Who's got a working VirtualDub Frameserver to CCE method ?

[maa]

Searched all over the place and read reams on MSVCR70.DLL and the alternative AVIsynth which seems too complicated - SO - can it be done without the "This is not an AVI file" message ?
(Reason: Joined AVI is to big for FAT32 output)


VirtualDub 1.5.4.1
CCE 2.66
thanks

maa

[maa]

So I gather the frame server in VDub doesn't work and one has to use it in a script for avisynth which outputs another signpost which then will work ?

Isn't there anything more workable than that ?
I didn't want to spend the whole weekend doing this - maybe I'll compress to MJPEG and feed that to CCE insetead....

[Kedirekin]

I've used the vDub frame server on several occasions (to CCE 2.5). It worked for me, though it is a bit slower than AviSynth. This was (and is) an older version of VDub (1.4.7), but I assume the frame server hasn't changed that much.

I don't know what steps you've already followed, so some of this might be repeat for you.

You have to go into the Virtual Dub directory, run AuxSetup and install the frame client handler. Then you
- start up vDub
- open your avi
- start the frame server
- save the signpost file (make sure you actually type the .vdr extension in the save-as box - I don't think vDub adds the extension for you and things won't work without it)
- leave the frame server running (I don't know if that is obvious, but it's very important)
- start CCE, click add, show files of all types and select your .vdr file

You can even start a second instance of vDub, open you .vdr file and confirm that the frame server is working properly.

As a matter of fact, just to make sure I'm not losing my mind, I just encoded a 300 frame segment of an avi frame served by vDub. It encoded just fine.

Oh, don't forget to run AuxSetup and uninstall the frame client handler when you're done. As I recall, it has a tendency to mess other things up if you leave it installed.

[DDogg]

So I gather the frame server in VDub doesn't work and one has to use it in a script for avisynth which outputs another signpost which then will work ? No, the vdr will load just fine directly into CCE and does not require avisynth. It is extremely easy as Kedirekin suggests.

However, if the source was a capture and you had edited out multiple commercials, you could serve the vdr directly, or if you wish, you could convert the saved VCF file into an avisynth script by use of bb's vcf2avs_gui.exe tool. It converts the multiple cut points stored in the VCF file into avisynth trim statements. Some people like doing it this way for various reasons specific to their methods.

[maa]

Oh dear!
I'm very sorry but I was misslead by some "guide" and have been naming the file *.vdr.avi all the time.
Of course it works as you describe - thanks for the tips.

One more question - what do I set in CCE for interlaced input ?
The product is for TV not Compter.

thanks

maa

[DSP8000]

alternate scan,DVD compliant,luminance level 16-255,intra DC precision 9,GOP sequence N=3,N/M=4,quality default 25 for CCE 2.50
DSP8000

[Kedirekin]

The only settings that apply specially to interlaced versus progressive (IMO) are Upper Field First and Progressive Frames.

Upper Field First needs to match your footage - I wish I could give advice here. How to tell up field first or down field first (http://forum.doom9.org/showthread.php?s=&threadid=69870) gives you a method for determining field order, but I can't remember what setting for Upper Field First corresponds to what. In my experience, unchecking it works most of the time, but that's with movies, not DV footage. A forum search might be advisable on this. And you can usually fix it using pulldown.exe without re-encoding if you get it wrong.

Progressive frames (I believe) changes the encoding strategy. Even if you check it, the encode will still work, but (presumably) the quality for interlaced footage will be lower (though I've never been able to see any difference).

I've seen web pages that indicate Linear Quantizer and ZigZag Scanning Order should be different between progressive and interlaced, but again I've never been able to see any difference. Because of that, I haven't bothered to remember what was suggested for what. My personal preference is Linear unchecked and ZigZag checked, but as far as I'm concerned you could almost flip a coin.

One other thing that is somewhat related to interlaced is GOP size (12 frames or 15 frames). It's really related to frame rate rather than interlaced, and truthfully it's another matter of personal preference. If you're doing NTSC interlaced (or 30 fps progressive for that matter), you might consider setting GOP to M=3, N/M=5. If you're doing film (24 fps - noninterlaced of course) or PAL interlaced, use M=3, N/M=4 as DSP8000 suggests. These settings keep the GOP length to approximately ½ second. Again, it's a matter of personal preference; there are opinions and arguments both ways on which setting produces better quality. I think quality is a wash (equal) either way, and ½ second GOPs just feels right.

[DDogg]

To this day I still have to reference Q10 and Q11 of RB's CCE FAQ (http://forum.doom9.org/showthread.php?s=&threadid=53770) to keep it straight in my mind. I have never understood why the authors of CCE made it so complicated.

From the faq:
CCE 2.5
So here is the rule of thumb: Always uncheck "Upper Field First" unless your video is interlaced AND bottom field first. Progressive material is always top field first.

CCE 2.67
Again here is the rule of thumb: Always set "Offset Line" to 0 unless your video is interlaced AND bottom field first in which case you set it to 1. Progressive material is always top field first.

[maa]

Thanks for the usefull tips again.
The possibilities here go into the thousands I'm sure.
Untill I've tried your suggestions, here is what I got so far.
Interlaced input is performing realy badly - I mean the encoded material is terrible. If I deinterlace and add 2 steps of timberal softener (niether of which I want because the original captured avi is fine) then the CCE output is excellent !!


So I'll try the above and post back.

maa

[DSP8000]

Well said,Kedirekin, I was reffering to a PAL source.I agree with DDogg about the CCE manuals,FAQ's,but that is how it works.Honnestly since I got Canopus Procoder Express I hardly even use CCE Basic.To my eyes CPE does better job for my DV AVI's.Funny thing is that if you choose PAL DVD preset in MainConcept 1.4.1 the GOP structure is the same as NTSC?Also about determining field order someone here mentioned that you can check with TMPGEnc,but I've never did it in this way.Is it accurate?
DSP8000

[Stabmaster-Arson]

To determine field order with tmpgenc:

Following will teach you how to determine the field order of your video for interlaced encoding. If the field order is wrong your movie will be jerky and techinically be ruined.

Tools needed: tmpgenc


Selecting the correct field order:
Fire up tmpgenc then do this.
Press the "Settings" button and go to the "Advanced" tab. Under the "Field
order" setting, pick "Top field first (field A)". Highlight the
"DeInterlace" option in the list and double click it to open the DeInterlace
dialog. Select the "Even-Odd field (field)" Method. Left-Click on the video
area and then using the keyboard arrows move to the right to frame-advance
the video. If the motion is smooth, then the field order is correct, if it's
all jumpy then exit the DeInterlace dialog, switch the Field order to
"Bottom field first (field B)", re-enter the DeInterlace dialog and check
that now the movement is smooth. When the correct field order is selected,
make sure that the DeInterlace option is checked OFF! We only use it to
check the field order, it must not be enabled for the final encoding.