Captured a 60gb 2hr .avi with Vdub/Huff, encoding to x264 takes 23+ hours??

[Groucho2004]

Well, this is weird. I took out all the MT lines out of the script (apart from SetMTMode() at the start) and removed SetMemoryMax and Distributor() as well, saved the script and am running the trimmed portion I did through AVSMeter, it's showing me Active MT Mode: 2.

Why would you leave SetMTMode in? Take it out.

[CZbwoi]

Yes, that should return frames 226614 to 230000 . Blue screen suggests some larger problem like memory corruption or HW issue.

I don't think he's talking about a BSOD. Maybe the frame range specified is outside the script's range.

@CZbwoi
We were posting at the same time but it seems I was right.

^ Yep. Also to now discuss another weird thing happening I just discovered, none of the videos I've encoded with MeGUI give me sound in WMP. In VLC there is sound, but not in WMP. A weird thing was happening in VLC with MeGUI encodes but I thought nothing of it until now: when I would skip far in the timeline upon clicking the sound would play and then go mute for a half-second before picking back up. Now I think there is some correlation since there is noise coming out when they're played through WMP. One step forward...two steps back...

I never changed the audio setting either, it's defaulted at FFmpeg AC-3: *scratchpad*

Why would you leave SetMTMode in?

If you read the other posts, I'm part of the group of people that has a problem if you take it all out. If I take it out the script doesn't work and I get that error I posted. Only solution is to switch out the .dll constantly or to add the SetMTMode() at the start like jagabo said. Some people like PDR don't need to it seems, but I do.

I still don't know why AVSMeter tells me I'm on Mode 2 though if all that's left in the script is SetMTMode().

My CPU usage was at 100% when I was encoding the 20min trim with MeGUI though (Constant Quality, 16 CRF, Slow), even though it was only at about 16-19 fps, so apart from the audio being wonky now at least there's hope there. The 20min clip took 1 hour and 4 minutes in total from the audio to the muxing. I don't know if that's a good encoding time for 20min and idk if it's happening and I got 100% CPU usage at the bottom of my task manager because I'm still (potentially) in MT Mode 2 while also...not being in it..?

[Groucho2004]

Why would you leave SetMTMode in?

If you read the other posts, I'm part of the group of people that has a problem if you take it all out. If I take it out the script doesn't work and I get that error I posted. Only solution is to switch out the .dll constantly or to add the SetMTMode() at the start like jagabo said. Some people like PDR don't need to it seems, but I do.

None of this makes any sense. I highly recommend you get familiar with at least the very basics of Avisynth. -> http://avisynth.nl/index.php/Main_Page

Why would you leave SetMTMode in?

I still don't know why AVSMeter tells me I'm on Mode 2 though if all that's left in the script is SetMTMode().

The answer can be found in the documentation:
Mode 2 is the default.
By placing "SetMTMode()" at the beginning of your script, you're running MT mode 2 with the number threads equivalent to the number of logical CPU cores of your processor. Moreover, you're running the source filter with MT mode 2 with is usually not recommended.

[poisondeathray]

I see the waveform appeared but I don't know how to decipher it atm and what you mean by Y<16 , Y>235, I'll have to look up guides or ask in another thread. Fix them with the tweak option though, right?

Yes, it's discussed in more detail in other threads. You can turn the histogram upright, so it looks more like traditional waveform like you would see in a NLE, this is discussed in the documentation page, and that's what you saw in your other thread.

http://avisynth.nl/index.php/Histogram

The brown/yellow areas are "illegal" . 16-235 is in the center section. It's generally a better idea to adjust it at the source capture. But you can use many methods , including a preset coloryuv(levels="pc->tv"), or smoothlevels(preset="pc2tv") . Those just "squish" everything into the legal range 16-235; some people might want to do a better job, maybe spread out the shadow contrast, highlight contrast - ie. do a more proper color correction. But this is a different topic, and lots of info covered in other threads

[_Al_]

That's the thing, I'm ways away from learning how batch files and commandlines work.

regarding that zip I posted, you create a INI file, or just use that one I posed for you, you can change temp and destination folder to your liking, different hardisks, for example D:\encoded\vhs .... whatever path . Or use notepad and copy that text I posted here as your INI, start notepad, paste it there,save it as my.INI, whatever name perhaps VHS.INI whatever. You can tell you windows to open INI files with Notepad so you just doubleclick INI and in opens, just like txt files.
Then you grab your AVI and that INI file and drop it on that AVI to MP4.BAT, that's it.
Then you can grab the same INI, different AVI and drop it on the same AVI to MP4.BAT. Then you would have running two instances at the same time.

Point is, you cannot grab two files from different folder , so perhaps you'd be forced to copy that my.INI into folder first where you AVI is.

