Thoughts on Silver Arrow?

Speed said:
That is full load temps if you look.
Click to expand...

To be fair RealTemp is displaying 0.0% load (but also 0.0MHz) and Prime95 says execution halted 1-2 minutes after the latest times displayed in 3 of the threads and the other one is also showing as stopped.

That said RealTemp and HWMonitor seem to show max temps ranging from 50ºC to 57ºC across the cores.
 
Last edited:
Speed said:
Looks fine, especially in this weather, now clock that 2500K! :D



That is full load temps if you look.



It is instock at another major retailer, so if OcUK purchased from the same production run it should be soon. Webnote them and ask, or get it from the other retailer.
Click to expand...

Real temp says load 0.0 or did i miss the load some other place lol.
 
So I just need to do a blend test on prime95 then? as that's what I did on the screenshot.
 
GoogalyMoogaly said:
To be fair Core Temp is displaying 0.0% load (but also 0.0MHz) and Prime95 says execution halted 1-2 minutes after the latest times displayed in 3 of the threads and the other one is also showing as stopped.

That said CoreTemp and HWMonitor seem to show max temps ranging from 50ºC to 57ºC across the cores.
Click to expand...

True, but its clear Prime95 has just ran and the max temps are indicative of what you'd expect from a SA in this weather on a 2500K running stock speeds.

chippy08 said:
Real temp says load 0.0 or did i miss the load some other place lol.
Click to expand...

Just because it isn't loaded at the time of the screenshot doesn't mean those max temps are wrong.


maj said:
I did try to overclock, but I just got a BSOD :( might give it another go sometime.
Click to expand...

It is well worth doing, 4.5GHz is easily achieveable, especially on a Silver Arrow. You should see huge gains in most games depending on the GPUs and title in question.

BSOD happen when it isn't 100% stable so you just have to test it more throughally. A tip I've picked up through many hours of overclocking is to use both Prime 95 Blend and LinX all memory in short bursts. Theres no point testing it with 8 hours of Prime 95 if LinX causes a BSOD in 30 minutes and I've seen it happen. They seem to cause BSODs and other crashes in different ways.

What I do is set a goal of the clock speed you want to run at and work towards it. Use short bursts of LinX All Memory and then Prime 95 Blend. Say 1 hour each, if it passes both one after another then its likely stable, you can dial down the voltage. Once you get to a point that it is unstable, up the voltage up a few increments and test for longer runs, say 3 hours of each. Then if it is stable up it again until you are happy. 8 Hours Prime 95 is the accepted standard but I'd add 100 runs of LinX All Memory in there as well as some gaming or 3DMark Vantage/11 Loops as well.

There is a good guide here as well if you get stuck on some terminology or methodology: http://www.clunk.org.uk/forums/overclocking/39184-p67-sandy-bridge-overclocking-guide-beginners.html

You can also use the BSOD codes to aid you: http://www.overclock.net/general-pr...0091-complete-list-bsod-codes-when-ocing.html. You may have to disable auto restart however or use the popup error message / event logs on the restart to see the code.
 
chippy08 said:
Yes but you did sceen shot after you stopped prime95 so your temps fell.
Click to expand...

Very confused.... It records the max temps, therefore it doesn't matter that he stopped it before taking the screen shot. :confused:

Not to mention the differences in room temps, CPU vCore, cases...without having both coolers in the same setup any comparisons you make will be inaccurate.
 
Speed said:
Very confused.... It records the max temps, therefore it doesn't matter that he stopped it before taking the screen shot. :confused:

Not to mention the differences in room temps, CPU vCore, cases...without having both coolers in the same setup any comparisons you make will be inaccurate.
Click to expand...

Also how long he ran the test for as temps normally rise the longer you run the test until it reaches its highest temp.
I know its not like for like also.
 
Speed said:
Very confused.... It records the max temps, therefore it doesn't matter that he stopped it before taking the screen shot. :confused:

Not to mention the differences in room temps, CPU vCore, cases...without having both coolers in the same setup any comparisons you make will be inaccurate.
Click to expand...

Also a bit confused. If you look at the times the Maximums were recorded only one of those times falls during the period in which the displayed Prime95 run was performed. So something pushed them more than the displayed Prime95 run.

As you say though, doesn't make too much difference as lots of other factors can play a part also which will make it hard to compare temps accurately.
However if IBT was used to get the other 3 cores to their maximums it'll be less relevant as a comparison to someone running Prime95.
 
Need to see one of these running on a 2600k HT on with same Vcore and clocks as mine really for me to make a comparison, as the 2600k will run hotter than the 2500k.Or I try mine with HT off.
 
chippy08 said:
Also how long he ran the test for as temps normally rise the longer you run the test until it reaches its highest temp.
I know its not like for like also.
Click to expand...

Max temps are reached pretty quickly, once they stabilise you should be able to make some rough comparisons. He ran it for 40 minutes by the looks of things, which is enough time imho.

But as you know there are too many variables, so you can't really use this info for direct comparison.
 
geeman1979 said:
Need to see one of these running on a 2600k HT on with same Vcore and clocks as mine really for me to make a comparison, as the 2600k will run hotter than the 2500k.Or I try mine with HT off.
Click to expand...

What cooler do you have? If my new PSU turns up tomorrow I might be able to do some testing for you. Just let me know the info.
 
You must log in or register to reply here.
Back
Top Bottom