Also, as a bounty platform, may I suggest bountysource.com, I've had good experience with it for free software projects. We can just point a github project and add bounties there.
Thank you for the offer, i hope we can work something out
btw, you don't need to get hardware for AmigaOS4.1 (at least you are not forced to). As soon as AmigaOS4.1 FE becomes available you can set up a working installation with your favourite UAE emulator
btw, you don't need to get hardware for AmigaOS4.1 (at least you are not forced to). As soon as AmigaOS4.1 FE becomes available you can set up a working installation with your favourite UAE emulator
Qemu does not emulate altivec, thus *UAE does not emulate altivec.
Unless someone adds that capability to Qemu...
After saying that I do see Altivec patch messages, but I said it based on TODOs showing up in Altivec instructions group in the source tarball.
in Qemu 2.1.50 file ./target-ppc/STATUS there are a lot of Quote:
INSN KO Altivec missing
And this is also the case after updating to 2.1.91
Edited by billt on 2014/11/15 2:31:09 Edited by billt on 2014/11/15 2:58:38 Edited by billt on 2014/11/15 3:01:54
... Anyway, regarding the Altivec work, I can definitely say that I would do a substantial discount if only because it's one way to help my first computer love!
Don't hesitate to contact me for a quote, and I also mention in my blog that I offer free evaluation -what needs to be optimized, how long it would take, etc. Please read:
Nice to see you here. It would definitely be nice to get a quote from you, or at least an opinion as to how much scope there is for improvement. You'll find a link to the ffmpeg PowerPC code in the first post in this thread. We're mainly interested in the H.264 at this stage, given its popularity and how computationally intensive it can be (depending on settings).
~Yes I am a Kiwi, No, I did not appear as an extra in 'Lord of the Rings'~ 1x AmigaOne X5000 2.0GHz 2gM RadeonR9280X AOS4.x 3x AmigaOne X1000 1.8GHz 2gM RadeonHD7970 AOS4.x
I was away for a few days for a business trip, I will look at the code for a few days and will let you know how long it would take to bring it in parallel to eg. arm or x86 levels, at least. Would that interest you? Should I post the quote here in public, or would you prefer it in private first? At the very latest I expect by Monday to have a clear idea of what has to be done.
@feanor Please make it public. We can create a bounty for it,so they make their money. The goal must be full h.264 playback and that your work will be implement into the RadeonHD2.4 driver without additional charge.
The goal must be full h.264 playback and that your work will be implement into the RadeonHD2.4 driver without additional charge.
Huh? The code to be worked on has nothing to do with the Radeon HD driver. It's code inside mplayer itself, and the optimizations would benefit all altivec machines.
That's a bit vague. What's the base hardware and what kind of h264 video would you consider as minimum for playback? Eg. even with full altivec, I guess getting 1080p 60fps streams on a G4@800Mhz is going to be a bit over-optimistic to expect. Plus, getting the 90% of the optimizations is going to be probably "easy", but if one wants to get the full deal, it's not uncommon to spend exponentially more time to get the rest done, most likely because you have to resort to heavily tuned asm, instead of C intrinsics, etc.
What is the expected outcome and what would make people accept such a challenge as solved?
Eg. pick a video stream (bigbuckbunny?) and a base hardware spec and expect that to work with no frames dropped?
I don't have an X1000 so I can't really benchmark, I need a lower base of reference. I have a G4@1Ghz so I can use that, and I have access to some Power7/8 VMs with VSX, but that's a different beast.
I think Hans should do the outlining and final aim, as he's got the idea and has the most experience due to his gfx driver development
Quote:
I don't have an X1000 so I can't really benchmark, I need a lower base of reference. I have a G4@1Ghz so I can use that, and I have access to some Power7/8 VMs with VSX, but that's a different beast.
Actually, whatever you can squeeze out of that G4 of yours will make the faster systems profit. (A G4@1GHz is close enough to an AmigaOneXE G4/800MHz, which is one of the first Amiga NG systems there were... ...to steal from "Ol Blue Eyes": "If you can make it usable there, you'll make it usable everywhere)
Is the altivec unit the same over all ppc generations? I mean, do newer altivec units have more means/possibilities to be sped up?
Is the altivec unit the same over all ppc generations? I mean, do newer altivec units have more means/possibilities to be sped up?
Well, yes and no, it's mostly the same, but some things are different between eg. a G4 (aka MPC74xx) and a G5 (aka PPC970), in general the altivec unit in the G4-class is more powerful, but the advantages of the G5 (clock speed, better memory bandwidth, etc) eliminate that. Differences are bigger in newer cores like the e6500 or in recent IBM cores like the Power6 onwards.
But we don't care about these cases, AFAIU, focus is on 32-bit big endian altivec, but I'll make it 64-bit safe ofc, and if possible, be/le agnostic so that it can run on a 64-bit LE OS as well, such as ppc64le. That's free of charge. I'll only charge for the 32-bit BE initial work.