Strongly annoyed about those closed-Micro$oft
WMV Codecs, that do not work in MPlayer?

I have seen ppc-users using the x86-Mplayer with win-dlls
via qemu. But I never did this effort, rather I would have
bootet osx on such occasions. (My last personal reason
for osx: gone now! )

There are reference code and very detailed instructions
online on how to join vc1_reference code with ffmpeg.
Updated On Feb-16 it now works with current ffmpeg-cvs.
Converting of those WMVs is no problem anymore!
Look here: http://multimedia.cx/eggs/?p=129

Don't forget to do step 12.5 in the footnote.

Converting works nice for me, although quite slow.

seeing as zambelli VC-1 ms coder posted the new thread
http://forum.doom9.org/showthread.php?t=117957
about the first non beta wmv9/vc-1, i dont know if it would work with one of the ways you write about but it says this
"“In case you’re wondering what has changed in the WMV3/WVC1 encoder since Beta 2, it’s fixes for some crashing bugs, VC-1 conformance bugs, interlaced encoding support, and performance improvements for higher complexity modes. The most interesting one is the interlacing fix: Using “VideoType” registry setting now correctly switches between progressive, frame interlaced, field interlaced, and mixed content encoding, and produces VC-1 compliant content - which was not the case in earlier betas.”

of course if BetaBoy can be convinced to port their full family of codecs/apps to the platform then playing AVC, VC-1 and others wont be a problem perhaps.
http://forum.doom9.org/showthread.php?p ... post876956

while im here and dont know what thread to bump so ill just add about NVIDIA rethinks the GPU with the new GeForce 8800
http://arstechnica.com/news.ars/post/20061108-8182.html
"Ideally, a program for the G80 would consists of hundreds of stream processing threads running simultaneously on the GPU's many arrays of tiny, scalar stream processors. These threads could do anything from graphics and physics calculations to medical imaging or data visualization.

Making a fully generalizable stream processor like the G80 required NVIDIA to include a feature that graphics programmers have desired for as long as they've wanted a unified shader model: a hardware virtual memory implementation for the GPU that enables seamless access to main memory for GPU-based programs."


i dont think we (the whole collective PPC)can yet convince them to OSS a driver for our use but the hardwares moving on and someone here might find a way.......

funny how that gpu reminds me of the KiloCORE description......

I don't think it is very prudent to go canvassing on forums like that.

If a developer wants to do something, usually he will go and do it. The problem is, if he is not, it is usually because he is not motivated and/or there is no time, or he just doesn't have the required skills.

We have proved out, by giving away hundreds of systems for free, that when a guy says "I can do that", something can always pop up and take away the motivation or the time needed. Children, financial problems, new job, old job, legal reasons. Pick one, any one, or maybe more than one.

Therefore it is not up to "some user on a forum" to present them with new work. What motivation does he have to support us? What motivation to port to Linux, support Power Architecture or maintain for Pegasos?

Because you asked?

Links to free hardware is not enough, there needs to be a want to put some effort behind it. I don't see that this is evident with these guys.

The basic moral of it is, if there was a developer out there who WANTED to do this, they would ALREADY be doing it, or they would most certainly have bugged me or Bill or Gerald or Thomas or anyone involved in the company about it. Or, they would be on these forums asking and they would be handled in the proper way.

_________________
Matt Sealey

of course,im fully aware of how you think it should work Matt, you have stated it enough over the years.

*snip*

Core stuff has really few stuff open source...

Still we consider it if is reported as faster as we considered apple's implementation before.

Please notice that h264 standard requires you to do some computational step that have very slight results in the decoding quality but that cost a bit (on ffmpeg you could cut them with a flag but then you won't have a compliant decoding), most of the times we got reports about blazing fast decoders it ends up in taking shortcuts that lead to non-compliance.

lu - that should at least complete and commit the weighted window stuff for altivec...

PS: the ffvc-1 has already some altivec optimizations and works pretty fine

Now, see here, 'popper'.

There is always a way to get a developer to do something and there are always a hundred ways to discourage them completely.

Posting on forums, thread after thread, that "you should port to this platform" is one of the hundred ways to discourage them. It is NOT your job to "promise" "boards from BBRV" to other developers on forums.

To delve deeper into this situation, mailing Bill to tell him that you posted a post on a forum telling a developer he should port to something else just for you, what was that for? Are you looking for a Gold Star? Kiss on the cheek?

If you disagree with how I interpret the best way to deal with developers and encourage them to our platform - then fine. You can disagree.

Any more "wah wah Neko is crap at his job" posts merely because you disagree, however, and I'll ban you outright. So far I did not see you contribute a single useful post here so I don't think anyone will miss you.

You may think it is draconian of me, but I view it as simply a button click.

~

Let us analyse the response to your request on Doom9. There have been.. zero responses to it on that forum, in this forum, in my inbox or in Bill's. Have you been contacted? I dunno, how shall I proceed to get this guy a board if he wants to do his port? Through you!?

How do you - at the end of the day - expect a guy who is concentrating on *DIRECTSHOW CODEC SUPPORT FOR WINDOWS* to even take your request seriously? You really just did go and stumble into a room, bug them with a question and then start advertising. It did nothing. Do you understand what image issue this creates for Genesi, when we have a hundred users bugging thousands of developers to sign up for a free board just because they hanker for some software? I don't think the overseers of a rabid community is what we want to be.

Why don't you pick up a compiler and do some coding yourself? Become a developer instead of a telemarketer.

_________________
Matt Sealey