MCTemporalDenoise [v1.4.20 - Update 2010/07/02]

[newface]

While i am adding to script MCTemporalDenoise(settings="high") getting error ther is no function named "Msuper" plz rectifie this error image attached below

[um3k]

Sounds like you need MVTools 2: http://avisynth.org.ru/mvtools/mvtools2.html

[newface]

Originally Posted by um3k

Sounds like you need MVTools 2: http://avisynth.org.ru/mvtools/mvtools2.html

Please give me more detiles

[MatLz]

Originally Posted by newface

Please give me more detiles

MSuper is one (surely the most important) of the functions in the MVTools2.dll plugin

[jordisound]

Originally Posted by Nephilis

Try MT("""MCTemporalDenoise(settings="high")""",threads=4) instead of SetMTMode()..

well, it works. With this sintaxis the speed increase, the 4 cores are working fine.

I am testing gpu=true vs. gpu=false (fft3dgpu vs fft3dfilter). fft3 gpu is 7% faster. analyzing the final video, the compressibility is 5% better with fft3dfilter. it says that the default values for fft3d and fft3gpu are different? or fft3d is more efficient than fft3dgpu?

[Poutnik]

Originally Posted by jordisound

well, it works. With this sintaxis the speed increase, the 4 cores are working fine.........

Interesting, I went the opposite direction MT() -> SetMTMode, realizing the spatial dividing by MT() was causing for me artefacts.

MC is said - seems obvious - to have gain from evaluation the whole picture.

[Joachim Buambeki]

I want to use MCTemporalDenoise to denoise my timelapse footage, but it always give me an error:

There is no function named "Sangnom"

The script is just:

MCTemporalDenoise(settings="medium")

for a 720p clip. As far as I understand there should not even be a call for Sangnom. I don't have the Sangnom.dll, since I am using the 64bit package from JoshyD.

Another question I have is:
What settings would you recommend me to denoise 4k footage, based on the documentation I figured larger block sizes should do no harm at 4k:

MCTemporalDenoise(settings="medium", protect=true, stabilize=true, enhance=true, bwbh=32, owoh=16, blksize=32, overlap=16)

Is this likely to work? I am afraid such a complex script will will not run very stable at 4k.

Best Regards
David

[LoRd_MuldeR]

http://manao4.free.fr/SangNom.zip

[Joachim Buambeki]

Thanks, but I need a 64bit Sangnom (though I wonder why Sangnom is needed for the above script.
From what I have read other people use MCTemporalDenoise with succes on with 64bit Avisynth.

Best Regards
David

[LoRd_MuldeR]

Originally Posted by Joachim Buambeki

Thanks, but I need a 64bit Sangnom (though I wonder why Sangnom is needed for the above script.

Well, if MCTemporalDenoise makes use of the Sangnom plugin, then you'll need that plugin for MCTemporalDenoise to work. It's that easy

However I don't know if a 64-Bit build of Sangnom exists...

[Joachim Buambeki]

Other people in the thread dedicated to Avisynth 64bit seem to use it without problems (last posts that mentions it in the thread). Since I use no anime preset or the anti aliasing function of MCTemporalDenoise I do not understand why Sangnom is even called. That is why I am confused...
Also I checked the 64bit builds of different filters by JoshyD and squid and have not found a 64bit version of Sangnom.

Best Regards
David

[Didée]

Try explictely setting MCTemporalDenoise(setings="medium",useEEDI2=true).

Some filtering with SangNom is initiated within the function, but not used later on. Even though SangNom is not actually used with the given call, the parser has the need to evaluate the SangNom() filter.

Taken strictly, this is sign of unkosher scripting ... which, in same or similar style, I might have done "a thousand" times myself.


Edit - Clarification by example:

Assume this little quasi-script:

do_filter   = false
use_filterX = false

vid = something

filter = use_filterX ? vid.filterX() 
\                    : vid.filterY()

out = do_filter ? filter : vid

return( out )

With given settings, neither filterX nor filterY are actually performed. The output result is just "vid".