Or just use command line to run it including paths and quotes into command line, paths included, so avi does not have to be in the same folder with that INI like in above example, just basic:

"F:\tools\Dmfs avi to Mp4\_AVI_to_MP4.BAT" "f:capture\huffyuv\vhs.avi" "F:\tools\Dmfs avi to Mp4\my.INI"

you can create this command line with drag&drop windows feature also, since win Vista I think, just click windows icon or START in desktop, down left corner, type cmd click cmd.exe, then drag&drop _AVI_to_MP4.BAT first from windows file browser
then hit space bar (there has to be space character between those dropped files)
then drag&drop there your my.INI
hit spacebar
the drag&drop your vhs.avi there
you just created command line, hit ENTER to run it

It is really nothing special, it is not some geek thing really, it just looks like it. But basic operations are used.

[CZbwoi]

Why would you leave SetMTMode in?

If you read the other posts, I'm part of the group of people that has a problem if you take it all out. If I take it out the script doesn't work and I get that error I posted. Only solution is to switch out the .dll constantly or to add the SetMTMode() at the start like jagabo said. Some people like PDR don't need to it seems, but I do.

None of this makes any sense. I highly recommend you get familiar with at least the very basics of Avisynth. -> http://avisynth.nl/index.php/Main_Page

Why would you leave SetMTMode in?

I still don't know why AVSMeter tells me I'm on Mode 2 though if all that's left in the script is SetMTMode().

The answer can be found in the documentation:
Mode 2 is the default.
By placing "SetMTMode()" at the beginning of your script, you're running MT mode 2 with the number threads equivalent to the number of logical CPU cores of your processor. Moreover, you're running the source filter with MT mode 2 with is usually not recommended.

If that's the case then I guess jagabo was wrong, if he meant that's how you turn MT Mode off without replacing the .dll. How is that a bad thing though? You're using up all the available cores, isn't that the main goal so it goes faster?

P.s. so is using SetMTMode() at the start still slightly 'degrading' the quality of it since you're not at the natural 1 thread avisynth setting?

The brown/yellow areas are "illegal" . 16-235 is in the center section. It's generally a better idea to adjust it at the source capture. But you can use many methods , including a preset coloryuv(levels="pc->tv"), or smoothlevels(preset="pc2tv") . Those just "squish" everything into the legal range 16-235; some people might want to do a better job, maybe spread out the shadow contrast, highlight contrast - ie. do a more proper color correction. But this is a different topic, and lots of info covered in other threads

So just make no part of the waves touch the brown? Have it right on the edge? I'll have to figure out how to pull that up in Vdub then so I can set it up right when capturing...


_Al_: I'll try it out but I don't know what x264_variables and audio/nero_variables I need to enter for myself first off. I don't know what bitrates to choose, what sizes, what tune film or buff sizes means. That's the stuff I have to learn first before firing up a commandline, because I usually would just let a GUI do it after picking a quality/preset

[jagabo]

If that's the case then I guess jagabo was wrong, if he meant that's how you turn MT Mode off without replacing the .dll.

I didn't say that SetMtMode with no arguments turned off multithreading. I said it get around the problem you were seeing in post #54. You don't have to call it with no arguments. You force single threaded operation by specifying one thread in SetMtMode, SetMtMode(x, 1). The number of threads can't be changed later in the script.

P.s. so is using SetMTMode() at the start still slightly 'degrading' the quality of it since you're not at the natural 1 thread avisynth setting?

The issue is that some source filters don't work right in mode 2. The solution is to set another mode first, then switch to mode 2 after the source filter.

SetMtMode(5,4)  # mode 5, 4 threads, mode 3 often works too
WhateverSource()
SetMtMode(2) # mode 2, can't change threads
OtherFilters()

[_Al_]

_Al_: I'll try it out but I don't know what x264_variables and audio/nero_variables I need to enter for myself first off. I don't know what bitrates to choose, what sizes, what tune film or buff sizes means. That's the stuff I have to learn first before firing up a commandline, because I usually would just let a GUI do it after picking a quality/preset

so why not to try it, that INI i gave you,
CRF does not set bitrate, it is quality setting, that what you should use, like CRF 16 or CRF 17, you are setting it in Megui also
tune film just keeps "certain" film noise in the video, for 8bit encoding and standard definition video I'd definitely keep it there, others might have different opinion, or you do not put it there but chances you see banding in your video are a bit higher, banding is bad in SD video, that video blown up on fullHD screens makes it look terrible, so any chance to reduce banding is good. Banding is those live , shapeless, constantly changing spots, of colors in dark areas or color gradients. tune film is a mild version of keeping noise, it was designed for film, there is even more powerfull setting -- tune grain , so you see, that you might even use if you are concern about quallity, video size would increase some 10% or more, but you want it look close to original, that is what you should use
buffers leave as it is there in your INI, for standard definition H.264 video, 20.000 is more than enough,
nero_variables are command line variables for AAC audio, -lc -cbr 256000 leave them as they are, -lc means nice and dandy profile,-cbr 256000 is constant bitrate 256kbps, you can change that to 384 if it is not enough for you for some reason. Instead of -lc -cbr 256000 you can use variable bitrate, quality settings like -q 0.5, that should be enough, you can set it higher to 0.7 or so, but audio is misery coming from VHS so anyway, but I do not like variable settings for audio if bitrates are negligent in the first place ... I'd recommend that CBR setting for audio and lc because it is more than enough here, no devices have problem with it, and also that 256 bitrate is nothing comparing to bitrates for video if you think of it