storm vs. apogee gt/gtx vs. fusion ( 2 Views )

1. So I keep hearing from everyone that the storm isn't great for multicore processors and I recently upgraded to a C2D E4300. My temps aren't great (43c idle, 60c load) but if i upgrade blocks would i really see that much of a difference? It would need to be a full 8-15 degrees C to make it worth my while in upgrading.
   
   I searched around a bit and found a review of a guy who had compared a storm (different revision than mine i think... are there many revisions of the swiftech storm?) with an apogee gt and gtx and a dtek fusion and his temps weren't that different between the 3 blocks (storm was the worst but not by much).
   
   Basically what i'm asking is: is it worth the upgrade? $60 for 3-4 degrees C seems a bit steep to me unless someone knows a different review with a more favorable conclusion.
   
   thanks
   
   
   (alparslan, El Salvador)


2. It would be more correct to say that the Storm isn't optimal for the mechanically flawed IHS implementation on Intel S775 CPU's. Its primary cooling contact patch size is sufficient to cool a dual-core CPU, but the mechanical loading of the mating pressure between the CPU die, IHS, and waterblock base is such that a large proportion of the heat load from the CPU actually gets distributed to the edges of the IHS, rather than toward the center, which is what the Storm design focuses on cooling.
   
   This is not a problem with AMD IHS capped CPU's. Just Intel's S775 CPU's.
   
   The Swiftech Apogee blocks may have their base-plates artificially bowed in a convex manner to address the mechanical flaws in Intel's S775 IHS implementation, and this has the effect of squeezing the IHS more firmly between the CPU die and the waterblock base. As a result of this an Apogee GTX can perform up 6C better than a Storm on a heavily overlocked C2D CPU under heavy load. Without the bowed base, the difference is more like 1-2C in favor of the Apogee, and then only because the S775 CPU's IHS doesn't distribute heat primarily in the center of the IHS unlike AMD's implementations.
   
   So it's really up to you. If you're going to be sticking with S775 for a long time it might benefit you to choose a block that can be warped to address the mechanical flaws in Intel's IHS solution for maximum cooling benefit.
   
   For everything else, the Storm block is still a solid performer, right up there with the best of the rest, and on bare-die CPU's, it is still unmatched.
   
   
   (helin, Morocco)


3. Would lapping the IHS address this issue? I'd think that if the IHS is concave and it's causing the issues with the Storm, a good lapping and flattening would cure it.
   
   
   (mýsra nur, Lao People's Democratic Republic)


4. most people prefer not to lap their cpu because it will void the warranty. In their case, the apogee would suit their needs. I have no qualms about lapping, so I personally find a DTek the best bet.
   
   
   (akýn, Macedonia, the Former Yugoslav Republic of)


5. its more about poor contact between the core and the IHS, than the IHS and the waterblock.
   
   
   (demir, Uganda)


6. I'm staying with my AMD 939 pin after reading this thread. How did something like this ever get out of the factory? :confused:
   
   
   (gülþen, South Georgia and the South Sandwich Islands)


7. by the nature of manufacturing, all IHS are slightly curved. It just seems the 775 is a little more flexed.
   
   
   (ahu, Bosnia and Herzegovina)


8. thanks for the reply, from the exact guy i was looking to hear from too :D
   
   i may lap the HS but I really like my storm block and don't really want to replace it... Might try to reseat it again though putting more pressure on one side to see if that helps at all. Or maybe i'll just lap the cpu.
   
   
   (sokaklarýn, Singapore)


9. First thing I did with my 185 was cut off the IHS once I'd tested it worked. Made a huge difference to load temperatures... huge!
   
   But it seems that both AMD and Intel are now cementing the things on with proper glue or solder rather than easy to cut rubber.
   
   
   (laybýku, Congo, the Democratic Republic of the)


10. Quote:

    Originally Posted by Gyrfalcon (Post 1031166342) its more about poor contact between the core and the IHS, than the IHS and the waterblock.

    Well, it is both. If I sandwich something between two things and push them all together with some degree of force, how much force is being applied at each point? The answer is the same.
    
    I think what you're trying to say, and I agree, is that lack of pressure between the center of the IHS and the waterblock is less of an issue because the contact of the surface area is large enough.
    
    I supposed that lapping would fix it to a fair degree, but you'd be wanting to deliberately make the surface of the IHS convex.
    
    
    (þeyda, Latvia)

Related Topics ... (or search in 1.720.883 topics !)