Vhs capture advice.

[jagabo]

A constant phase-shift video is easily recognizable and can be fixed with a simple:

DoubleWeave()
SelectOdd()

Or:

SeparateFields()
Trim(1,0)
Weave()

[Sharc]

I prefer the AviSynth approach because you examine the true fields without applying any deinterlacing and then conclusion on processed material. I found (rarely) in the past that it can lead to misinterpretations.

Yeah, but one can examine the true fields in VDub2 as well. I am not defending here anything but rather give the OP a hint because he is not very familiar with Avisynth (yet)

[lollo]

Sure, I know OP preference, just pushing him to upgrade 😉

[lollo]

A constant phase-shift video is easily recognizable and can be fixed with a simple:

DoubleWeave()
SelectOdd()

Or:

SeparateFields()
Trim(1,0)
Weave()

Yes, that's even more intuitive and easier to understand!

[lordsmurf]

. Not everybody is a perfectionist.

You don't have to be a perfectionist to avoid crappy work.

That Youtuber doofus (and there are sadly too many other them) has zero knowledge of video. He may think he's helping others, but he not whatsoever. If this was a car, he'd be telling people how to pour sugar in the tank to "make the gas sweeter, to make the car go faster", or some such BS. He doesn't know what he doesn't know. Newbies are fooled, equally knowing nothing. But the rest of us know he's an idiot.

If you just want "a picture" (with audio), no matter how awful it looks, sure, I guess. But it looks bad even on a tiny cell phone. Everything is royally screwed.

If you don't want to help him, .

Sometimes help is telling people what NOT to do. Not coddling bad questions, based on false premises.

I think a $12 no-name Chinese composite-to-USB converter with bundled software will do as good or better job than his double-adapter-to-OBS approach.

And you would be correct. Both methods are awful, but the cheap Easycap is less shitty.

[Sharc]

That Youtuber doofus (and there are sadly too many other them) has zero knowledge of video. He may think he's helping others, but he not whatsoever.

Don't be so negative. He is generating new forum members asking for help to "fix" their crap

[Leanoric]

I captured this tape the other day and have noticed that there is a bar that runs across the middle of the picture. Its not visible on the first half of the tape. It starts at about the half way point and gets worse towards the end. The clip is taken from the end of the tape where its really bad.

Just wondering if you guys know what the cause of this could be?

[dellsam34]

That's tape drop out, The VCR's DOC is trying the best it can to replace it with good lines but the damage is too big for its capability, If you open the cassette flap and look at the tape you should see a crease on the surface.

[Leanoric]

Ok thanks, thats good to know, the tape is unimportant so Id rather it be bad than something wrong with the vcr.

[Leanoric]

Hi all, Ive got a couple of questions for Hybrid users. The QTGMC deinterlacing looks much better than what vdub was producing, Im struggling to get the denoising right though. At the moment, Im just guessing what boxes to tick and numbers to use, so have ended up with some videos that look very plasticky.




Im unfamiliar with Avisynth scripts so I was wondering if the values Lollo has used in the script below are visible? I can see that he used temporal degrain and FFT3DFilter, but I cant work out what numbers to use in Hybrid.




Also, when processing a video, the order things should go in if Im not mistaken are 1st colour correction, then denoising, then deinterlacing, and then crop and resize?

When I look at the avisynth script in Hybrid, they seem to be in the wrong order, do I need to adjust this somehow?

Thanks

AviSynth script:

# plugins directory
plugins_dir="C:\Users\giuse\Documents\VideoSoft\MPEG\AviSynth\extFilters\"

	# FFmpegSource
loadPlugin(plugins_dir + "ffms2_87bae19\x86\ffms2.dll")

video_org=FFmpegSource2("Clip.avi", atrack=-1)

# cropping 
	crop_left=12	# | rimozione esatta delle bande nere sinistra, sopra, destra e del disturbo sotto	
	crop_top=2	# | 720-(12+16)x576-(2+14)=692x560
	crop_right=16
	crop_bottom=14
video_org_crop=video_org.crop(crop_left,crop_top,-crop_right,-crop_bottom)

video_org_crop_adj=video_org_crop.ColorYUV(off_u=+12, off_v=-6)

	# QTGMC
Import(plugins_dir + "QTGMC.avsi")
	# Zs_RF_Shared
Import(plugins_dir + "Zs_RF_Shared.avsi")
	# Nnedi3
loadPlugin(plugins_dir + "NNEDI3_v0_9_4_55\x86\Release_W7\nnedi3.dll")

