stress testing - orthos and prime give different full load temps

latercomer2 · 5 Mar 2009 at 11:28

I'm currently in the process of overclocking for the first time and I've got a question about the different stress tests.

I was using Orthos when simply ramping up the FSB and it was hitting 60degree on full load. Following the basic sticky, I'm now attempting to find the max FSB by droppin the FSB and ram multipliers and using Prime Large FFTs as directed. But the max temps I'm getting are about 5 degrees lower than when using Orthos. I realise its a slightly different test but as they both run the CPUs at 100% I wouldn't expect as big a difference in temps.

Is this normal? Any advice?

JTrickle · 6 Mar 2009 at 13:19

Yes, its normal.

I've found prime to be useful for testing memory and fsb. Useful as I dont think any of my previous AMD SA/939 builds ever had matched memory or even same stick sizes.

latercomer2 · 6 Mar 2009 at 13:57

Thanks for the reply. I've been doing a fair bit more testing over the last couple of days and tend to use Prime first as it seems to cause problems far quicker than Orthos. Still once I get close to settling on the final settings I'll probably use both just to make sure.

Diggsy · 6 Mar 2009 at 20:47

I don't know why it states that Large FFT's create maximum heat in Prime, as I have always found that Small FFT's will heat it up a lot more than Large.

Give Small FFT's a shot and see what temps you get. I'd bet they will be a lot higher than Large.

latercomer2 · 6 Mar 2009 at 21:20

I've had to back off the clock as it was was hitting around 68-70 when running the small FFTs. The small FFTs definitely runs hotter but then as its CPU intensive thats no great surprise. I'm going to have to see about getting a better cooler so I can up the Vcore and see how it goes.

Diggsy · 6 Mar 2009 at 21:28

What CPU speed was that? 68-70 for a Quad is not too bad at all with Small FFT's. There is pretty much nothing else that you will run that will ever create that much heat.

latercomer2 · 6 Mar 2009 at 23:39

It was ~3.5Ghz but its only an E5200 not a quad. Still I'm fairly happy with the results so far given its a first attempt and on stock cooling in my old case

Much more experimenting coming up in the near future me thinks