show your sandy bridge 24/7 clocks and volts

Hello my first post, be gentle hah

Just finished my new i5 core rig and wanted to post the details, I've been building PCs and overclocking since I convinced a 486DX2-50 to run as a 486DX2-66 - that's was 33% overclock don't knock it! :p That was also my first experience of having to take the side of the case to keep it cool in the summer!!

My last rig was a [email protected] on air cooling, the new one uses an H70 (my first attempt at "sort-of-cheapy" water cooling heh.

I think I have a good chip, I'm seeing 4.6Ghz on 1.2vcore and topping out at 4.8Ghz on 1.25vcore - which is very nice! Temps are 75C under prime95 at 4.8Ghz in an annoyingly warm room (in the summer anyway). and 70C at 4.6Ghz. Can't *quite* convince the chip to do 5.0Ghz without really jumping up to naughty temps in the 85C range, so good enough for me!! I'm amazed at how little power these things use...! The Q6600 system used ~100W at 3.3Ghz... and the i5 is around x2 performance in benchmarks that I've compared them in. Super happy with the new setup!!

Cheers,

Kev
 
shkwav, are you certain those are the voltages? Have you checked they're reported as that in CPU-Z? Your temperatures don't line up with those voltages at all on that cooler. I get about 75 degrees at 4.8 on a mid range air cooler, and that is at a much higher voltage. Something doesn't sound right. If you download CPU-z and check the voltages under load and they tie up with that,AND it is 8 hours prime stable, you have a very good chip, but I'd remount your cooler. But something about your figures doesn't add up. :confused:
 
Last edited:
Ok you are correct <bows to your superior skillz> I am not used to this new MB yet, appears that it is automagically upping the vcore when under load. Cheeky monkey. CPU-z is reporting vcore of 1.28v @ 4.6ghz and 1.33v @ 4.8ghz. I am a bit worried I've messed up the paste when I fitted the cooler it was a bit of a sweaty-palms-gritted-teeth moment...!

Also above I meant to say the Q6600 chip used 100w MORE juice @ 3.3ghz than the i5 @ 4.8! What a little trouble maker!

Cheers!

Kev
 
That's still a good chip! Slightly edges mine out at 4.8GHz, I need 1.344 or worker 4 decides to fail after about 6 hours. That voltage makes your temperatures more reasonable so your cooler is probably fine. :)
 
I'm have a plug that measures the power usage that travels through it, I've plugged the whole system into it (including the monitor, but I switch that off when checking the wattage levels). The i5 system in total draws only ~150W when idle (4 SATA HDs, 1 SATA DVD drive, 8GB DDR3 and radeon 4870 in there also), and ~240W under full CPU load @ 4.8Ghz. With the graphics card maxed out also it draws ~300W total, it fluctuates a bit but you get the idea... For comparison my previous Q6600 system @ 3.3Ghz with the same graphics card and HDs etc. in it was using ~250W at idle, 350W under full CPU load and ~400W with the GPU maxed out.

Cheers,

Kev
 
How much do they cost to buy.

Fancy lending it me if i pay for delivery both ways.

Need to know if my psu will handle 3 gtx 480s Ive got two now so i could guest the power a extra one would pull
 
You can pick them up for around 12 quid in most hardware stores - I got mine from Wilkinsons, I've seen them in Robert Dyas also - definitely a tool worth having for over-clocking fun! :D
 
I'm have a plug that measures the power usage that travels through it, I've plugged the whole system into it (including the monitor, but I switch that off when checking the wattage levels). The i5 system in total draws only ~150W when idle (4 SATA HDs, 1 SATA DVD drive, 8GB DDR3 and radeon 4870 in there also), and ~240W under full CPU load @ 4.8Ghz. With the graphics card maxed out also it draws ~300W total, it fluctuates a bit but you get the idea... For comparison my previous Q6600 system @ 3.3Ghz with the same graphics card and HDs etc. in it was using ~250W at idle, 350W under full CPU load and ~400W with the GPU maxed out.

Cheers,

Kev

So the [email protected] uses about 110W less under load than a [email protected]? That doesn't sound right to me. Real world may well be different, but using the good old calculator at http://extreme.outervision.com/psucalculatorlite.jsp

2500K at 4.8/1.33 = 145W
Q6600 G0 at 3.3/1.34 = 150W
Q6600 B3 at 3.3/1.34 = 166W

Edit: link was wrong
 
Last edited:
Yep it does indeed. I am reading the values straight off the LCD panel during testing, I wrote them down when testing the Q6600 before performing the upgrade to the i5. The only other notable difference worth mentioning is that I bought a new PSU for this rig (as my old one was ATX2.2 only) - I was previously using a HIPER Type R 580W (which was definitely struggling) and now have a Corsair TX-650 which is barely turning over... Maybe the MB is taking less also now? The plug is showing the *total* draw for the complete system...

Cheers,

Kev
 
I thought it might, and I'm glad then that I went for a nice new 80% certified one - it is *so* much quieter under load also.
 
Hmm, so if you look at the efficiency curve on the corsair product page then the system load must be about 260W (40% load), which at 86% efficiency in the PSU gives you your 300W load at the wall. I was just reading up on this 80+ certification, and it's performed at an ambient temp of 23 C.

It looks like the Hiper was similar, about 82% at that load.
 
Last edited:
I had decided that both my bloomfield i7 920 @ 2.66GHz with turbo disabled and sandybridge i7 2600K @ 3.40GHz with turbo disabled as both staying on stock speed (no overclocking) as I had enough of overclocked as it isn't worth it anymore! :) Just enjoy my pc and play gaming for longest lifespan!
 
I had decided that both my bloomfield i7 920 @ 2.66GHz with turbo disabled and sandybridge i7 2600K @ 3.40GHz with turbo disabled as both staying on stock speed (no overclocking) as I had enough of overclocked as it isn't worth it anymore! :) Just enjoy my pc and play gaming for longest lifespan!

Why did you bother to buy a 2600K then ?

I Really don't understand this, almost as if the upgrade is for no apparent reason other than keeping up with the latest gizmo.......
 
Back
Top Bottom