[x264-devel] Re: [Tom Jacobs <T.R.Jacobs at lboro.ac.uk>] Re: Re: Rif: Re: JM 9.6 vs. x264 rev 228: results

[Tuukka Toivonen]

>> i have attached 2 of the graphs (not sure if attachements work in the
>> mailing list or not) showing how the multisliced implementation works.

Kind of; they seem to go into a webpage.

>> these 2 graphs are taken with 25frame click with 1 ref frame. the major
>> difference between the sliced version seam to be at lower psnr the relative

Hmm... is the last number the number of slices, ie. 25-1-all-1 is no-sliced 
version?

>> bitrate of the multislices are bigger. this is understandable since the rate
>> control for multislice isnt that good.
>> 
>> having said this what do people think? is it aceptable (assuming im getting
>> good benifits from using my multi-prossor system)?

Yes. I think it is acceptable. I'm quite sure that this works relatively 
better than the ffmpeg MPEG-4 SMP encoding; I'm quite surprised that it's 
this way. The quality loss is less than 1% at high bit rates; with just two 
slices it's practically the same as no-slices version.

With ffmpeg, there was a significant quality loss with just two threads, 
and worse, the speed improvement wasn't that much (far from 2x).

At low bit rates it's a bit worse (<32 dB) but you wouldn't encode movies 
with that low quality... if that is your target.

I wonder when this multislice version gets integrated with x264? The 
results look quite good.

-- 
This is the x264-devel mailing-list
To unsubscribe, go to: http://developers.videolan.org/lists.html