Which one is better?

[jagabo]

Of course, nobody can say for sure which of your sources is better without seeing them. The h.264 clip is potentially better because h.264 can compress better than Divx, the particular file has a higher resolution and higher bitrate. But what if the Divx file was well converted from a DVD rip and the h.264 files was converted from a VHS recording of an over-the-air broadcast from 200 miles away by someone who didn't know what they were doing? Now which do you think would be better?

In any case, in the week since the original post, if it wasn't obvious just by looking at the sources, you could have converted the files a dozen times over and seen for yourself.

[src2206]

Nope jagabo

I have really decided which one to go for, and i indirectly mentioned my choice while crying over my System Specs. Moe over I found very little difference to go for a nearly 14hr conversion process rather than using HEnc and a 5hr conversion process [+ Energy Bill]

Get the point!!!!

[jagabo]

5 to 14 hours to convert a 20 minute video is attrocious. An A64 XP 2800+ should be much faster than that.

CCE encodes 2 to 4 times realtime on my system (probably around two times faster than yours at video encoding, a Core 2 Duo E6300). A 20 minute video takes 5 to 10 minutes to encode (single pass constant quality).

X264 sources do take longer to convert than Xvid (without deblocking). About 50 percent longer. But Xvid with deblocking (note that h.264 decoders perform deblocking automatically, it's an option with Xvid) is slower than h.264.

[guns1inger]

With CCE I can do a 1+3 pass (analysis + encoding) VBR encode of a 90 minute film in around 5.5 - 6 hours on my system, with source and targets on separate physical discs. This is using avisynth the resize, and apply blockbuster and asharp filters.

The last time I saw anything like 14 hours was an eleven hour single pass encode to lagarith. It was a 2 hour 5 minute VHS capture that was undergoing to very heavy noise processing and color correction in avisynth. It then too another 6 hours to do a 1 + 3 VBR encode in CCE.

[src2206]

With CCE I can do a 1+3 pass (analysis + encoding) VBR encode of a 90 minute film in around 5.5 - 6 hours on my system, with source and targets on separate physical discs. This is using avisynth the resize, and apply blockbuster and asharp filters.

Would you please tell me about the process of setting 3+1 CCE?

Secondly, can you please enlighten me on the two filters [blockbuster and asharp filters] that you mentioned?

jagabo , the specs which are displayed here is not for the use of this kind of jobs, thats more of a professional type and without DVD Burner. The present system specs I have mentioned in one of my previous post.

[jagabo]

Would you please tell me about the process of setting 3+1 CCE?

Select Multipass VBR, 3 passes.

[src2206]

VBR?! But I thought CBR was better!!

[jagabo]

Then select Multipass CBR and 3 passes if that's what you want.

[Cornucopia]

The axiom is that, for all other things being equal (esp. motion search precision and GOP structure), VBR gives smaller filesizes than CBR for a given quality
-or-
VBR gives better quality than CBR for a given filesize.

But...
CBR should encode much quicker.

Personally, I see no point to multipass CBR. If I'm getting that picky, I'll be using VBR anyway.

Scott

[src2206]

OK thanks for clearing things yp.

It has been a great learning experience

Now I need to master the AVISynth Script.