Discussion in 'General Hardware' started by newtekie1, May 19, 2009.
15th Warlock | Intel | I7 3930K@5Ghz | 104.6134 | 20.92268 | 64-bit |
Like I just said, AVX. Your scores are exactly what you'd get without AVX.
I'm running W7 SP1 and the latest IBT (2.52). Is there anything else I should configure to enable AVX?
How can I configure the test for AVX?
Just download the latest IBT - it supports AVX (properly this time; older versions either didn't support it, or did so in a flakey way)
EDIT: current version of IBT is 2.52
If you have SP1 and the latest IBT AVX should work by default. If it's still not working I can only think to blame either windows or IBT. I know in the 4.5 GHz range I get 60s without, in the 100s with, which matches up with these numbers being posted. How long ago did everyone getting these low scores grab their copy of IBT? Maybe someone messed up and uploaded it with outdated libraries.
Not sure why. All I do know is that even Asus stated to not rely on IBT for OC stability testing, due to nil or flakey support for AVX.
Without a system change, going from IBT 2.50 to 2.52, has seen my results change from 48GFlops, to 91+
I'm guessing it was just flakey on my system
2.51 and 2.52 are AVX aware
As far as the downloading, I got all mine fresh within the last 3-4 weeks along with a fresh W7 install.
I also thought the CPU could be throttling, but my stock runs are lower than my OC'd runs, so I have to assume I dont throttle stock as well.
That is low, seriously.
I get 98 Gflops on a stock 2600k. You should be close to 160-170 with such high OC and 2 extra cores
Yes, but that's using AVX instructions, I made this run without AVX, and right now it's higher than a dual Xeon system without AVX try your 2600K without AVX and you'll see what I'm talking about (I also have a 2600K)
I will run the test later after dinner
TRWOV | Intel | Core i5-2320 3.0Ghz | 080.498875 | 026.8329583 | 64-bit | AVX
TRWOV | Intel | Core i5-2320 3.0Ghz | 044.105125 | 014.7017083 | 64-bit | No AVX
TRWOV | Intel | Core 2 Quad Q6700 2.66Ghz | 023.7305 | 008.92124 | 64-bit | Conroe865PE board with DDR1 RAM
Chicken Patty | Intel | Core i7 2600K 4.5 GHz | 093.7511 | 20.8 | 64-Bit | Hyper Threading Enabled / bunch of crap running in the background
Lost Hatter | AMD | Phenom II x4 925 3.24Ghz | 041.3233 | 010.4574 | 64-bit | AMD770/rx780
Vnhill1981 / Intel / i5 2500 (Non K) @ 3.95GHz / 101.0784 / 025.5895 / 64-bit / Average over 10 tests, Multiplier set at x41 in BIOS, comes in at x38 during test
i had a few friends run it and both their computers broke. LOL
I'll try to update this more regularly, at least once a month...
Here's another run of my I5-2400, my 2400 scores pretty well Who needs a 2500k
i5 750 | 3420Mhz | 19 x 180 | 4C | 1,11875v | Asus P7H55-M/USB3 | Air | Scythe Rasetsu | 65c
TRWOV | AMD | Turion 64 X2 Mobile TL-62 2.1Ghz | 006.4612 | 003.07677 | 64-bit | hp pavilion dv2500 SE
i5 750 | 3600Mhz | 18 x 200 | 4C | Vcore 1,16875v | VTT 1.3V | CPU PLL 1.85v | Asus P7H55-M/USB3 | Air | Scythe Rasetsu | 68c
IBT has become my favorite stability testing tool.
Working on over clocking my i5-2500k. I'm IBT stable at 4.8ghz right now.
Here are my results.
Got 1.432v on my 2500k for 4.8ghz.
Completely Bonkers|Intel|E5420@2.5Mhz|058.14| 023.25 | 32-bit | E5420+L5420, stock, W2K3 sandboxed
Separate names with a comma.