to run. But ngu pixart on low runs just as fast and looks better than super-xbr for low quality sources. Regular ngu on low looks pretty much the same as super-xbr for high quality sources, and runs at the same speed. So i don't see the need for super-xbr anymore.This could (almost) be used as an argument to stay with super-xbr. I guess you talk about luma only so what chroma are you using?

[Arm3nian]

This could (almost) be used as an argument to stay with super-xbr.
How so?

I guess you talk about luma only so what chroma are you using?
I don't see any differences between the chroma upscaling options, from different passes to different algorithms, on low resolution low quality video, so I used the same one.

For chroma doubling, you are restricted to choosing what you want based on the luma option, so I left that on default as well.

The changes are obvious in luma doubling. In the mario picture, pixart low doesn't look that great, but neither does nnedi3 compared to pixart very high. In my sources, with actual people, pixart low performs better than in the mario. I would choose it over any configuration with nnedi3 or superxbr. Again though, with my sources, veryhigh pixart looks very good, just like in the mario.

[madshi]

yeah, but since I hardly do any scaling atm, I am more interested just in changes of chroma upsampling and wondered if there are any differences between the regular algo and pixelart as well when it comes just to chroma upsampling
It's a different algorithm, so of course there are differences.

I'd say the picture quality with NGU pixart Very High looks a little better, or equal, depending on the source material. Definitely not worse with any material. :) Shaving off a few milliseconds is always a nice bonus!

There is slight loss of sharpness to my eyes with NGU pixart set to Medium. What really surprised me is there isn't a huge drop in quality between Very High to Medium. It also saves an additional 5 to 7 milliseconds of render time at Medium which is impressive. If I had less capable hardware, I'd have no issue with using the Medium setting. I'll stick with Very High for now with SD, because I can. :p
Sounds good!

Holy! I like "NGU pixart" with games!
Thank you - coming from you this means high praise! :) I'll reply to your question via PM.

I only use NNEDI3 at 16 neurons - NGU very high is slower than NNEDI at 16 neurons in same conditions.

Example :

NNEDI3 at 16 neurons is slower than NGU at high

NGU pixart at very high takes double the time to process compared to NNEDI3 at 16 neurons.

More precise examples:

NGU high ~5ms

NNEDI3 at 16 neurons ~8ms

NGU very high ~16ms

If everything else was kept the same.
Yes, but how does quality compare? FWIW, NGU pixart "high" currently doesn't exist. If you activate it, it uses NGU pixart "medium" for luma, but it uses a higher quality chroma setting compared to NGU pixart "medium". What would be interesting to me is if you prefer NNEDI3-16 or NGU pixart medium (in terms of image quality), when testing with real world sources?

Actually when I first tested this I don't think that NGU pix low actually activated and I was looking at Lanczos. I noticed this same thing occurring when I played with it on my GTX 960.
NGU pix low isn't terrible, but NNEDI3 16 handles thick low res lines more elegantly.
I'm glad to hear that NGU pix low isn't terrible. It certainly wasn't terrible in my own tests, so I was a bit puzzled when you said it was. I'm not sure why it didn't activate in your case. If you can reproduce that somehow reliably, maybe you can find out the circumstances in which it doesn't activate? Does the OSD then still claim it's active?

So I had a look vs NNEDI3 on the HTPC, the high preset is the closest in terms of visual appearance to NNEDI3 with how it enhances lines. Very high is more aggressive with parts of the image which is thinks are not lines and sometimes removes them (yes they're blurry but NNEDI3 does not do this) So in this way it seems to operate very much like NGU. Basically NGU just comes off as a tad aggressive to me even with this version of it.

When I compare the original image upscaled with Lanczos to NNEDI3 64 neurons + my Avisynth profile it looks similar in appearance just more refined, lines are similar in shape and appearance only slimmer. With NGU pix in places there's shrinking of lines, some areas it works nicely with line edges curving them naturally and in other places it just looks a bit off. NNEDI3 is more natural with lines and that's more important to me than anything else. NGU pix might be a better option for upscaling 720p content due to the extra detail there (will have to take a look) but for DVD level stuff I'm sticking with NNEDI3. Basically you'd have to improve on what I almost consider to be basically perfect for my use case. I guess I'll have to upload pictures to show you, my guess it will be hard to make me replace NNEDI3, it's a great alternative only IMO because obviously in certain situations it shines, just not for me where I need it most.
Interesting. Yes, screenshots would be helpful (please always include the unscaled source image, too).

