I took IHS off my X2 3800+!

L

Link

Gawd
Joined
Jul 28, 2002
Messages
699
Taking the IHS off an A64 was very easy. It only took me about 15 minutes.However, it didn't help bringing down temperature for my X2 3800+. :(
I'm aircooling it with the SI-97A, and I thought taking the IHS would help bringing down further even with the same HSF. Wasted my 15 minutes and 3 year warranty. :mad: So mad at myself for not being able to resist the temptation.
 
lol..... the ihs probably made REALLY REALLY good contact with the core lol :p
 
I for the life of me will never be able to understand why someone would want to do this.

I mean come on people, taking the IHS off such an expensive processor is suicide. especially considering the fact that most of the times it doesnt do dick anyways.

oh well.... sorry for your luck. :)
 
I can't imagine that your temps didn't change one bit after removing the IHS.

Regardless of how good the contact was, shoving a piece of aluminum in between a core and your HSF is never good.

All the results I've seen (and they aren't too many) show around a 5-7C drop in temps.
 
You could have just lapped it, would've been alot less risky than taking the whole thing off.
 
ozziegn said:
I for the life of me will never be able to understand why someone would want to do this.

I mean come on people, taking the IHS off such an expensive processor is suicide. especially considering the fact that most of the times it doesnt do dick anyways.

oh well.... sorry for your luck. :)
Click to expand...
You gotta be kidding me, removing the IHS always drops temps, you obviously don't know anything if you think it doesn't do "dick" most of the time :rolleyes:

If you haven't seen a temp drop after removing IHS, first of all you should use a probe instead of trusting the mobo sensors since they're always way off. If you measure temps with a probe and they still haven't gone down then you must be getting really bad contact.
 
YmkFX said:
You could have just lapped it, would've been alot less risky than taking the whole thing off.
Click to expand...
But it also wouldn't have achieved the same objective. It's an option but not an equal alternative.

Link, does your heatsink press very firmly against the CPU when you tighten it down, or only enough to hold it up?
 
I made very sure that HSF is sitting very firmly against the CPU.
 
The only thing I can think of right now is that maybe your heatsink needs lapped. If its curved your contact area might not be much better.

But like HiJon said, if you remove the heatspreader you should always get lower temps. It definitely means something funny is going on.
 
RawsonDR said:
The only thing I can think of right now is that maybe your heatsink needs lapped. If its curved your contact area might not be much better.

But like HiJon said, if you remove the heatspreader you should always get lower temps. It definitely means something funny is going on.
Click to expand...
Tomorrow, I'll take the HSF off, and see how well it is making contact with the CPU. At least, so far, temperature isn't higher than before.
 
Link said:
Tomorrow, I'll take the HSF off, and see how well it is making contact with the CPU. At least, so far, temperature isn't higher than before.
Click to expand...

And please, take a picture too. :)
 
Doesn't it also take some time for the thermal compound to settle? If you're using AS5, maybe giving a few hundred hours would help out in the temps.
 
mikelz85 said:
Regardless of how good the contact was, shoving a piece of aluminum in between a core and your HSF is never good.
Click to expand...

Isn't it copper with a nickel finish?
 
mikelz85 said:
Regardless of how good the contact was, shoving a piece of aluminum in between a core and your HSF is never good.
Click to expand...

but then again, taking off that useless piece of aluminum thus leaving the core very vunerable to getting damaged VERY EASILY is even worse.
 
HiJon89 said:
You gotta be kidding me, removing the IHS always drops temps, you obviously don't know anything if you think it doesn't do "dick" most of the time :rolleyes:
Click to expand...

okay bigshot.

show me some kind of hardcore proof that proves that removing the IHS does any good. no, I'm not talking about lowering temps by a measly 5 or 7C difference. I'm talking about something like being able to add a 200 or even 300MHz O/C to someone's speeds by doing this VERY risky mod.

every single person who I have seen that has removed their IHS from their X2s MAYBE and I do mean MAYBE got an extra 100 or 150MHz O/C from their chips. if that...

now, you have to ask yourself: is this really worth the potential damage to your expensive chip? or even worse yet, is this worth throwing your $350 or $500 chip's warranty out the window?

NOT.
 
ozziegn said:
okay bigshot.

now, you have to ask yourself: is this really worth the potential damage to your expensive chip? or even worse yet, is this worth throwing your $350 or $500 chip's warranty out the window?

