F5F Stay Refreshed Power Users Overclocking Beginner in Overclocking. Some Questions

Beginner in Overclocking. Some Questions

Beginner in Overclocking. Some Questions

Pages (2): Previous 1 2
S
Sim_Crafter
Junior Member
7
08-27-2020, 11:58 AM
#11
Prime95 smallFFT is a benchmark, not a reliable foundation for other tests. It mainly checks core count and L3 size differences. RAM isn't important here. The test is too simple to assess CPU or overall system stability.
S
Sim_Crafter
08-27-2020, 11:58 AM #11

Prime95 smallFFT is a benchmark, not a reliable foundation for other tests. It mainly checks core count and L3 size differences. RAM isn't important here. The test is too simple to assess CPU or overall system stability.

T
71
08-31-2020, 02:49 PM
#12
Could try Asus Realbench
There is a test that runs 8hrs and uses half your ram.
RealBench v2.43 - New Version Available Now!
RealBench v2.43 - New Version Available Now! - Republic of Gamers
rog.asus.com
T
TheRealNoob123
08-31-2020, 02:49 PM #12

Could try Asus Realbench
There is a test that runs 8hrs and uses half your ram.
RealBench v2.43 - New Version Available Now!
RealBench v2.43 - New Version Available Now! - Republic of Gamers
rog.asus.com

I
iFitchy
Member
53
09-01-2020, 08:24 AM
#13
Half your RAM isn't sufficient, and eight hours seems excessive. To properly assess stability, you should perform a HCI Memtest at least up to 400%, using Testmem5 with the Anta777 configuration. Other options include Karhu Ramtester, though I haven't used them yet. The quicker you reach the point of failure, the better. You don't want to wait eight hours just to discover instability; aim for a final stability check in 1 to 3 hours at most. Make sure to test every RAM stick and all its channels—otherwise it's like running your CPU at half speed and calling it stable.
I
iFitchy
09-01-2020, 08:24 AM #13

Half your RAM isn't sufficient, and eight hours seems excessive. To properly assess stability, you should perform a HCI Memtest at least up to 400%, using Testmem5 with the Anta777 configuration. Other options include Karhu Ramtester, though I haven't used them yet. The quicker you reach the point of failure, the better. You don't want to wait eight hours just to discover instability; aim for a final stability check in 1 to 3 hours at most. Make sure to test every RAM stick and all its channels—otherwise it's like running your CPU at half speed and calling it stable.

S
Siphan0
Member
74
09-05-2020, 07:24 PM
#14
Sorry, I was only using Cinebench for a brief test. I've run stability checks (OCCT memory and CPU for an hour), Aida64 for an hour, and Prime95 for an hour. Everything was successful. I'll attempt some of your recommendations next.
S
Siphan0
09-05-2020, 07:24 PM #14

Sorry, I was only using Cinebench for a brief test. I've run stability checks (OCCT memory and CPU for an hour), Aida64 for an hour, and Prime95 for an hour. Everything was successful. I'll attempt some of your recommendations next.

E
EmmaForLife
Member
201
09-11-2020, 02:21 PM
#15
I obtained MemTestPro and executed the test at 200%. I also performed the OCCT memory test and CPU tests (Large set, Variable load type, Extreme Mode) for two hours each. No errors have occurred so far. My games and general work function smoothly. I experienced two unexplained reboots at night (bug check reboot) and once during a restart where I got a blue screen message that eventually resolved. I think the issue might be connected to curve optimization because the problems appear during idle periods. My plan is to run MemTestPro at 400% first, without curve optimization, to assess the situation before increasing the test to 200%. So far, I've made three adjustments: 1) PBO with auto overclocking (175 MHz), 2) Curve Optimization per core, and 3) DOCP (3600), fCLK at 1800, DRAM V increased from 1.35 to 1.36. Does addressing the CO first seem sensible?
E
EmmaForLife
09-11-2020, 02:21 PM #15

I obtained MemTestPro and executed the test at 200%. I also performed the OCCT memory test and CPU tests (Large set, Variable load type, Extreme Mode) for two hours each. No errors have occurred so far. My games and general work function smoothly. I experienced two unexplained reboots at night (bug check reboot) and once during a restart where I got a blue screen message that eventually resolved. I think the issue might be connected to curve optimization because the problems appear during idle periods. My plan is to run MemTestPro at 400% first, without curve optimization, to assess the situation before increasing the test to 200%. So far, I've made three adjustments: 1) PBO with auto overclocking (175 MHz), 2) Curve Optimization per core, and 3) DOCP (3600), fCLK at 1800, DRAM V increased from 1.35 to 1.36. Does addressing the CO first seem sensible?

N
Nishaa
Junior Member
5
09-23-2020, 04:28 AM
#16
Does fixing the CO first appear logical? Yes, these are the signs. Stalling at idle. Definitely address that first.
N
Nishaa
09-23-2020, 04:28 AM #16

Does fixing the CO first appear logical? Yes, these are the signs. Stalling at idle. Definitely address that first.

Pages (2): Previous 1 2