FWIW, it seems most other users actually prefer NGU pixart over NNEDI3, or at least don't prefer NNEDI3, so I might end up removing NNEDI3 soon, unless I find evidence that NGU pixart is really clearly worse in some situations.

1) Even on low, I think NGU pixart does a better job of making a crap source look better, compared to max nnedi3 settings. It looks the cleanest, which nnedi3 was the king of.

3) Yes because NGU pixart looks better for low quality sources and NGU already looks better for high quality sources.

4) I've liked super-xbr because it looks good and requires little power to run. But NGU pixart on low runs just as fast and looks better than super-xbr for low quality sources. Regular NGU on low looks pretty much the same as super-xbr for high quality sources, and runs at the same speed. So I don't see the need for super-xbr anymore.

[...]

In the mario picture, pixart low doesn't look that great, but neither does nnedi3 compared to pixart very high. In my sources, with actual people, pixart low performs better than in the mario. I would choose it over any configuration with nnedi3 or superxbr. Again though, with my sources, veryhigh pixart looks very good, just like in the mario.
I'm happy to hear that - thanks! :)

[mogli]

How so?Well, you say they are pretty much the same except that NGU smooth looks better for low quality images. But then again speed does depend on chroma so super-xbr is still faster in some cases. Since I would prefer one algorithm that handles more cases over several different ones, overall I still prefer super-xbr for now.

[ryrynz]

FWIW, it seems most other users actually prefer NGU pixart over NNEDI3, or at least don't prefer NNEDI3, so I might end up removing NNEDI3 soon, unless I find evidence that NGU pixart is really clearly worse in some situations.


They're probably not looking as closely as I am at the same sort of content, I'm picky about how natural upscaling and processing must look and NGU just appears a tad too artificial compared to NNEDI3 with lines.
I'll be able to show you this, but you may not deem the differences significant enough which has me a little concerned. Adding blue edges and a touch of sharpness seems to bring things closer in line to NNEDI3 but there's still that slightly more aggressive line creation/deletion going on which I don't like. Why does NGU pix very high look a good step more unnatural than high? Scaling up NNEDI3 looks the same regardless, the lines just get better though beyond 64 neurons almost seems pointless in many situations but it least it looks the same, with NGU you get one step forward and one step sideways, they're all quite different in appearance.

I guess I'm going to have to take a lot of screenshots..

Any way.. I can duplicate an issue with scaling although this applies to NNEDI3 on screen not NGU. If I start a video using NGU pix (let madVR decide everything) and then switch to NNEDI3 (any neurons, let madVR decide everything) NNEDI3 doesn't activate and I get whatever madVR chose (in this case bicubic 60 all the way). If I change neuron count then NNEDI3 applies properly. This is using D3D9 on the Intel HD4000.

[madshi]

Why does NGU pix very high look a good step more unnatural than high?
There is no real "high" atm (see release notes). "high" is the same as "medium", but with better chroma doubling.

Unnatural or not, I'll wait for your screenshots.

Any way.. I can duplicate an issue with scaling although this applies to NNEDI3 on screen not NGU. If I start a video using NGU pix (let madVR decide everything) and then switch to NNEDI3 (any neurons, let madVR decide everything) NNEDI3 doesn't activate and I get whatever madVR chose (in this case bicubic 60 all the way). If I change neuron count then NNEDI3 applies properly. This is using D3D9 on the Intel HD4000.
OpenCL initialization takes a while. So when activating NNEDI3 for the first time, madVR starts with a simple linear interpolation (so the image appears without delay), while OpenCL initialization is done in a background thread. Once OpenCL is ready to go, the image should automatically be re-rendered with NNEDI3.

[Arm3nian]

Well, you say they are pretty much the same except that NGU smooth looks better for low quality images. But then again speed does depend on chroma so super-xbr is still faster in some cases. Since I would prefer one algorithm that handles more cases over several different ones, overall I still prefer super-xbr for now.
I said NGU low looks and performs similarly to super-xbr. You can still choose higher settings for NGU if your hardware allows, which can produce a better image.