	# TemporalDegrain2
Import(plugins_dir + "TemporalDegrain-v2.2.1_modGMa.avsi")
	# RgTools
loadPlugin(plugins_dir + "RgTools-v1.0\x86\RgTools.dll")
	# MaskTools2
loadPlugin(plugins_dir + "masktools2-v2.2.23\x86\masktools2.dll")
	# MVTools
loadPlugin(plugins_dir + "mvtools-2.7.41-with-depans20200430\x86\mvtools2.dll")
	# FFT3DFilter
loadPlugin(plugins_dir + "FFT3dFilter-v2.6\x86\fft3dfilter.dll")
	# FFTW
loadPlugin(plugins_dir + "LoadDll\LoadDll.dll")
loadDll(plugins_dir + "fftw-3.3.5-dll32\libfftw3f-3.dll")
	# dfttest
loadPlugin(plugins_dir + "dfttest-v1.9.6\clang\x86\dfttest.dll")
	# dither
loadPlugin(plugins_dir + "dither-1.28.0\win32\dither.dll")

	# LSFmod
Import(plugins_dir + "LSFmod.v1.9.avsi")

### de-interlacing
deinterlaced=video_org_crop_adj.AssumeTFF().QTGMC(preset="slow", matchpreset="slow", matchpreset2="slow", sourcematch=3, tr1=2, tr2=1, NoiseTR=2, sharpness=0.1)

### convert to YV16
deinterlaced_yv16=deinterlaced.convertToYV16()

### denoising
denoised_yv16=deinterlaced_yv16.TemporalDegrain2(degrainTR=3)

### convert to YUY2
denoised=denoised_yv16.convertToYUY2()

### convert to YV12
denoised_yv12=denoised.convertToYV12()

### sharpening
sharpened_yv12=denoised_yv12.LSFmod(defaults="slow")

### convert to YUY2 with chroma from YUY2 color space
sharpened=sharpened_yv12.convertToYUY2().MergeChroma(denoised)

### add borders
video_restored=sharpened.addborders((crop_left+crop_right)/2,(crop_top+crop_bottom)/2,(crop_left+crop_right)/2,(crop_top+crop_bottom)/2)

return(video_restored)

[lollo]

I can see that he used temporal degrain and FFT3DFilter

To be more accurate, I do not use the "real" FFT spatial denoising (it should be enable in TemporalDegrain2 with postFFT=1). Only intrinsic TD2 denoising. The loaded FFT3Dfilter is used internally by TD2 for its processing.

Also, when processing a video, the order things should go in if Im not mistaken are 1st colour correction, then denoising, then deinterlacing, and then crop and resize?

1- level correction (if needed) and preliminary color correction

2- deinterlace

3- denoise

4- upscale (I never resize), but only for YouTube.

5- final color correction

For cropping I prefere to do it before deinterlacing to have clean borders without "blacks", but it must be done accordingly to mod2 requirements. Many do at the end before resize, which is also ok.

4 and 5 can be swapped

I do not use Hybrid, but Selur is here, he's a very nice person and will be help you if you have any doubt, just ask.

[Leanoric]

Ive been playing around with Hybrid the past few days, and have had some very good results denoising.

HQDN3D and TemporalDegrain have produced really good results on very noisy video, far better than what I could achieve in vdub. I can understand now why the advice is to learn Avisynth.

For anyone like myself, that would be out of their depth with Avisynth, I would say definitely give Hybrid a go.

[rgr]

Ive been playing around with Hybrid the past few days, and have had some very good results denoising.

HQDN3D and TemporalDegrain have produced really good results on very noisy video, far better than what I could achieve in vdub. I can understand now why the advice is to learn Avisynth.

For anyone like myself, that would be out of their depth with Avisynth, I would say definitely give Hybrid a go.

Check out Neat Video, but it's a paid program/plugin.

[lollo]

Check out Neat Video, but it's a paid program/plugin.

Some comparison from master poisondeathray (using a basic SMDegrain approach!): https://forum.videohelp.com/threads/403073-Why-is-Neat-video-the-best-video-denoiser/p...12#post2634000

The whole thread is very interesting.

Leanoric, now that you've taken the first step stay with AviSynth (but NeatVideo has some interest anyhow)