Can not See screen saver
Anonymous
1 May 2005 14:50:06 UTC
Topic 12942
(moderation:
> Da ich hier 3 Rechner laufen habe, 2xATI und 1xnVidea, meine ich, dass die Programmierer von
> Einstein nicht optimal gearbeitet haben. Oder die ATI-Karten arbeiten nicht nach dem Standart.
> Trotzdem sollte Einstein das Prgm. auch an den meist verbauten Grafik-Karten anpassen.
Letztlich benutzen wir nur die GLUT, in diesem Fall von Windows, die auf OpenGL aufsetzt (wobei wir die OpenGL-Ebene nicht mal anfassen). Anscheinend haben bestimmte, wohl eher neuere, ATI-Treiber Probleme mit bestimmten OpenGL-Befehlen, wahrscheinlich zugunsten von DirectX. Kompetente Hilfe bei der Loesung des Problems ist gerne willkommen, wir haben z.Zt. alle Haende voll zu tun und keine Zeit, um die Bugs von ATI herumzuprogrammieren. Wenn sich da drausen jemand darum kuemmern kann und moechte, moege er sich im Forum oder per eMail an mich wenden.
BM
Can not See screen saver
)
Do you guys care to translate that please?
There is quite some discussion about the Einstein App having problems with some ATI cards (actually drivers, I think). Apparently these problems doesn't occur with NVidia cards or the SETI graphics (with the same client).
"K1100LTSE" suggested that we should adapt the Einstein App to work with the ATI cards, even if they might behave non-standard in certain aspects. I replied that we are currently only using the standard GLUT (of Windows, in this case) and don't have time to fiddle around with certain bugs certain driver versions of certain manufactures might show with some of their graphics cards. [Edit: In addition to that none of the developers by now has observed this problem on an own machine, so it's quite hard to get hold of it. I have two machines running ATI graphics cards and don't see it]
But any help on this would be very welcome, so if there is a skilled graphics programmer out there who wants to hack the GLUT (or actually, I'm afraid, the OpenGL underneath it) to make it work even with these setups (without, of couse, breaking all the other possible ones which do work now), please show up in the forum or drop me a line. It might even help to get a list of things (functions or sequences) to avoid on the GLUT or OpenGL level.
[Edit: I think if there is such a person with enough insight into the ATI drivers he would probably work for ATI and be better off fixing the problems in the driver than work around it in the GLUT or somewhere else]
BM
BM
> but is the SETI ss so much
)
> but is the SETI ss so much different from the Einstein and CPDN ss' that the
> former will work but the latter won't?
Oh, you see this with CPDN, too? That's news to me. Good to know.
I think it is some sort of run condition or memory allocation problem in the driver that is triggered by a combination of certain driver versions, memory configurations (on the card), and sequences of OpenGL commands. Apparently there are some people who also see this problem with SETI, and some don't, I didn't see it at all on my machines wit ATI cards, some strange workarounds do work for some people, but not for others...
I don't know much of CPDN, but I know that E@H and SETI are using just plain vanilla OpenGL (with the GLUT).
BM
BM