Same for pixart, to a greater extent probably, higher NGU levels look quite a bit better than low/super-xbr.

As for having one algorithm that works for all cases, you still have to make profiles for different content if you want optimal settings, so it doesn't really matter.

[ryrynz]

There is no real "high" atm (see release notes). "high" is the same as "medium", but with better chroma doubling.


Yup, am aware.


OpenCL initialization takes a while..

This is what I'm seeing, thanks.

[YxP]

Pixart doesn't seem to do 4X upscaling yet so I can only assume quality will be even better later. Tested it quickly watching older Simpsons and Futurama (meaning: closer to that hand-drawn style) and it works very, very well. Definitely will use it instead of NNEDI myself.

[madshi]

Pixart doesn't seem to do 4X upscaling yet so I can only assume quality will be even better later. Tested it quickly watching older Simpsons and Futurama (meaning: closer to that hand-drawn style) and it works very, very well. Definitely will use it instead of NNEDI myself.
Sounds good! I'm not sure yet if I can make 4x pixart, we'll see. So you seem to prefer pixart quality over NNEDI3? Could you maybe post a comparison screenshot or two?

[YxP]

Sounds good! I'm not sure yet if I can make 4x pixart, we'll see. So you seem to prefer pixart quality over NNEDI3? Could you maybe post a comparison screenshot or two?

I guess I could, but I'm not sure if it's worth it since I can only use NNEDI16 before my GPU fan goes nuts. I can use NGU or pixart medium with almost completely silent fan, so it's more of a performance thing for me, although I DO like NGU more also quality-wise. I'll check later if I can find some good examples anyways.

[Damien147]

I am happy that I chose rx 470 over rx 460:)
Still can't use NGU Very High but by seeing some people mentioning ''unnatural'' look I am calm with NGU med.
@Madshi confess it,you get paid by AMD and NVidiots....
I see in the future GPUs with ''MadVR optimized'' sticker in the box.:p

[pose]

Still can't use NGU Very High
Why? What resolution youre scaling to?

[huhn]

because NGU is very slow on polaris GPU for still unknown reasons.

[cork_OS]

With NGU pix in places there's shrinking of lines, some areas it works nicely with line edges curving them naturally and in other places it just looks a bit off. NNEDI3 is more natural with lines and in places it's sharper too which just gives it a more drawn feel to it where as NGU just looks rather bland almost artifical in appearance by comparison, this is more important to me than anything else.
Seems I see what you mean (EDIT: NNEDI is sharper and lines are darker or more expressed/emphasized): http://screenshotcomparison.com/comparison/199707
Source sample (http://www.mediafire.com/file/fei452045t546p4/VTS_01_1.mpg) (frame 80).

[pose]

because NGU is very slow on polaris GPU for still unknown reasons.
I have RX 470 and NGU very high is as fast as NNEDI3 128 in the same scenario. Probably not as fast as it should be but good enough for me.

[huhn]

if want to see some light at the end of the tunnel.
the NGU very high x4 scaler performed very well on my RX 480.

i still don't know how people can accept such a huge regression compare to the R9 390 but not my problem.

[pose]

i still don't know how people can accept such a huge regression compare to the R9 390 but not my problem.
NGU on R9 390 is much faster than on RX 480? Ive heard multiple times since the NGU implementation that Polaris is bad at it but yet to see any decent comparison with older cards of the same tier.

[Damien147]

Why? What resolution youre scaling to?

HD to 4k(4096x2160)

I have RX 470 and NGU very high is as fast as NNEDI3 128 in the same scenario. Probably not as fast as it should be but good enough for me.

This means that you can run NGU very high ?
Same scenario as mine you mean?
The best I can do is NGU high for HD to 4k but for some occasions I get dropped frames with high(only a few).

I am like HD to 4K--->NGU med(chroma and luma)-->Jinc AR.You?
Tried NNEDI3 128(chroma and luma) to check your saying and I get dropped frames.

