the si97

[thiago]

http://www.newegg.com/app/ViewProductDesc.asp?description=35-109-120&depa=0

It says it's only supported for barton 2800+ and up.. I have a 2500

I had a thread on how I should cool my cpu and everyone voted this one to be the best... Why did they tell me to get the si97 if it wont work on my comp, Unless this is bad information which i doubt.. what should i do about buying a good heatsink now should i get the sp97 ? help me out

 

[redhalo]

Ok, I'm not trying to insult your intelligence cause I can see how thats misleading, but think about it. If it can cool a 2800+ which should run hotter than the 2500 then it would be even easier for it to cool the 2500. I personaly have a 2500 Mobile and the Si97 with a Vantec Tornado on a Vantec fanbus set as low as it will spin. Wonderfull combo. Currently have it overclocked from it's original 1.8ghz to 2.5ghz. I don't know why they say 2800+ but if it can cool something hotter then what you have then it should be no problem.

 

[c00z]

redhalo what temps are you getting with it for idle and for load as well as what voltage are you using for 2.5ghz

 

[redhalo]

redhalo what temps are you getting with it for idle and for load as well as what voltage are you using for 2.5ghz

At work right now so I can only go off of memory. Before with factory heatsink I was getting idles of 47 and loads around 55 after Si97 and OC I'm doing 40 idle, 45 load with my vcore set at 1.65. Now realize that my fan is at half speed. When I get home tonight I'll get some readings for ya at stock clock speed and fan at full blast to see other comparisons. That Vantec Tornado is a beast turned up. I want to say at 1.8ghz with full blast I was idleing somewhere in the mid 30's. I may be wrong though.

 

[undertheradar]

heatpipes sometimes have a minimum heat requirement or else the liquid in the pipes doesnt absorb enough heat and doesnt move as fast as it should...meaning you arent cooling as well as it could if it were a little hotter. Ever notice how in tests, heatpipe coolers often get outperformed by their metal-fin only counterparts when under idle? But when under load temps then the heatpipes really kick in and are able to transfer the heat better than anything else almost as well as water cooling? Thats because the speed that the liquid can carry the heat away inside the pipe is depending on the amount of heat that the processor generates. This is one reason why the thermalright XP-120 doesnt seem to outperform its XP-90 counterpart. I bet that once both of them are strapped onto hotter procs then the advantage of the extra fins space and heatpipe will become more evident. To be on the safe side with that XP chip, be sure to overclock it some to be sure it hits about 100degreesF under load. I am not sure that this may/may not be an issue with the SI-97, but my experience with heatpipes at work tells me this might be the reason for the minimum requirements...nothing a little OC cant fix however...

 

[redhalo]

Well I just learned something All the same, Si97 works great with my 2500 Mobile so take it for what it is.

 

[Python14]

Yeah dude, you'll be happy with the SI-97. I've got one on my 3000+ and it's beautiful.

 

[c00z]

yeah i am looking to get some ideas of what people are getting for idle temps as well as load temps for SI-97s that are ontop of heavily oc'ed mobiles. I specifically want to know case temps, ambient temps, voltages and mhz of the chip since I think mine is idling way to high for one wiht a 120mm fan on it, but just want to check.

 

[chile7236]

i was getting a 38C idle at 1.85 when i had mine...i've seen more than a few others lower...the SI97 is a great HS...

 

[Rad777]

It says it's only supported for barton 2800+ and up.. I have a 2500

Why did they tell me to get the si97 if it wont work on my comp,

Dude... Thats sad... Someone is lacking some common sense ^_^

Rad777

 

[thiago]

Dude... Thats sad... Someone is lacking some common sense ^_^

Rad777

Well just help me out instead of saying im sad... maybe i smoke way too much weed whats your problem! will a si97 work for my pc ?

 

[redhalo]

Dude... Thats sad... Someone is lacking some common sense ^_^

Rad777

Wow that was highly inapropriate. If everyone knew the answers there wouldn't be much use for these forums. Someone is lacking common sence, and I'm not refering to thiago.

