![]() |
1 Attachment(s)
Quote:
With NTSC = 240 Quote:
An example,see screen |
I think we're having a language barrier issue.
I know the proper settings (and 288 works as a threshold for both PAL & NTSC if you stick to standard SD resolutions). I was deliberately trying to find out which wrong settings can trigger the decode error. |
4 Attachment(s)
Quote:
Huffyuv 2.1.1 encodes and decodes any Huffyuv file according to the rule that interlaced = true when lines > 288. The original CCE-SP patch replaced this with a global setting stored in an INI file. Quote:
Quote:
For ffmpeg's Huffyuv, the patch removing the hardcoded threshold (src code) mentions the incompatibility with older versions: Quote:
--- EDIT: I successfully decoded the file, at any rate. Huffyuv's debugging option was helpful. The "Ignore is-interlaced flag" setting causes the decoder to behave the same way as the CCE-SP patch versions before 0.2.4; the global Field Threshold setting from the INI is used to determine the decoding mode. I used AVISource("test pentrex 5 ATI MMC 7-10-15.avi",fourcc="HYMT") to force Huffyuv-MT to decode this HFYU file, since 2.2.0 lacks the "Ignore is-interlaced flag" setting and I don't feel like installing another version. Attachment 4974 Attachment 4976 288 / Field processing enabled Attachment 4975 576 / Field processing disabled (with Avisynth Info) Attachment 4977 The frame rate is bad, and I hear clicking in the audio track. |
Thanks, msgohan. Answered many questions.
|
3 Attachment(s)
From your post #37:
Quote:
The image you posted uses the script for 03.avi to process the dark clip 01.avi. The same script can't be used for both videos, they have different problems. The script that I used for the video shown in your image was different and involved 2 steps in 2 scripts. The script for Step1 is below. The original black borders (which are too bright and too noisy) were cropped from the video and the core image was run in Avisynth. The borders are replaced later in Step2. At the same time I loaded four VirtualDub filters and saved the output of that Step1 process as YV12 lossless (I used Lagarith lossless compression). Step1: Code:
AviSource("path\to\video\Huff_MT.avi")Step2. The saved YV12 video fom Step1 video was opened using script #2 below: Code:
AviSource("path\to\step1 video\HuffMT_01_Step1.avi")autolevels_0.6_20110109.dll, AvsRecursion.dll, awarpsharp.dll (for awarpsharp2), ContrastMask.avs, DeHalo_Alpha (as "DeHalo_alpha_MT2.avsi"), FFT3dGPU.dll, gradfun2db.dll, GradFun2DBmod.avsi,hqdn3d.dll, RemoveDirt_v09 (with RemoveDirt.dll and RemoveDirt.avs), RemoveSpotsMC.avs, rgTools-x86, TemporalDegrain.avs, and VariableBlur_070. These filters are core plugins and support files for TemporalDeGrain, ContrastMask, awarpsharp, etc. I have also attached "For_VirtualDub.zip". It contains two VirtualDub plugins you might not have: CamcorderColorDenoise 1.6MT and DeFlicker1.3beta1. Also attached is "hufftMT_01_Step1_VDub.vcf" with the settings I used for the 4 VirtualDub filters run with Step1. You could try running both scripts together as one, but the processing will be too slow and could sometimes hang VirtuaLDub. I often use two separate steps for very slow processing. |
Big thanks... am in the work.
I have used most filters long since, but only seldom, because I primarily only Capture. 50 devices always in order hold a lot of time needs. Yes. with 2 steps I know since RemoveNosieMC from heini011 http://forum.doom9.org/showthread.php?t=110078 But here mostly I do not have so bad material like now with Woodstock. The fact that with "QTGMC (InputType=1" I know of course... that "Mode=1" for progressive material is, in contrast to "Mode=0" this is for interlaced material. One reads a lot..., however, everything can not always keep in the brain if the brain is small.:o |
Video that is in decent condition requires much less work. Usually the video I work with doesn't require this much effort or time. But sometimes the video is so bad, you have no choice if you want better results. In any case, it is very rare that I can use exactly the same filters for every video, especially with analog source because it differs so widely.
|
2 Attachment(s)
Yes, is quite clear, there is not only a Script for all springs.
Although I have VDub and also still VDubMod, I have landed in the postprocessing with Edius. this has the advantage, I see the picture like it is given correctly also thus in the connected view device and not in the PC monitor which is good for Interlaced nothing. First becomes gecapturt, in YUY2... then only picture] with Avisynth TGMC is prefiltered of the Stream [in the fashion MT a little bit. This claps rather fast and I can go on working on another PC. |
2 Attachment(s)
As well as it looks I must remove a lot of newer Plugins to load old Plugins.
TemporalDeGrain from old 0.9 packages. my version is from 2011. ContrastMasc does not clap here, finds not all Plugins. RemoveSpotsMC also does not clap. In principle, no rhomb [#] of the line should be in front in it :D Must load all Plugins in a separate folder and then appeal. |
Quote:
|
2 Attachment(s)
Supplement
sanlyn... I must take for a ride the Plugins in a separate unterfolder and appeal separately, thus klappts. However, I may not cut so much, see picture |
Quote:
When you're ready to post another sample, I would like to see one direct from the VCR to the capture card, to eliminate more hardware variables. |
2 Attachment(s)
I believe there come with the 1st step a few wrong "dll" to the use.
Processing claps with from 3.40 to 3.60 fps. With my easy "first settlement" here also as the 1st step, it works in the MT fashion with from 19 to 20 fps. On the left in the picture = TGMC [the first settlement] On the right in the picture = 1st step/Avisynth with VDub File as mpeg2.... folds so it with the high-level store in the net. 2. Film clip = in Edius the Croma something reduces in left picture half. -------------------- Now this is only one example, as a rule it is encoded to AVC + AAC + in MBAFF. |
Quote:
Quote:
|
Quote:
|
Quote:
The default value of 288 works perfectly fine for both NTSC and PAL so long as you stick to only standard capture resolutions: those with heights of 240, 288, 480, and 576 (and there hasn't been a reason to capture below 480 since the early 2000s). Even the author of the CCE-SP patch who criticized the fact that 288 was hardcoded and originated the Field Threshold option admitted this! Quote:
|
Quote:
Quote:
|
Quote:
|
rockovids
Read sometimes the 2nd contribution Since you're capturing NTSC you should use field treshold of 240 lines, 288 is for PAL. http://forum.videohelp.com/threads/2...-about-huffyuv |
:argue:(Hee-Hee!) Thanks Goldwing ,This is the best smile icon that I could find for feeling like a ping-pong ball right now, but I will take what everyone says, and try it out, One thing for sure, I will take notes this time, about the final settings that do work on my system!... I do have the system restore option from 2012,but that is a last resort. Only about 5 years experience with computers, but I have learned from this site, that every situation is different, with VHS capture/software/hardware..for a while I was/am guilty of easy monkey-see,monkey-do!:yikes:
|
Site design, images and content © 2002-2026 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2026 Jelsoft Enterprises Ltd.