[huhn]

there are numbers in this thread.
it was something like 3-5 times slower.

my 960 out performances the rx 480 of cause only in NGU.
i have a r9 270 amd and only the 960 at the time so no R9 390 or something like that but the R9 270 is fine in term of NGU performance.

[pose]

HD to 4k(4096x2160)
This means that you can run NGU very high ?
Same scenario as mine you mean?

Ooh, i see. No, im a pleb and my monitor is a 1080p. I run NGU med chroma/ NGU very high luma -> Jinc AR upscaling / SSIM 1D100 AR LL downscaling. Good to know that you at least can do NGU high, it should look great :)
Tried NNEDI3 128(chroma and luma) to check your saying and I get dropped frames.

Dont use such excessive amount of neurons for chroma when youre upscaling something. Luma is the most important.

[kitame]

"heavy" image enhancements responds quite well with NGU, it wasn't much useful with Lanczos or NNEDI3 before.

current settings:
sharpen edges = 1.0
LumaSharpen = 0.75
AdaptiveSharpen = 1.0
anti-bloating filter = 75%
anti-ringing filter = yes

chroma upscaling = NGU (low)
image upscaling = NGU (high+high)


on a side note, whats "algorithm to use after doubling" for? if i don't check the 200% supersampling does it mean it remains unused?
also, theres a second "Chroma NGU(low)" on the OSD, is this normal?

[Damien147]

Ooh, i see. No, im a pleb and my monitor is a 1080p. I run NGU med chroma/ NGU very high luma -> Jinc AR upscaling / SSIM 1D100 AR LL downscaling. Good to know that you at least can do NGU high, it should look great :)

Looks great indeed.Good to know that my GPU works as intended because you gave me second thoughts.There's gonna be a confusion with NGU pixart and standard NGU for what to choose in the future I guess but we'll see how it goes.


Dont use such excessive amount of neurons for chroma when youre upscaling something. Luma is the most important.

I just used NNEDI3 to compare with what you said.Thanks for the tip though.

there are numbers in this thread.
it was something like 3-5 times slower.

my 960 out performances the rx 480 of cause only in NGU.
i have a r9 270 amd and only the 960 at the time so no R9 390 or something like that but the R9 270 is fine in term of NGU performance.

That gives as hope that things will hopefully improve in the future.:)

[Georgel]

@madshi

I continued testing and results are mixed.

On some or most sources, NGU is better than NNEDI.

To be honest, the new NGU looks really good and this is the sweet spot.

NGU at those settings will distort the image, but less than the other algorithms. The sharpening has one of the least ringing and added aliasing / noise so this makes most sense to be kept. I still wouldn't dismiss NNEDI entirely, someone might need it, and NGU will still add some of that GPU high pitched noise. If there is any way to make it even faster, that would be helpful, after some point it's not efficient enough and will introduce dropped frames.

http://i.imgur.com/C7NRhiH.png

http://i.imgur.com/oFGJ5tz.png

[Warner306]

"heavy" image enhancements responds quite well with NGU, it wasn't much useful with Lanczos or NNEDI3 before.

current settings:
sharpen edges = 1.0
LumaSharpen = 0.75
AdaptiveSharpen = 1.0
anti-bloating filter = 75%
anti-ringing filter = yes

chroma upscaling = NGU (low)
image upscaling = NGU (high+high)


on a side note, whats "algorithm to use after doubling" for? if i don't check the 200% supersampling does it mean it remains unused?
also, theres a second "Chroma NGU(low)" on the OSD, is this normal?

Scaling from 720p -> 1080p, for example, overscales to 1440p, so downscaling is required after doubling -- 1440p -> 1080p.

[kitame]

Scaling from 720p -> 1080p, for example, overscales to 1440p, so downscaling is required after doubling -- 1440p -> 1080p.

that makes sense, however how does one trigger it?

edit: i think i got it, it automatically triggers depending on the source's scaling requirement.
e.g. if source requires more than double it triggers the "upscaling" input, where as if the source requires less than double it triggers the "downscaling" input.

then does this mean the image upscaling does 2x scaling by default?
if this is the case then shouldn't "image upscaling" be renamed into "image doubling" to make an accurate description?

edit2: i just noticed these doubling requirements only applies to Super-XBR, NNEDI3 and NGU.

