• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

TPU's Core i7 Overclocking and Feedback

you have to figure the time it takes fr the air to pass from one to another wouldn't be enough to overheat them and they would find a very nice equilibrium without taking up a lot of extra space, man that would work excellent in a MM case....
 
Guys can you help me. I tried to chaange the north and south bridge heatsinks on this MSI eclipse x58 mobo. Now the new ones did not fit correctly. So I went back and put on my old ones. I took off the plastic one the bottom of the heat sink and then cleaned the old thermalpaste and re applied with AS5 making sure none got on the transistors. Now I try to boot up and I can not even post or get in bios. The LED is stuck on CPU Lni.
 
first limit all your chances of messing up by narrowing down your points of Failure.
\\

Make sure your 12v motherboard pwr cnnector in in along with the big 24 pin connector.

Only have the HD hooked up
RAm, video card, cpu and PS
and just try for a post


unfortunatly it sounds like a short, so check all around
the area you have been messin with..

if it stil doesn't work, take back off the heatblocks you changed and inspect the chips..
 
it seems like the stackers would need a good high flow pump to keep them running good.

for the price though... you could just get a better rad like thermochill, feser one, or hw labs.
 
With radiators being less than half as restrictive as an average CPU block i would love to see some tests with the stackers, i saw Gabe from Swifetch saying they were going to post some figures that will help highlight what is needed to run those babies
 
I have to say i love to get every degree, but i always felt that te PA 120.3' were over hyped and overpriced. I would see people telling some newbie's that they should get a PA 120.3 when a swifty would be fine, paying ove 2x for only a 5% to 15% gain at best seems pretty shoddy to me,,,(not Talking about anyone here)


But thanks for bringing up the Stakables T-ski, i forgot all about them....

Very interesting product.... very interesting...

http://img.techpowerup.org/090418/rad2.jpg

Yep thats what I was planning on getting. What do you guys use for tightening barbs?

for anyone thinking about stacked rads... check this out.

http://www.realredraider.com/vbulletin/showthread.php?t=5043
 
Just sold my Gene, going to give the Blood Rage another shot. Couldn't help myself...
Something just didn't feel right using an matx board. :laugh:
 
I don't see how in the hell you would get worse temps. I would love to see someone else test these.
Lack of flow comes to mind. Easily compensated for by using better fans, if you ask me. Why in the hell would you run 2 3x120mm rads on just 3 fans anyway, stacked or not?

You figure, if you didn't stack them, you were most likely going to use 6 fans. Might as well still use 6 in a stacked setup.
 
I think I am finally starting to get the hang of this setup, I am just playing around with voltages and combinations of HT on and off plus Turbo on and off to see if I can get the best balance of speed and voltage/temps on air.

it would seem that if I can keep my memory on the 2:8 divider with this board then she will be a lot more forgiving with overclocks, this run is at just over 4.1gig on air with turbo on and the 19x multi, HT off, voltage at just 1.29V idle, 1.30V load :D, temps good!

I want to try this setting now with HT enabled and see if she will hold temps to benchable limits (which I doubt).
 

Attachments

  • 4.1gig.jpg
    4.1gig.jpg
    153.3 KB · Views: 316
Lack of flow comes to mind. Easily compensated for by using better fans, if you ask me. Why in the hell would you run 2 3x120mm rads on just 3 fans anyway, stacked or not?

You figure, if you didn't stack them, you were most likely going to use 6 fans. Might as well still use 6 in a stacked setup.

I would definately use 6 fans. I'm just dumbfounded on how he got worse temps with two rads. Your right, the lack of flow does play a role in this, but I would think a 655 could handle it fine.
 
I think I might have a decent chip here, have managed 15 minutes prime at 4.2gig, it didnt crash then, got upto 74C... I just get bored of waitin :D...... But check the voltage...that might give me a 2006 run :rockout: Turbo enabled, HT disabled and yay! I broke 10 seconds for SuperPI 1M....................
 

Attachments

  • 4.2SuperPI.jpg
    4.2SuperPI.jpg
    136.4 KB · Views: 306
Nice job tatty. Looks like a nice votage for that overclock.
 
for anyone thinking about stacked rads... check this out.

http://www.realredraider.com/vbulletin/showthread.php?t=5043

Interesting post, I reserve all my comments for a follow tests from swiftech and or someone using swiftech rads.


