AMD Athlon 5350 / 5150 & Sempron 3850 / 2650

[mczak]

So these results are way better than the earlier ones for the 5350 and much more like they should be. So what was the problem with the earlier numbers? While I understand the issues with DPM for the gpu tests, the cpu results also look like the cores were stuck at 800Mhz or something when comparing the numbers (in the multithreaded tests the "old" 5350 ends up roughly where the sempron 2650 now does which would support that theory). Was that a testing error or do you really need 3.14 kernel for it to work?

[mendieta]

I would love to see, in these kind of tests, other integrated GPU's, perhaps a Haswell, as a benchmark. Cheers!

[Veto]

Thanks for the test!
It would have been nice with idle power draw numbers as well. If used for a NAS 24/7 the CPU will be idle most of the time.

[AJSB]

Unlikely, too time consuming, etc, unless there were requests from many Phoronix Premium readers for that... besides the overclocking is limited and just getting about an extra 100MHz if lucky.

No problem, thaks Michael, i know i was asking too much
...but a GTX750(Ti) is/was tested, right ?

As for OC, it seems that the 5350 can go up 156MHz more in total for CPU....

[dungeon]

So these results are way better than the earlier ones for the 5350 and much more like they should be. So what was the problem with the earlier numbers? While I understand the issues with DPM for the gpu tests, the cpu results also look like the cores were stuck at 800Mhz or something when comparing the numbers (in the multithreaded tests the "old" 5350 ends up roughly where the sempron 2650 now does which would support that theory). Was that a testing error or do you really need 3.14 kernel for it to work?

From my testing with AM1M-A and Athlon 5350 with Lubuntu 14.04 and Debian Sid kernel 3.14 is not explicitly needed , so xUbuntu users can stay with default kernel and default mesa, etc... but they have to tweak few settings: radeon.dpm=1 as kernel parameter and ColorTiling together with ColorTiling2D in xorg.conf - that way performance is roughly the same like you see in this article .

I am using it right now with 3.13 kernel no problem, and performance is like you see in this article i even overclocked it to 2206 MHz and it works fine .

[dungeon]

As for the first testing article he did, i think Michael does not apply these tweaks and maybe it set something wrongly in bios - maybe silent mode/power saving preset or something similar - that could easely explain low results .

[dungeon]

From my testing with AM1M-A and Athlon 5350 with Lubuntu 14.04 and Debian Sid kernel 3.14 is not explicitly needed , so xUbuntu users can stay with default kernel and default mesa, etc... but they have to tweak few settings: radeon.dpm=1 as kernel parameter and ColorTiling together with ColorTiling2D in xorg.conf - that way performance is roughly the same like you see in this article .

I am using it right now with 3.13 kernel no problem, and performance is like you see in this article i even overclocked it to 2206 MHz and it works fine .

Of course for benchmarking games with radeon driver, one must also disable vsync features with: SwapbuffersWait in xorg.conf and vblank_mode=0 env... further one can enable hyperz to boost graphic performance i don't have issues with it ( i wanna see these issues, but i don't have it with games i use so i enable it) R600_DEBUG=hyperz env, just please be aware that many have issuses with it mainly rendering artifatcs but even gpu lockups - see the bugs .

[smitty3268]

but they have to tweak few settings: radeon.dpm=1 as kernel parameter and ColorTiling together with ColorTiling2D in xorg.conf

As for the first testing article he did, i think Michael does not apply these tweaks and maybe it set something wrongly in bios - maybe silent mode/power saving preset or something similar - that could easely explain low results .

Those settings should only affect graphics - and i'm sure they were the problem in his gpu tests. But they shouldn't have any effect on the cpu-only tests, which were also much slower than expected.

[dungeon]

Those settings should only affect graphics - and i'm sure they were the problem in his gpu tests. But they shouldn't have any effect on the cpu-only tests, which were also much slower than expected.

He tests what one Ubuntu user can expect from 14.04 but right out of box with no tweaking in that article and with that in mind that is correct i think GPU wise. CPU wise i said maybe he enable some power saving feature:



He maybe enable silent mode which triger the issue there or something, who knows Governior is also setted to performance earlier as i see, but now he testing it with ondemand which is default, etc.

[dungeon]

What i see on that picture also wrongly setuped is also that he plugged one DIMM in DIMM_A1 , but manual explictly says: Ensure to insert the DIMM into DIMM_A2 socket if you install only one memory module .