[davide1983]

Hi, why madvr scaling algorithms don't work whit 4k resolution ?

Thanks

[Damien147]

Try restoring default settings,for me they work.

[MS-DOS]

madshi, Is D3D11 version of NGU still planned?

[jkauff]

Not well enough compared to what? The only important question for me right now is if NGU pixart can replace NNEDI3 or not, because that's what it was designed for.
After testing with several SD movies (none animated), NNEDI3 still wins. NGU pixart High processing does not look as "natural" for live-action as NNEDI3 128/32. It resembles the look I get when I throw SuperRes into the mix--too processed-looking. This is especially true for black-and-white films.

I have no doubt it works fine for anime, although I have no anime to test with. I hope other testers will be trying it out with SD live-action movies to confirm or refute what I'm seeing.

BTW, I'm using a GTX1060 with the latest drivers.

EDIT: NGU still looks great for higher resolution sources, as it has since you introduced it. However, I watch a lot of SD.

[ryrynz]

I have no doubt it works fine for anime, although I have no anime to test with. I hope other testers will be trying it out with SD live-action movies to confirm or refute what I'm seeing.


It's fine.. Not as good as NNEDI3 64 neurons, but even with HD anime I still prefer the more natural look to its lines over NGU, it just has a flat and processed look to it. Glad to see other people picking up on it too.

Try NGU with edge blurring on 1 or 2 and ~0.5 edge sharpening and IMO it looks a bit better.. But then your performance benefit goes out the window..

[leeperry]

I don't watch anime so I'm not sure if my testing would help but why do you wanna ditch sxbr and/or NNEDI3? So far I like the latter for very low-res SD and the former in combination with SR for stuff that's too noisy to get the NGU treatment, I don't think there will ever be a one-size-fits-all doubler that will work in any given situation. I rest my case that in combination with FRC I love what SR does to motion-blur and NGU turned this combo into bloatland last time I checked =/

[khanmein]

@Georgel what kinda graphic card u using now? cheers.

[khanmein]

because NGU is very slow on polaris GPU for still unknown reasons.

i personally think related to ROPs value cos RX480 (32) compare with R9 290/390/x (64)

[madshi]

Seems I see what you mean (EDIT: NNEDI is sharper and lines are darker or more expressed/emphasized): http://screenshotcomparison.com/comparison/199707
Source sample (http://www.mediafire.com/file/fei452045t546p4/VTS_01_1.mpg) (frame 80).
FWIW, your zoom factor is very weird. It's not 2x and not 4x, but something in between. And I can say that with zoom factor, the current madVR builds happens to apply NGU pixart twice, but NNEDI3 only once, so it's somewhat of an apples to oranges comparison. If you pick a slightly smaller zoom factor, NGU pixart is applies only once, too, which makes the images nearer to each other. There still are differences, though.

"heavy" image enhancements responds quite well with NGU, it wasn't much useful with Lanczos or NNEDI3 before.
Are you talking about NGU (one of the official madVR builds)? Or the special "NGU pixart" test build? If it's the pixart test build, please make sure you say "NGU pixart", otherwise there might be confusion.

I continued testing and results are mixed.

On some or most sources, NGU is better than NNEDI.
NGU or "NGU pixart"? Those are 2 different things!!

madshi, Is D3D11 version of NGU still planned?
Maybe.

After testing with several SD movies (none animated), NNEDI3 still wins. NGU pixart High processing does not look as "natural" for live-action as NNEDI3 128/32. It resembles the look I get when I throw SuperRes into the mix--too processed-looking. This is especially true for black-and-white films.

I have no doubt it works fine for anime, although I have no anime to test with. I hope other testers will be trying it out with SD live-action movies to confirm or refute what I'm seeing.

EDIT: NGU still looks great for higher resolution sources, as it has since you introduced it. However, I watch a lot of SD.
Screenshots (with unscaled source frames), please?

I don't watch anime so I'm not sure if my testing would help but why do you wanna ditch sxbr and/or NNEDI3?
Because NNEDI3 is very slow (and relies on unstable OpenCL). If I can make an algorithm which is both faster *and* better looking than NNEDI3 with the same sources NNEDI3 shines with, then that should be a benefit for everybody.

