For instance, yes.

:mrgreen: :rotf:
Ever heard of a WOW :D / Damm :x girl :?:
When you see them from behind, you say WOW :D :!:
When they turn their face towards you, you say DAMN :x

:hihi:

Kwag, can you tell me what should I change to the original KVCD templates (for example KVCDx3 ntsc) if I use constant bitrate? I changed the GOP to 18 but I'm not sure that this is everything.

You should use the standard MPEG matrix if you are using constant bitrate.
The matrix ("Notch") is designed to work with variable bitrate, and take compression to it's advantage. So you'll get better results setting the standard matrix. Also, the GOP shouldn't be too long, and a value of 18 should be about the largest value for a CBR encode.
If you insist on using CBR, because you don't want to do prediction, then may I suggest something?
Instead of using 1,150Kbps, use 1,450Kbps MAX and 300Kbps MIN and use MVBR mode.
This will give you better quality that ANY CBR VCD, and still, I'm alost 100% sure that it will give you 80+ minutes on one CD-R. The trick is the low bitrate on MIN, which is way below the MAX.
So for 1 hour TV series, you can almost be sure that you'll always fit it on one CD, probably with a lot of space left. This way, you can use KVDC's "Notch" matrix, and a GOP of 24.
This settings are excelent for 352x240. If you are going to encode at higher resolutions, then the MAX bitrate of 1,450Kbps will not help you, and you'll get blocks on high action scenes.

Edit: For KVCDx3, you'll need AT LEAST 2,000Kbps MAX bitrate for a descent picture (Using the MA script ), and don't expect more than ~45 minutes per CD-R.


-kwag

Thank you very much, Kwag

I use mpeg-1 CBR of ~2400kbps on my captures of TV episodes. Running prediction on a file that is always going to end up wih an average bitrate that is about 100kbps less than my max doesn't make sense. I know I could raise my max bitrate to say 3000 or 3500 so as to help in high acion scenes but, the MA script already takes care of this, it sort of does in filtering what an encoder does with vbr.

Sorry, but that is just not right.
MA filtering in our case only allows the encoder to use a lower bitrate for high action scenes, without a visible degredation in quality by blurring the picture. That's the purpose of the MA script, so it would make absolutely no sense to use a high CBR together with the MA script, as it will only degrade quality of high action scenes, and will not lead to a better compression for that scene! :roll:

The types of scifi tv shows that I capture have lots of action, explosions, and stroboscobic effects that come out looking terrible at less than DVD bitrates. The MA script helps a lot since I usually encode at 704x480. Perhaps I should analyze the captures individually and adjust my scripts accordingly, any tips would be welcome. My sources are hardware encoded mpeg-2 captures at between 8 to 10 Mbps CBR and each episode is encoded to one CD at about 42 to 44 minutes per disc. Thanks in advance.

Well that's exactly what I said, if your play-time per media is at only 45 minutes, of course you're able to have a great quality EVEN in CBR encodes at 704x480.
Well now try the same captured clip in CQ mode, with prediction and everything and the MA routine, and you'll be amazed by the quality.
It's just logical thinking that it has to be better, because the high bitrates are used in those parts of the movie where they're needed... :wink:
That's my tip! 8)

This sounds good. 8)