|
|
#1 | |
|
Registered User
Join Date: Oct 2010
Posts: 20
|
Is support for overclocking even planned?
|
|
|
|
|
|
|
#2 | |
|
Registered User
Join Date: May 2008
Posts: 199
|
Maybe if you ask nicely you will be answered...
|
|
|
|
|
|
|
#3 |
|
Registered User
Join Date: Dec 2005
Location: My own little world
Posts: 117
|
|
|
|
|
|
|
#4 | |
|
Registered User
Join Date: Apr 2003
Posts: 17
|
Hopefully soon.
Right now my 7600 GS card without overclocking on the 260.19.21 driver runs about as fast as a 5600 LE on a F11 system with the appropriate drivers. While 2D speed has improved (via gtkperf), GL performance has really become painful. I'll run glxgears on 256.53 later on today (without overclocking) for comparison. |
|
|
|
|
|
|
#5 | |
|
Registered Fox
Join Date: Oct 2007
Location: Confoederatio Helvetica
Posts: 530
|
Quote:
Jokes aside: if you want to compare 3d performance you could use the phoronix test suite or something similar, glxgear results are not suited for any kind of comparision (google for glxgears is not a benchmark) Fuchs |
|
|
|
|
|
|
#6 | |
|
Registered User
Join Date: May 2008
Posts: 199
|
Well it is in some circumstances good for comparisons. But not a comprehensive opengl test for sure.
I remember reading here in the forums somewhere that even nvidia devs are using it sometimes... |
|
|
|
|
|
|
#7 |
|
Registered Fox
Join Date: Oct 2007
Location: Confoederatio Helvetica
Posts: 530
|
No.
The one and only reason for glxgears is to see whether you have any 3D acceleration. The results should never be used for comparisons of any kind. They depend on a whole lot of stuff aside the GPU/Driver and the tests are not even near to real live situations. http://wiki.cchtml.com/index.php/Glx...ot_a_Benchmark second hit when googling only for "glxgears", not even mentioning something like not benchmark. Fuchs |
|
|
|
|
|
#8 |
|
Registered User
Join Date: May 2008
Posts: 199
|
Whatever. For me every time was spot on (and i know it is not a true benchmark) - if glxgears score was lower, all opengls apps i used were slower or did not even work (on Intel, Ati and nvidia).
That is why i said it is a good comparison in some circumstances. |
|
|
|
|
|
#9 |
|
Registered User
Join Date: Apr 2003
Posts: 17
|
Agreed. I'm looking at the SPECviewperf benchmarks right now. I'll look at the Phoronix suite in a bit.
I just use glxgears as a point in seeing whether or not I'm being impacted by changes. Hard benchmark numbers using this are not to be trusted. It looks like either Lightsmark or SPECviewperf are the weapons of choice. |
|
|
|
|
|
#10 |
|
Registered User
|
gradinaruvasile
With GF 7900 and 17x drivers I've got ~20k fps in glxgears (without composite) Now I've got ~5k without on GF 470GTX. But it doesn't mean that GF 7900 better then 470GTX ![]()
__________________
OS: Gentoo Linux, ~amd64. VideoCard: GeForce GTX 470. ASUS P8P67 Deluxe, 2x4GB DDR3-1600 CL9, Core i7 2600k. |
|
|
|
|
|
#11 |
|
Registered User
Join Date: Jun 2006
Posts: 681
|
Phoronix benchmark does too much, you can simply measure performance by running any Unigine demo or/and Lightsmark 2008 demo or/and UT2004.
|
|
|
|
|
|
#12 |
|
Registered User
Join Date: Oct 2010
Posts: 20
|
At least we can still flash the BIOS...
![]() ![]() ![]() |
|
|
|
![]() |
| Thread Tools | |
|
|