[Lossless] Ut Video Codec

[Lirinis]

another was done by Zarxrax and can be found here.

What colorspace did he use? What version of HuffYUV? How did he measure those fps? How to do it right?

I tried encoding to YV12 and then rendering a preview in AE and adaptive HuffYUV slightly outpermormed Ut in rendering time and significantly in encoding. Maybe I was doing something wrong. I'm trying to find out.

Intel CPU.

[mirkosp]

YV12 colorspace. Not sure about the version of HuffYUV, might have been neuron2's though. I know that ffdshow's HuffYUV is better than neuron2's, but I am *sure* that DeathWolf's comparison had that too and UTVideo still outperformed it (though DeathWolf's comparison had the source on ramdisk for better comparison purposes).
As for measuring fps I'm not sure how he did do it.

EDIT: Found DeathWolf's comparisons, first one is older without UTVideo, latter has utvideo but fewer codecs:

Code: Select all

Bored lossless encoding test
YV12 848x480@23.976, source on ramdisk, x264 is in lossless mode, destination null

Encode Test, sorted by size
===========================
Codec        FPS     Size(MB) Comment
X264         58.26   180.81   (multi threaded,default)
X264         16.98   180.81   (mono threaded,default)
X264         65.24   182.40   (multi threaded,fast)
X264         18.74   182.40   (mono threaded,fast)
X264         486.81  234.41   (multi threaded,ultrafast)
X264         155.42  234.41   (mono threaded,ultrafast)	
X264         147.47  284.27   (mono threaded,ultrafast, -I 3)
FFV1         54.90   289.07   (default, yv12 colorspace)
X264         145.77  306.96   (mono threaded,ultrafast, -I 2)
Lagarith     110.51  340.63   ()
FFHuffYV12   287.33  383.39   (default, yv12 colorspace)
X264         136.96  390.37   (mono threaded,ultrafast,all intra)
HuffYV12     615.80  443.91   ()
VBLE         261.22  458.22   ()
Null Codec   1724.4  1274.9   ()


Decode Test, sorted by speed
============================
Codec        FPS     Comment
FFV1         42.46   (default, yv12 colorspace)
VBLE         46.34   ()
Huffyv12     86.20   ()
Lagarith     88.86   (yv12 mode)
FFMS2 X264   99.78   (default)
FFMS2 X264   100.23  (fast)
FFMS2 X264   159.63  (ultrafast,all intra, much faster seeking)
FFMS2 X264   187.40  (ultrafast,-I 2, faster seeking)
FFMS2 X264   195.92  (ultrafast,-I 3, fast seeking)
FFHuffYV12   221.03  (default, yv12 colorspace)
FFMS2 X264   222.09  (ultrafast)
Null Codec   1725.8  ()I changed the thread title to be more specific so hopefully you won't be objecting anymore.

Note: please note that h264 is slow to seek:) so if you intend to browse around the stream, either fixing stuff/looking for stuff, it's not an ideal choice

Code: Select all

Bored lossless encoding test V2
YV12 960x540@23.976, source on ramdisk, x264 is in lossless mode, 6000 frames

Encode Test, sorted by speed
===========================
Codec        FPS     Size(MB) Comment
FFV1         39.02   985.24   (default)
Lagarith     72.73   1166.94  ()
VBLE         233.22  1292.13  ()
FFHuffYV12   292.68  1196.25  (default, yv12 colorspace)
HuffYV12     521.74  1412.96  ()
Ut Video 420 727.36  1340.66  ()
Ut Video 420 774.29  1199.44  (predict median)
Null Codec   2181.8  4449.46  ()


Decode Test, sorted by speed
============================
Codec        FPS     Comment
FFV1         29.74   (default)
VBLE         35.66	   ()
Huffyv12     69.97   ()
Lagarith     72.73   (yv12 mode)
FFHuffYV12   173.91  (default, yv12 colorspace)
Ut Video 420 399.39  (predict median)
Ut Video 420 523.95  ()

[Mister Hatt]

For the record, YV* is not a colourspace, it's a shorthand way of writing the chroma subsampling ratio for planar colourspaced video. 420 will always be smaller than 422 which will be smaller than 444. Exact ratios are 1.5 bytes per pixel in YV12, 2 bpp in YUY2, and 3 bpp without subsampling. From that you can tell that your YV12 video should be half of the size of your fully sampled video, or 3/4 of the size of your YUY2. Good way to know if your lossless codec configuration was done properly in comparisons too. Just sayin'.

[Ikore]

I was wondering, why exactly do you all say that UT is so fast. Did you test in real projects? Under what conditions would I see that?

this is my infommercial, I started using UT in my crappy pc with 2 gb ram and a really bad pentium 4 processor.
with lagaryth I was unable to use 720p to edit, right now Im editing at 720p with UT video codec, almost no lag, so yes this thing is freaking faster.

[HalOfBorg]

My netbook plays UT files when it can barely handle Lagarith files.

[Lirinis]

For the record, YV* is not a colourspace, it's a shorthand way of writing the chroma subsampling ratio for planar colourspaced video. 420 will always be smaller than 422 which will be smaller than 444. Exact ratios are 1.5 bytes per pixel in YV12, 2 bpp in YUY2, and 3 bpp without subsampling. From that you can tell that your YV12 video should be half of the size of your fully sampled video, or 3/4 of the size of your YUY2.

So what?

Good way to know if your lossless codec configuration was done properly in comparisons too. Just sayin'.

I was experimenting on the 9th episode of Ore no Imouto (24mn 30s, 1280x720).
Results:

By rendering here I mean rendering a 30 sec 1-layer full-quality preview in AE.

Lagarith is slow but small. HuffYUV is a bit faster than Ut and even smaller if compared to speed-optimized Ut. And it is way faster in encoding.

[Cannonaire]

I personally choose UTVideo optimized for decoding speed. Generally when I'm encoding it isn't a very time-sensitive process. When I;'m sitting down to edit, I like to have as much speed as possible. Yes, ~76fps is more than you need for editing, but you also should take into account everything else that's using your CPU time. The quicker it is to decode, the less CPU cycles it takes to decode in real-time, which helps out a lot when you're using any kind of effects and fades.