- last edited on by
ROGBot
With Maximus VII Hero I'm now trying to influence the performance of the CPU to the BIOS version. Everything takes place under the same conditions in terms of temperature measurement settings of CPU and BIOS settings. I set the processor to 4.5 GHz / 4.5 GHz core frequency and cache to 1.3V. OS was Windows 7 SP1 64-bit. Graphics just out of curiosity GTX 980T Poseidon.
For testing, I tried 3 BIOS versions. 0401, 0802, and the latest BIOS 1,001 [/ b]
I was interested in performance SuperPi 32M, CINEBENCH R11.5, 3D Mark06 CPU performance, 3D Mark11 Performance and 3D Mark Firestrike. Older tests such as SuperPi because I wanted to 🙂
I. BIOS 0401
Generally BIOS showed OK, the voltage on the CPU LLC5 perfectly kept, also stability seemed very good even at a voltage of 1.3V.
Superpi 32M = 7m 0.685s
3D Mark 06 CPU = 9 671p
3D Mark11 P = 20 360b, physics=13 314p
3D Mark Firestrike = 16 262p, physics = 14 587p
Cinebench R11.5 = 10.48p
II. BIOS 0802
The BIOS convinced me more balanced performance across all tests, but stability was unpredictable, with 1.3V in SuperPi frozed despite several attempts (I set after 1.345V). Despite a slight decrease in performance CPU 3D06 and in SuperPi - differences are acceptable, other benchmarks in turn are noticeably better in terms of score. However, the stability of pity ... It takes more stress on the same settings than the previous case. Now therefore, not this BIOS
Superpi 32M = 7m 1.278s
3D Mark 06 CPU test = 9522p
3D Mark11 Performance = 20 608p, physics = 13 312p
3D Mark Firestrike = 16 297p, physics=14 686p
Cinebench R11.5 = 10.93b
III. BIOS 1001
The BIOS me a little disappointed, I forgot to measure a lower voltage, but what Im still to testing, 1.3V holds now. This is positive. Although the behavior of PC stability and performance is not bad, I'm sorry that SuperPi is much worse and I tested it several times, even after restarting the PC with the same result within tenths of a second.
Superpi 32M = 7m 4.648s
3D Mark 06 CPU test=9484p
3D Mark11 Performance=20 655p, physics=13 334p
3D Mark Firestrike=16 286b, physics=14 656p
Cinebench R11.5 = 10.91p
There is overview table with the results.
Repeatedly measuring the current I reached there with the last BOSS 10.93 points in R11.5 and another test of SuperPi is 7m 3.478s and for 3D06 9488points per CPU. Thus, I would recommend for the average user to choose between BIOS 0401 and 1001 BIOS 0802 to avoid. On the same frequencies requires a higher voltage even when served on average the best results, at the cost of "instability" of the same strain is not worth it. If you like SuperPi :), then 0401 is your choice. Conversely, if the test elderly hot food, will be better in 1001, which is better in newer tests.
For testing, I tried 3 BIOS versions. 0401, 0802, and the latest BIOS 1,001 [/ b]
I was interested in performance SuperPi 32M, CINEBENCH R11.5, 3D Mark06 CPU performance, 3D Mark11 Performance and 3D Mark Firestrike. Older tests such as SuperPi because I wanted to 🙂
I. BIOS 0401
Generally BIOS showed OK, the voltage on the CPU LLC5 perfectly kept, also stability seemed very good even at a voltage of 1.3V.
Superpi 32M = 7m 0.685s
3D Mark 06 CPU = 9 671p
3D Mark11 P = 20 360b, physics=13 314p
3D Mark Firestrike = 16 262p, physics = 14 587p
Cinebench R11.5 = 10.48p
II. BIOS 0802
The BIOS convinced me more balanced performance across all tests, but stability was unpredictable, with 1.3V in SuperPi frozed despite several attempts (I set after 1.345V). Despite a slight decrease in performance CPU 3D06 and in SuperPi - differences are acceptable, other benchmarks in turn are noticeably better in terms of score. However, the stability of pity ... It takes more stress on the same settings than the previous case. Now therefore, not this BIOS
Superpi 32M = 7m 1.278s
3D Mark 06 CPU test = 9522p
3D Mark11 Performance = 20 608p, physics = 13 312p
3D Mark Firestrike = 16 297p, physics=14 686p
Cinebench R11.5 = 10.93b
III. BIOS 1001
The BIOS me a little disappointed, I forgot to measure a lower voltage, but what Im still to testing, 1.3V holds now. This is positive. Although the behavior of PC stability and performance is not bad, I'm sorry that SuperPi is much worse and I tested it several times, even after restarting the PC with the same result within tenths of a second.
Superpi 32M = 7m 4.648s
3D Mark 06 CPU test=9484p
3D Mark11 Performance=20 655p, physics=13 334p
3D Mark Firestrike=16 286b, physics=14 656p
Cinebench R11.5 = 10.91p
There is overview table with the results.
Repeatedly measuring the current I reached there with the last BOSS 10.93 points in R11.5 and another test of SuperPi is 7m 3.478s and for 3D06 9488points per CPU. Thus, I would recommend for the average user to choose between BIOS 0401 and 1001 BIOS 0802 to avoid. On the same frequencies requires a higher voltage even when served on average the best results, at the cost of "instability" of the same strain is not worth it. If you like SuperPi :), then 0401 is your choice. Conversely, if the test elderly hot food, will be better in 1001, which is better in newer tests.