However, by the time when the parser is parsing the "filter = ... ? .. : .." line, it needs to know "filterY", because it needs to fully evaluate the code line. The path given by "use_filterX=false" requires that the parser evaluates "filter = vid.filterY()". Albeit the action never is performed, it is still required that "filterY" is *known* at least.

[Joachim Buambeki]

Originally Posted by Didée

Try explictely setting MCTemporalDenoise(setings="medium",useEEDI2=true).

Thanks Didée, that seemed to be one of the culprits.
But now I get the error:

EEDI2: maxd must be < 30!

EDIT: I searched for "maxd" and it seems to be named "maxdiff" in the script and the the default is 48 for medium. Changing it to something below 30 does not seem to help, the script seems to ignore my user specified setting. When using

MCTemporalDenoise(setings="low",useEEDI2=true)

it works finally, even if I illegally set maxdiff to 36 with settings="low". There seems to be something borked in the presets.
I will look deeper into it tomorrow.

Regards
David

[Gavino]

Originally Posted by Didée

Taken strictly, this is sign of unkosher scripting ...

Quite so. As I'm sure you know, that sort of thing is best done inside a conditional expression, especially if it involves a less frequently-used external plugin whose use is optional, eg
s = (param == xxx) ? SangNom(...) : NOP()

EDIT: Your clarification came after I posted this.
Your example (as you know) would then be more cleanly written as
filter = do_filter ?
\ use_filterX ? vid.filterX() : vid.filterY()
\ : NOP()

(However if filterX and filterY were both built-in functions, I wouldn't bother doing this unless they were known to have a significant time or memory cost on instantiation.)

[LaTo]

Originally Posted by Gavino

Quite so. As I'm sure you know, that sort of thing is best done inside a conditional expression, especially if it involves a less frequently-used external plugin whose use is optional, eg
s = (param == xxx) ? SangNom(...) : NOP()

I will add this into the next version, was removed between 1.3.xx -> 1.4.xx (cleaner code) but... it seems to be a problem

[Gavino]

Originally Posted by LaTo

was removed between 1.3.xx -> 1.4.xx (cleaner code) but... it seems to be a problem

It's ironic that to clean it up in one respect (readability), it had to become less clean in another (loading unused filters).

Of course, the root of the problem is the limitations of Avisynth's conditional syntax. What you really want to be able to write is something like:

if (AA) {
  caa = (useEDDI2 ? .... : ...SangNom(...) ...)
  mA = ...
  smA = ...
}
else {
  ...
  smA = ...
}

[At this point, I will make my usual plug for GScript. ]

Incidentally, on the subject of code cleanup, is there a good reason for using code like

Assert((radius>=1&&radius<=6)         ? true : false,  ...)

instead of just

Assert(radius>=1&&radius<=6, ...)

Perhaps to get round some perceived bug in Assert?

[Joachim Buambeki]

Hi LaTo,

can you confirm (or deny) the problem I have with "maxdiff"?

Your script is awesome by the way!

Best Regards
David

[LaTo]

v1.4.16 :
added switch to avoid loading of unused filters
fixed a bug in anti-aliasing part

[LaTo]

Originally Posted by Gavino

Incidentally, on the subject of code cleanup, is there a good reason for using code like

Assert((radius>=1&&radius<=6)         ? true : false,  ...)

instead of just

Assert(radius>=1&&radius<=6, ...)

Perhaps to get round some perceived bug in Assert?

IIRC I had some problems with Assert in the past, so maybe... but it's a bit old for me, I don't remember

Originally Posted by Joachim Buambeki

can you confirm (or deny) the problem I have with "maxdiff"?

EEDI2 maxd is maxd in MCTD
Sangnom AA is maxd in MCTD
maxdiff is for another stuff

[Great Dragon]

For some reason MCTD doesn't process last and first frames in a sequence for me.

Here is an example:
2 last frames in a sequence:

2 first frames in a next sequence: