Help with Best x264 Settings and Virtualdub Avidemux

[ZZen]

Hey everyone, i need lots of help lol. The default x264vfw default settings in virtualdub seem quite good to me to archive clips and post to youtube ect, I dont need to put the CRF rate factor higher its good as is default 26.0 but is there anything else i can do do increase the quality even more while maintaining near the same file size? I don't mind waiting longer for encoding for better quality.

someone said to change x264's ME algorithm in [Analysis and Encoding] tab to Tesa instead of the default UMH takes longer but gives better quality,is that true should i do that? anything else i can do like that too to improve quality, long as there should be no sync issues, by that i mostly mean trouble joining clips

One of the things i'm worried about is changing settings that might cause sync issues/ trouble joining my clips since i will be connecting so many clips, and will it be fine to cut edit join the x264 avi mp3 clips i make in virtualdub? I will have to because i get an error when i try to append my source videos in virtualdub, vdub says the video streams have incompatible data formats, though they are recorded with the same program, Xfire not Fraps, and in avidemux my source videos are just green, and im nooby with avidmeux.

and if i do have to use lame mp3 i should use CBR i heard so there wont be sync issues or trouble joining clips?

I'm a bit unsure about 2pass encoding and dont like that it isn't automatic with x264 in virtualdub would that be better than
default single pass CRF? im sure 2pass will give better quality but is that basically at the same file size? or its always bigger, or maybe smaller ^^ I also dont know what bitrate will be equal to the default 26.0 CRF single pass if i use 2 pass, probably near 2500 3000k, All info you all have on 2 pass is welcome ^^

Im not even sure of the proper way to do 2 pass, name the 1st pass then overwrite that same file with the 2nd pass or?

also someone said you have to delete 1st passes stats file for each video because the next videos 1st pass wont overwrite the previous videos 1st pass data it will keep the first videos data, is that true? any easy way to auto 2 pass in vdub x264?

Handbrake seemed exactly like what i wanted for encoding automatic 2 pass h.264 and acc audio but it does its CLI file reading thing then says "No titles found please make sure you have selected a valid non copy protected source. your source may be copy protected,badly mastered or a format handbrake doesn't support" another Xfire video codec problem i guess
Can SUPER give the same quality that Handbrake can? they seem similar


And in vdub i must use mp3 cant use AAC audio right? since its avi.
I also cant seem to add lame to virtualdub 1.8.8 how do i? in past versions it worked for me

is there any big advantage in quality/size ect using mp4 with avid or something, instead of x264 avi in virtualdub?

mp4 trouble,with mp4 files in avidemux the file stutters when i play it is that normal? do i need a different h264 codec for avidemux than the virtualdub one?

When i add x264 files created with vdub to avidemux it says h264 detected something about b frames and asks to use safe mode an xvid file it asks if u want to unpack streams, i dont know what to choose for these and with either file i can barely scroll through it i just found out now u have to hold shift but still it freezes a bit at some parts, all these thingss just make avidemux seem odd to use. maybe ill try YAMB later.

i would use avidemux h264 AAC audio mp4's but im scared of sync issues or some other problem with mp4's when i join all my clips or cut parts should i just stick with vdub mp3 h264 avi's, or is cutting adding editing mp4's just as safe and easy as it is with AVI's?

in virtualdub there is save as avi, and save as old format avi?? i used F7 to save often whats the difference between these

there's not really any better codec than the h.264 x.264 one is there?

Thanks for all you help guys, sorry for all the questions

can virtualdub avidmeux or other free things give video quality as good as say sony vegas? this guy uses vagas im trying to match his quality it's very clear in HD

http://www.youtube.com/watch?v=buiAqo8UHBs&feature=channel_page

[ZZen]

Oh i forgot 1 other weird thing, any codec i compress with, the video looks kind of grey with less color or too bright played in vlc wmp. in wmp i can turn down the brightness below default and it looks almost normal, but when posted to xfire's site or youtube that same file looks normal like the source video not over bright/Grey or anything, why's that?

[ZZen]

i was wrong about the trouble with trying to scroll through a file in avidemux, its only with shorts clips, longer clips seem to be easy to scroll through and don't need to hold shift on them

[jagabo]

Oh i forgot 1 other weird thing, any codec i compress with, the video looks kind of grey with less color or too bright played in vlc wmp. in wmp i can turn down the brightness below default and it looks almost normal, but when posted to xfire's site or youtube that same file looks normal like the source video not over bright/Grey or anything, why's that?

Open the exact same video in two players at the same time. Do they look different?

i was wrong about the trouble with trying to scroll through a file in avidemux, its only with shorts clips, longer clips seem to be easy to scroll through and don't need to hold shift on them

This is more likely the GOP length of the source video. Long GOPs can take a long time to seek to random frames.

someone said to change x264's ME algorithm in [Analysis and Encoding] tab to Tesa instead of the default UMH takes longer but gives better quality

In CRF mode the more motion vectors the encoder finds the smaller the file will be. In my experience, the difference between UMH and Tesa is typically less than 1 percent. In 2-pass mode the difference in quality (ie, using 1 percent higher bitrate to get the same file size) will not be visible.

I'm a bit unsure about 2pass encoding and dont like that it isn't automatic with x264 in virtualdub would that be better than default single pass CRF? im sure 2pass will give better quality but is that basically at the same file size? or its always bigger, or maybe smaller ^^ I also dont know what bitrate will be equal to the default 26.0 CRF single pass if i use 2 pass, probably near 2500 3000k, All info you all have on 2 pass is welcome

They are two sides of the same coin. In single pass CRF encoding you are selecting the quality you want and the encoder delivers whatever file size is necessary to give you that quality. In 2-pass encoding you select the file size you want and the encoder delivers whatever quality it can with that file size.

2-pass may have a slight advantage in that it can take into account short temporal changes. For example, in an otherwise boring shot where there is a sudden flash of light (like a photographer taking a flash photo) constant quality encoding might waste a little bitrate on the sudden bright frame and the sudden change going back to normal (because you've asked for all frames to meet a particular quality). 2-pass encoding might look at that shot and decide you won't really notice if that sudden bright frame in the middle of the shot is encoded with a lower quality. So it will encode the frame and change back to a dark frame with fewer bits.

[ZZen]

I see what it is with avidemux now it just scrolls by every 250 frames,unlike virtualdub every 3 or 4 frames that im used to.

And yeah in VLC WMP they look the same, but looks the same as the source in virtualdub or when uploaded, its not much of a problem though.

[ZZen]

Played in VLC WMP the video looks a little bit gray little less color and over bright, but if i take a snapshot of it in VLC it gives a 1.4 mb PNG image the looks normal
And my recorded game footage uncompressed also looks normal

[jagabo]

Adjust your graphics card's video overlay settings.

[ZZen]

Thanks guys, yup jagabo is right.

The only mystery is why they look normal rendered from vegas

[jagabo]

The only mystery is why they look normal rendered from vegas :)

Editors usually don't use video overlay -- except sometimes in preview mode.

[ZZen]

I meant to say if i render with MainConcept Avc/Aac.mp4 in vegas the files don't end up with that brightness effect it looks more like the source.

Video overlay doesn't effect that codec? I thought it was the same or similar to H264 X264

[jagabo]

I meant to say if i render with MainConcept Avc/Aac.mp4 in vegas the files don't end up with that brightness effect it looks more like the source.

I suspect you were viewing two videos at the same time when you made this observation. Only one program can use video overlay at at time. The other has to use windows GDI and isn't effected by video overlay settings. One other possibility I can think of -- you were using Quicktime Player to view the AVC video. It doesn't use video overlay at all.

[ZZen]

Yup you are right I've noticed playing two three videos at a time after changing the graphic card settings only the first will stay normal looking all the rest will be the old bright grey.

But with just 1 open MainConcept Avc/Aac.mp4 in vlc kmplayer wmp it really is dark, actually a bit darker than the source its okay i probably wont use that codec anyway

Ill try Quicktime then. btw you are so smart such a video pro

[jagabo]

Quicktime player is the worst. It's a CPU hog and it tears a lot. I wouldn't bother installing it unless you just want to test.

[jagabo]

You asked in a PM but I'll respond here in case anyone else wants to know.

Tearing is when you see part of one frame of the video at the top of the image and part of another frame at the bottom of the image at the same time. This occurs when a program is updating video memory as it's being sent to the monitor. It's most obvious during panning shots:



In this (simulated) example the player started copying the new frame to video memory to the frame buffer from the top down, but the frame was sent to the monitor before the update completed. So the top part of the image contains the new frame, the bottom half contains the last frame. The tear line will move up and down the frame and you may even see more than one tear.

Smarter players will use hardware double or triple buffering. The swap from one frame to the next (at the monitor) only happens when the frame buffer has been completely updated.