Because as N4cot1c said, i can't imagine it loosing ground to single, unless it impeedes flow so much as to drop a lot, hmmmmm got to go back and study that article... But he us=ually does some good tests.

Also I am sure Gabe will chim in with some numbers too, i don't think this "earl of sandwich" drama is over yet!


another idea is that under a 300w load which is was tested at doesn't evenput stress on the radiators which are rated at higher heat loads, kinda like the MO-ra, it shines under hotter loads. It would be cool to ask him to run the same test on the mo-ra to see where i stands because that type of setup is for higher loads way beyond 300watts.
 
Last edited:
Just sold my Gene, going to give the Blood Rage another shot. Couldn't help myself...
Something just didn't feel right using an matx board. :laugh:

Still love the bloodrage, I like the space saving nature of the Matx and for a system that will stick with one video card, i think it's a cool setup, but that board needs custom cooling solutions on the southie and norhbridge, So i think what you gain in space you still need to do extra work to coool the rest of the componets... I'm still on the bridge with mine...

Glad to see your still banging away on the Bloodrage.
 
I would definately use 6 fans. I'm just dumbfounded on how he got worse temps with two rads. Your right, the lack of flow does play a role in this, but I would think a 655 could handle it fine.

When there is a problem with restriction a 655 is less worthy of the job than a DDC 3.2

As for seeing someone else's results. I try to find as many sources as I can but frankly the guys on RRR are the few who actually have the money and correct test equipment to look into these issues. If you notice it's how the rads are configured and not that there are two rads. Two rads spaced out with 6 fans did swimingly. I think it partially has to do with sending your heat dump through the 2nd rad. See thermal mass can overload certain volumes of transfer. What I'm saying is that after water or air gets heated up to a certain point it has a harder time effectively transfering heat. When you dump hot hair consistantly through the fin array of another rad there will be less heat carried away from the 2nd radiator.
 
+1 DDC 3.2

My only point is to test the stackers with a heat load hat dosn't even push it is a irrelevent test, kinda like testing a 440 cuda on city driving, it's ment for higher loads and then you will see positive differences.
 
Well that still doesn't make sense that stacked rads would perform better on higher load than lower load. The MO-RA 2 has significantly cooler water temperatures than a pa 120.3.

Since it doesn't seem like my point got across... it's all where the heat goes. If it goes from one radiator into another then how is that efficient? The freaking single radiator with 6 fans does better than all of the stacked solutions because the heat is not being dumped back into the loop. Try putting 6 fans on a pa 120.3 and you might get the same results or better than the 2x swiftech rads dumping heat in opposite directions. The problem is tha that HES didn't test that option. I have no idea why... but it doesn't matter. He's a very credible source and it makes total sense when you think about heat transfer. With a sandwitched rad system you're basically heating up the 2nd rad with the first. While the water is moving in a longer series the heat taken off of the first rad is already present in the 2nd rad, so less heat can be dissapated from the 2nd rad. It's the same kind of phenomena with multiple components on the same loop. The first component will dump a lot of heat into the water making it harder for the second component to move the same ammount of heat into the water. Either the flow rate has to increase drastically for this to be effective or the water has to be chilled.
 
I think I might have a decent chip here, have managed 15 minutes prime at 4.2gig, it didnt crash then, got upto 74C... I just get bored of waitin :D...... But check the voltage...that might give me a 2006 run :rockout: Turbo enabled, HT disabled and yay! I broke 10 seconds for SuperPI 1M....................

good job tatty, thats a nice clock for voltage buddy :toast:
 
Those are great overclocks. 4.2GHz is really FAST for just about any application.
 
Those are great overclocks. 4.2GHz is really FAST for just about any application.

im back to 4.2 GHz daily as well.
 
im back to 4.2 GHz daily as well.

For VISTA BOINC works the same with HT on as HT off, just an FYI. Apparently HT slows it down drastically, but only in VISTA
 
Still love the bloodrage, I like the space saving nature of the Matx and for a system that will stick with one video card, i think it's a cool setup, but that board needs custom cooling solutions on the southie and norhbridge, So i think what you gain in space you still need to do extra work to coool the rest of the componets... I'm still on the bridge with mine...

Glad to see your still banging away on the Bloodrage.

I'm going to be running the included water block on NB, and I have a EK block that may work on the south bridge.
 
Back
Top