[BluesFanUK]

NGU is very good, but to me it's a pointless function atm purely because of the noise it generates on the GPU. It's EXTREMELY annoying, sounds like the PC is frying.

Headphones on when I use it.

[khanmein]

NGU is very good, but to me it's a worthless function purely because of the noise it generates on the GPU. It's EXTREMELY annoying, sounds like the PC is frying.

this proven that your GPU is reaching the limit so i suggest upgrade it.

previous my GTX 970 have coil-whine but now i upgraded to GTX 1070 no more coil whine anymore.

[kitame]

Are you talking about NGU (one of the official madVR builds)? Or the special "NGU pixart" test build? If it's the pixart test build, please make sure you say "NGU pixart", otherwise there might be confusion.

i'm using the official 0.91.5 build, so i'm pretty sure its not the pixart build.

[leeperry]

Because NNEDI3 is very slow (and relies on unstable OpenCL). If I can make an algorithm which is both faster *and* better looking than NNEDI3 with the same sources NNEDI3 shines with, then that should be a benefit for everybody.
I hear you but options are good, NNEDI3 really shines at detecting edges in say 360*288 movies. I'm sure some ppl will always prefer NNEDI3 to anything else depending on their needs, taste and display. Anyway, please don't ditch sxbr coz I like it especially on SD that's too noisy for NGU.

:thanks:

[Georgel]

@Georgel what kinda graphic card u using now? cheers.

GTX 1080



NGU or "NGU pixart"? Those are 2 different things!!




I copied latest madVR then copied the test files for NGU Pixart over the install like you said. Those settings are for that situation...

http://i.imgur.com/ygPfORE.jpg

[Thunderbolt8]

360*288 moviesbut how many people are watching movies in this resolution, let alone care about good scaling then?

[leeperry]

^ If you can't count on mVR no more for extreme hopeless upscaling then what has this world come to eh :D

[CruNcher]

but how many people are watching movies in this resolution, let alone care about good scaling then?


SD/D-1
HD
FHD

are the most typical inputs these days only handling sub that is crazy somehow to even take into account anymore, no TV advertises with sub SD improvements and few only with good SD Handling ;)

Also if you think which codecs and/or analog sources have been used for those ;)

But still DVD (Mpeg-2) is a viable target they still exist and want to be upscaled to up to 4K now ;)

[Shiandow]

but how many people are watching movies in this resolution, let alone care about good scaling then?

Well, watching 360x288 on a 1080p screen is the same as watching 1440x1152 (i.e. >1080p) on an 8k screen, so it will become relevant eventually, probably. That said I don't think you can expect an image to look well when scaled more than 3x in each direction, eventually you just run out of detail. I am however somewhat optimistic that for smaller scaling factors you can make it hard to distinguish between a low quality high resolution image and a high quality low resolution image.

[CruNcher]

Most probably you first have to deartifacting before even thinking about the upscaling, every TV Hardware (MediaTek Based) these days has at least a simple MPEG improvement mode mostly even straight from the decoder part itself and Mali Txx onboard ;)

And much faster as you can say A all the cheap Smart TVs will be Vulkan Render based ;)

[cork_OS]

FWIW, your zoom factor is very weird. It's not 2x and not 4x, but something in between. And I can say that with zoom factor, the current madVR builds happens to apply NGU pixart twice, but NNEDI3 only once, so it's somewhat of an apples to oranges comparison. If you pick a slightly smaller zoom factor, NGU pixart is applies only once, too, which makes the images nearer to each other. There still are differences, though.
My bad, I didn't checked that. I have 1680x1050 display and just used "touch window from outside" in player. Additionally, I didnt'n unchecked "add grain".
I redid the comparison with zoom factor around 4x and found that NGU pix output is much closer to NNEDI3 with some presharpen (like LumaSharpen 0.3) added to NGU.
NNEDI3 64+32 vs. NGU pix med+med (http://screenshotcomparison.com/comparison/199827)
NNEDI3 64+32 vs. NGU pix med+med (LumaSharpen 0.3) (http://screenshotcomparison.com/comparison/199828)
So I carefully assume people may prefer NNEDI3 due to its sharpness, not line thinning differences.