Custom CD sizes
 

Hi SansGrip,

I'm only new to this "file prediction thing", but it is way better than encoding a whole movie over night just to find you have to do it again! Very tedious.

The Predictor looks great... I do a lot of TV episode convertions to CD and a 1 hour show will normaly come out at 40-43 mins less ads. So 2 episodes per CD. Could the Predictor be modified to allow for custom size CD's? May even get 3 on a CD if I'm lucky!

An open question... Do you really need the resize/crop/border commands in the .avs file created from FitCD? Would not TPMG be performing this function?

The only dumb questions are unasked ones! We all gotta learn...



Hi Kwag,

Thanks for all the work on the kVCD templates! Best to ask this by example...

If I use one of the kVCDplus templates and its starting CQ is say 50, and I use the kVCD Predictor and to squeeze a movie onto 1 CD and the CQ to do this is say 22, how far can the CQ be reduced before the quality starts to really degrade? I guess the real question is at what point are 2 CD's better than 1?

Re: Custom CD sizes
 

Hi jackl,

The best way to find out is to encode a sample with prediction. This way you know exactly what the movie is going to look like, once you find your correct CQ or CQ_VBR value. If the sample quality is not good for you, you can then select two CDs for target instead of one, or you can lower the resolution by selecting a lower resolution template. Right now with the current GOP, I believe we can fit ANY 2+ hour movie in two CDs with quality almost equal to the source. I've seen the samples for LOTR, targeted for two CDs, and it looks just like the DVD :D
So just play with KVCDPredictor for your movies, or use the manual formula if you want to target 2 or 3 episodes on a single CD-R. :wink:


-kwag

Does the .81 apply with tp the sticky formulaa ?

Hi LadyMiles,

You can set the factor to 1.0 with the latest GOP and prediction ;)

-kwag

Sorry, my brain is a little dead from packing ? But how do you apply a factor of 1.0 the formula in the sticky that uses Sansgrips Sampler

Yes. 1.0 = nothing.
No factor ;)

-kwag