 

[ST|FFY]

Well just help me out instead of saying im sad... maybe i smoke way too much weed whats your problem! will a si97 work for my pc ?

redhalo already answered this and the answer is yes

 

[undertheradar]

well, if you read my response, you will see that there can be such a thing as minimum heat generation when we are looking at heat pipe coolers. The only people to know for sure would be the engineers who made the SI-97, but in a more extreme example lets say we put this giant on an old socket 370 like about 500mhz. not alot of heat. The heatpipes wouldnt get enough heat to start the phase transformation process before the heat just escaped, making the cooler's design pretty much useless. Thiago was right to question if there was a reason for a "minimum" rather than just assuming that bigger is always better, as the people who assign that rating might/must have a reason, and overlooking that is an even bigger error in 'common sense'. So rather than attacking him for his logical question, you might want to step back and realize that he might be showing more common sense than you...and making such remarks in this forum shows an even bigger lack of common sense in the first place for treating someone like that...

AHEM, so...what was that you were saying?!

 

[redhalo]

well, if you read my response, you will see that there can be such a thing as minimum heat generation when we are looking at heat pipe coolers. The only people to know for sure would be the engineers who made the SI-97, but in a more extreme example lets say we put this giant on an old socket 370 like about 500mhz. not alot of heat. The heatpipes wouldnt get enough heat to start the phase transformation process before the heat just escaped, making the cooler's design pretty much useless. Thiago was right to question if there was a reason for a "minimum" rather than just assuming that bigger is always better, as the people who assign that rating might/must have a reason, and overlooking that is an even bigger error in 'common sense'. So rather than attacking him for his logical question, you might want to step back and realize that he might be showing more common sense than you...and making such remarks in this forum shows an even bigger lack of common sense in the first place for treating someone like that...

AHEM, so...what was that you were saying?!

Ok, work with me on this one, if the temps are too low the heat pipes won't work to thier full abilities, but with that regard they arn't needed to work fully untill they reach the prescribed temp anyways. So, It will still work for his needs unless the chip has and ungodly low threshold for burning out. (shouldn't). Like having a car that can go 200mph, it may be stated that it performs it's best on the track, but it will still get you to the grocery store. Just not as fast (unless you speed). Through practice though, like I said before, it works fine for the 2500.

 

[undertheradar]

maybe the example i gave wasnt the best, as the chip could cool w/o even using the heatpipes (lets use HP). There is a fine threshold however, where HPs work less than regular metal (copper, aluminum, silver, gold) heatsinks. There is not enough heat to get the HPs going effectively, yet too much that the buildup required to get the pipes going could hurt the chip (or at least not give you the best cooling potential like a regular fan/sink would). An example of this is how with many HP-coolers the idle temps can be up to a few degrees warmer than a non-HP counterpart. Testing Aerocool 101 Towers shows this the most because of the big (1" diameter) pipe in the center. It makes it look as if the HP is less effective...and it is with smaller deltas in air temp and proc temp. But as the proc generates more heat under load, the heatpipe gets the heat it needs to become more effective, to get the fluids moving and vaporizing to carry heat away faster. Much like the turbo on a car, it requires a certain amount of input to get the 'wheels moving' and get the reward of more cooling capacity. This is how HP-coolers dont often perform as well on idle, but pull away in the lead when under load or overclocking. You should see the big HPs we are developing where I work. We are doing testing on HPs for automotive applications as well as jet engines. HPs can move alot of heat...much more than 'common sense' might suggest...almost like having a liquid/refrigerant cooled setup...but they perform so well because the difference between the hot end in the engine and cool end at the radiator is do great...with computers it is more difficult because the delta between the room temp/case temp and the proc is a much smaller margin. Thermaltake might be suggesting that while the XP2500 might work with an SI-97, it might not generate enough heat to make the SI-97 an effective cooler as say...the 2800 on up. Perhaps a stock 2500 would be better cooled by a regaular heatsink like a volcano or SLK model. But just to be sure, OC that baby to be a 3000 and that will be pretty much what you have...