If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
re: VDPAU, I think that was what the existing open source decode acceleration code used so we stayed with it.
re: flash, not sure. My recollection was that flash was moving away from decode acceleration support on Linux but I haven't really looked at it. Are you talking about an older version of flash maybe ?
Latest version that still supports firefox is the worst, but chrome wasnt much better frankly. Always assumed that it was radeon's fault because my other laptop, which has an intel sandy bridge graphics card, didnt get like that. (CPU in the radeon laptop is an Intel Core series as well, but not with a graphics unit integrated -.-)
I thought I clarify some things before everybody starts to speculate and in the end blames me for faking the moon landing (or something like that).
First of all for the firmware: It actually contains relatively few code, everybody who programmed a entropy decoder before knows that you need quite a bunch of lookup tables to do so. So that's what this is mostly about, on top of that there is a bit code that controls how data is pumped between different internal blocks.
So it is NOT possible to fix large fundamental bugs or implement new codecs that the hardware doesn't support by just hacking on the firmware. For that the hardware is just to hardwired (well that's why we call it hardware, don't we?).
On the other hand it is possible that the hardware can support more codecs than the firmware currently has implemented. This happens when the firmware programmers at AMD are not finished when the hardware gets out, so it still makes sense not to burn it into some kind of a ROM block or something like this, but instead just load it into normal RAM on startup.
The good news is that I can confirm VP8 is on the todo list for future UVD hardware generations, but on that list is also practically any video decoding feature that UVD currently doesn't support, so that's not such an surprise also.
Now to the supported hardware: Well actually supporting the newest hardware generations is usually more simpler than getting the "old" ones to work, cause the greatest obstacles are hardware bugs you somehow need to work around. And while SI is pretty much bug free regarding this everything before has a couple of different things you need to worry about.
The plan is to support every new hardware generation with patches as soon as it is available, but it is the next large hardware block we are going to support besides the output engine, 3D and DMA so I can't guarantee that there wouldn't be any problems and/or delays.
Currently supported is everything with UVD2 and newer on it, with the only exception of RS780/RS880 which have two outstanding problems. Additional to that it used to work on RV770 and RV790, but a couple of people reported that it currently doesn't. I'm already investigating in it.