Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > Video Encoding > New and alternative video codecs

Reply
 
Thread Tools Search this Thread Display Modes
Old 15th March 2011, 14:44   #1  |  Link
whipdancer
Registered User
 
Join Date: Jul 2009
Posts: 104
WMV HD Success

I realize this is not a very popular codec or topic, but given the help I received from this forum, I thought I would post about some success I've had...

Just thought I would throw this out there, as I've been looking at methods of encoding to WMV HD for the last couple of years.

After some updates to Win7 broke compatibility with the Windows Media 9 & 11 tools, I was no longer able to use cscript & Zambelli's WMCmd script, WMEncoder, or any of the tools originally available with the WM9 and WM11 SDK's.

After many frustrating hours of trying to get the original tools to work again - I gave up and decided to reimage my pc, create a baseline installation image and test various software & codecs.

If anyone is interested, I can post the full results, but the working solution I have right now is Expression Encoder 4 (Free Edition), Haali splitter, CoreAVC codec, AC3Filter. With this combination, I have successfully transcoded HD (both AVCHD & VC1) m2ts files into WMV HD, using CBR, CBR 2 pass, VBR, VBR constrained. This has also worked with VOB (regular DVD) files as well.

Again, thanks to many on the forums for being part of my education on video and encoding/decoding.

~Whip
whipdancer is offline   Reply With Quote
Old 17th March 2011, 00:36   #2  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 2,984
Well, that's nice to hear!

Expression Encoder is the official replacement for all the now-deprecated Windows Media 9 Series tools. Make sure you've got SP1 installed, which among other things fixes some bugs with reading sources.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 19th March 2011, 12:29   #3  |  Link
CruNcher
Registered User
 
CruNcher's Avatar
 
Join Date: Apr 2002
Location: Germany
Posts: 4,949
Im currently trying to buildup a framework to capture the windows screen content efficiently (with low frame drops as possible near 0 @ 30 fps) so near realtime (lossy) i tried different methods as of yet and it seems a daunting task doing it multithreaded efficiently especially on non DWM (Direct3D) XP (Directdraw), i also tried Expression Encoder in that and i can say its heavy how much resources it takes even on low complexity even on Sandy bridge, even @ 800 kbps i get heavy frame drops with Expression Encoder and the 4 cpu cores are crunching like crazy with x264 i already got nice results and next is evaluating vp8 for this task, im also experimenting with taken of load and latency on the capturing side via the GPU (utilizing Quicksync on 7). Also tried Microsofts Experimental Screen Capture Codec but that one is not really usable for so High Motion switches @ all it drops even crazier
So my question would be is the Core Encoder still adapted and improved Performance wise (especially Multithreaded Performance) or has that process finished on Microsofts side and is only continued with 3rd party implementations, outside of Expression Encoder ?
__________________
all my compares are riddles so please try to decipher them yourselves :)

It is about Time

Join the Revolution NOW before it is to Late !

http://forum.doom9.org/showthread.php?t=168004

Last edited by CruNcher; 19th March 2011 at 14:49.
CruNcher is offline   Reply With Quote
Old 19th March 2011, 17:42   #4  |  Link
Leeloo Minaļ
Registered User
 
Join Date: Nov 2007
Posts: 50
Did you try lossless codecs for your capture job ?
Since you have a good CPU, take Ut Video codec which works well with multi-cores, and you will get your 30 fps...
Leeloo Minaļ is offline   Reply With Quote
Old 19th March 2011, 18:21   #5  |  Link
CruNcher
Registered User
 
CruNcher's Avatar
 
Join Date: Apr 2002
Location: Germany
Posts: 4,949
I know but i would prefer AMV (YV12) instead of UT (YV12) for this based on my own tests i conducted (on Sandy Bridge)
anyway doing it via the GPU seems the best way and i have a test framework running based on Nvidia currently
But doing it via Quicksync could be even much better in terms of Power Consumption and efficiency though all of those would be H.264 based
and Quicksync would only work on Vista/7 and im not a fan of this restriction my current Nvidia Framework runs on Vista/7 as well as XP same efficiency as Google or Opera have in front of Microsoft and Mozilla when it comes to 2D canvas acceleration on XP
Which funnily my first Video will be about done with the GPU Framework, though i need to test how the impact on the rendering speed is first and how Nvidias driver manages this GPU Multithread wise im not sure i gonna reach the same framerate as the GPU canvas rendering for now and all this under XP which GUI never was Designed for the GPU is a real adventure

Here is a example i did for showing a bug to the developer of Realtemp (rtcore.dll Rivatuner Plugin), though this was done still via Software x264 (Capture & Encode CPU) in High Motion switching to Video Playback 30 fps fast (VBR) this doesn't survive the frame drop problem (though i have to say it was non vbv restricted, so it heavy spiked causing drops on the capture side) http://mirror05.x264.nl/CruNcher/rtcorebug/

Just had a look @ Expression Encoder 4 again and didn't realized that a SP1 was out which brings in GPU Encoding via MainConcepts Cuda Encoder also found this in the Documentation

Quote:
6. Output quality differences between GPU-assisted encoding and pure CPU-based encoding should be expected since they are implemented very differently. For example, the GPU output doesn’t blur as much, and thus is harder to encode, which in turns can introduce more compression artifacts if the bitrate isn’t adequate.
And im really surprised by this :P i guess many here will know why but it seems someone mixed something up here @ Microsofts Expression Encoder Team about the efficiency of MainConcepts CPU Encoder and (early) Cuda Encoder still from December last year and what perceptible differences those have in General, saying that the GPU Encoder blurs less then the CPU is against the law of physics unless your CPU Encoder is majorly bad in Psy which MainConcepts isn't it has AQ like x264 (complexity based) (and we aren't in the age when B-frames introduced heavy blur anymore either, so what is written their is somehow bending stuff,at least stota PSY wise)
Though if the Microsoft Guy who wrote that really experienced this then something is majorly wrong with the Expression Encoder Mainconcept Preset tuning
Which is surprising as Microsoft did a good job Psy tuning their own VC-1 but implementing 3rd party stuff seems to be different for them ,especially as they implemented one of the Best CPU H.264 Encoders and say it Blurs more then the GPU counterpart that's really strange

PS: I might have to take back that the Titanium Codec is actually not suited for Capturing High Motion , seems its CBR only and so you need to set a high enough bitrate or it looses frames, it looks ok now also cpu utilization wasn't that high worth a deeper look .
__________________
all my compares are riddles so please try to decipher them yourselves :)

It is about Time

Join the Revolution NOW before it is to Late !

http://forum.doom9.org/showthread.php?t=168004

Last edited by CruNcher; 20th March 2011 at 17:04.
CruNcher is offline   Reply With Quote
Reply

Tags
avchd, vc-1, wmv, wmv converter, wmv9

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 23:22.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2019, vBulletin Solutions Inc.