Sorry about the delay. The news story is posted now.... I had written it up in advance, but when hospitalized it did not work out so well for posting the story
Another "best wishes, get well soon" message here. We were hoping nothing had gone wrong -- normally we have to hustle to keep up with you
Originally Posted by rohcQaH
yeah, "eventually" every card is going to be supported. Just wondering if the RV740 has some architectural changes that need lots of additional work, or if it can be driven by the existing code with slight modifications. In other words: will RV740 owners wait longer for 3D support than RV730 owners?
We do not discuss unreleased hardware, but all discrete GPUs in the 7xx family have very similar 3D programming characteristics so I would not expect any delays in support other than getting production boards into our developers systems. Each different GPU normally needs some code tweeks and I expect future 7xx parts will be no different.
So... yes, the wait will be longer, but only because our devs already have production rv730s but don't have rv740s yet.
I'm not totally sure of the status; we mentioned it at CEBIT IIRC, and I'm sure that OEMs have them, but didn't think there were any in public other than leaks. The lead time from part availability to laptops-in-customer-hands is pretty long; usually months not weeks.
Tell you what... if it's up on amd.com I'll talk about it, OK ? Hold on a minute... OK, so looks like the mobile parts are up on amd.com :
So, yeah, the 3d engine shouldn't be hard to get running but the display blocks changed again IIRC. I expect display will take more fiddling than 3D to get running. Realistically we'll need to wait until production desktop cards are shipping but shouldn't take long after that.
I somehow assumed no one would start planning purchase decisions for something that doesn't officially exist yet. Anyway.
There are some previews of test samples around the net, and some sites mention a release date of 28.4. No idea what parts of those information are "official", "leaked" or even "plain wrong".
Was just wondering if it's worth keeping an eye on, or if it wouldn't be an option due to driver trouble anyway. Sorry for being a bit impatient.
Thank you for your reply john. If it takes until may or june until the hardware is available and 7xx 3D works, I'll wait. Feels good to have a palpable date
Finally, the code has only really been run on developer machines, ie we haven't done the "put card A in, test, fix, put card B in, test, fix, put card A back in, test, curse, fix the previous fix, put card B back in, test, put card C in, test, fix, put card D in, test, wow it works, put card E in, test, fix..." dance yet.
FWIW glxgears almost works on my HD3850 AGP, it segfaults every time at 30 seconds and there is so much debugging output that I have to redirect stderr to /dev/null, but I do see something - not quite rendered properly and it doesn't resize properly (the gears resize but do not stay within the window if reduced and if enlarged look like the shot below).
It's great to have something to test though - so thanks to all concerned.
Heh, looks like the shadows on the gears were calculated horribly wrong in all three gears. (Kinda interesting that it seems some stuff in the red gear is rendered *completely* black)
Edit: Oh, wait. Actually all three gears have some of their lower-right portions rendered completely black. Me blind.