AMD Phenom II questions

[jstokes]

has there been any testing or benchmarking of AMD new Phenom II 920 or 940 with Linux? I'm planning to build a new system in the next couple of weeks with a 970GX Motherboard Phenom II 920 or 940 ,4Gb Ram, a HD4850, and Ubuntu 8.10 I'm not sure if I'm going to do 64 bit or not, and have been looking for benchmarks to help with my decision. Also has there been any testing of ATI catalysis 9.10 with a system like this.

[Kano]

Well an Intel E6600 + Nvidia 8800 GTS 512 beats that config with Linux games like nexuiz *g*

http://global.phoronix-test-suite.co...4926-3248-8692

[deanjo]

Well an Intel E6600 + Nvidia 8800 GTS 512 beats that config with Linux games like nexuiz *g*

http://global.phoronix-test-suite.co...4926-3248-8692

That's more due to video card then processor.

[RealNC]

CPU is good. Graphics card is not.

[d2kx]

Kano's statement is based on a benchmark I made which had lower results due to a fglrx bug I ran into (very high CPU usage during the test due to EDID & modeline request spamming by the driver, causing short freezes every few seconds).

I've got a Phenom II X4 940 on a 790GX/SB750 based motherboard (MSI DKA790GX) with a HD 4850 and like my system

[deanjo]

Kano's statement is based on a benchmark I made which had lower results due to a fglrx bug I ran into (very high CPU usage during the test due to EDID & modeline request spamming by the driver, causing short freezes every few seconds).

I've got a Phenom II X4 940 on a 790GX/SB750 based motherboard (MSI DKA790GX) with a HD 4850 and like my system

Well it's still your video card that is holding you back. You proc is not your bottleneck. Compare your latest run:

http://global.phoronix-test-suite.co...745-27234-5490

vs a system with more graphics horsepower with a 9850 OC'd to 3 Ghz running at a higher res.

http://global.phoronix-test-suite.co...376-32355-7283

You can see that it's not the proc holding you back.

[Kano]

Well a benchmark with 1680x1050 would be better, not everybody has a 1920x1200 to bench against.

[deanjo]

Well a benchmark with 1680x1050 would be better, not everybody has a 1920x1200 to bench against.

Happy now? http://global.phoronix-test-suite.co...23-30010-25949

[Kano]

Well I still think that your system has a timer issue as your hd benchmarks are too high compared to others too. A 8800 GTS 512 is usually faster than a 8800 GT or did you use SLI? nexuiz runs only on one core basically so quad would not gain anything - you could test that with max_cpu=2 kernel option.

[deanjo]

Well I still think that your system has a timer issue as your hd benchmarks are too high compared to others too. A 8800 GTS 512 is usually faster than a 8800 GT or did you use SLI? nexuiz runs only on one core basically so quad would not gain anything - you could test that with max_cpu=2 kernel option.

My system is SLI (running afr), thus why I said the CPU is not the bottleneck and on a system with more graphics power gains are had, if the cpu was a bottleneck there would be no gain in SLI mode.