NOT.
Click to expand...


My feelings exactly.
 
So don't do it. That simple. What kind of point are you guys trying to make? Just b/c you don't feel comfortable doing that to your processor doesn't mean the next guy feels the same way - or that he really cares how you feel for that matter.

I for one would love to see pics of the nekkid cores. Please post a couple.
 
Elias said:
What kind of point are you guys trying to make?
Click to expand...

my point is this. all I'm saying is that removing the IHS simply isnt worth the minimal speed gains (if any at all) compared to the high risks that could and probably will go wrong from damage to the core.

I always like to express my opinion on this subject just in case somm n00b decides to try this out but doesnt really know what he (or she) could be getting themselves into.

you know, an ounce of prevention is worth a pound of cure....

I for one would love to see pics of the nekkid cores. Please post a couple.
Click to expand...

here's an X2's IHS removed from someone at the OC forums.
 
ozziegn said:
my point is this. all I'm saying is that removing the IHS simply isnt worth the minimal speed gains (if any at all) compared to the high risks that could and probably will go wrong from damage to the core.
Click to expand...
it might not be worth it to you, but that's not gonna stop people from trying :rolleyes:

my experiences
http://www.hardforum.com/showthread.php?t=851884
http://www.hardforum.com/showthread.php?t=857220

tenchi4u:
http://www.hardforum.com/showthread.php?t=839108
http://www.hardforum.com/showthread.php?t=840062


also.. you can't say that the cores are fragile...

OWNED BY SANDPAPER!
and it still works ;)

clawhammer.JPG


the only way to break a core is if you're just being stupid and careless with your heatsink installs, or if you like to drop the case on the floor..
 
^ LOL - as if removing the IHS wouldnt be enough, but now you're repeating the old sandpaper story?

good Lord.... I guess that makes you all that much [H]arder than I am ( :rolleyes: ) cuz I damn sure cant afford to throw my $400 CPU's warranty out the window.
 
I really don't see a reason to remove the ihs. If 1-2 deg C is causing your instability you have far greater thermal management issues.

Now if you hav etons of money and just want to play with your cpu be my guest.
 
personally, a speed gain of "100-150mhz" is pretty high imo. Hell, I'm all about the numbers. So I got 2.93ghz with chilled water (25C load temps) on my 3000+ venice. Then I bought a used vapochill. After I get the vapo up and running, I'm sure the IHS will come off of the 3000+, leaving me with (hopefully) more FX trumping [H]ard Superpi times.

Does it do a damn thing for performance? Not really, no. Maybe I could use my weekly dvd recompressing as an excuse for the needed power, but I won't. I honestly have no more use for a 3.2ghz chip than a 2.4ghz chip, except superpi times and other synthetics. Just like I'll tell anyone that buying overclocking ram is stupid, unless you're doing it just for benchmarking, and I'm sitting on 2x512 twinmos SP @270 1.5-2-2-0. It sure is fun to do, but when it comes down to it, 90% of my posting come from either a barton or duron box, both of which seem plenty fast to me.

Point blank, if it's worth the risk to you (most of which comes from not having the protection on the core) then it's worth the risk.
 
Well,,,did some more testing.
The idle temperature is about the same, but the load temperature got lowered, and cpu needs lower voltage to run at the same speed to be Prime 95 stable.
It's been 4 hours doing 2 instances of Prime at 2.6ghz on 1.48V. Temperature is at around 45~48C (80F room temp)
 
ozziegn said:
okay bigshot.

show me some kind of hardcore proof that proves that removing the IHS does any good. no, I'm not talking about lowering temps by a measly 5 or 7C difference. I'm talking about something like being able to add a 200 or even 300MHz O/C to someone's speeds by doing this VERY risky mod.

every single person who I have seen that has removed their IHS from their X2s MAYBE and I do mean MAYBE got an extra 100 or 150MHz O/C from their chips. if that...

now, you have to ask yourself: is this really worth the potential damage to your expensive chip? or even worse yet, is this worth throwing your $350 or $500 chip's warranty out the window?

NOT.
Click to expand...
Measly 5 or 7C? WTF are you smoking, 5-7C is huge. And 100-150Mhz extra, people pay big money for better cooling to gain 100-150Mhz, and just by removing the IHS you can gain 100-150Mhz for free.
ozziegn said:
is this really worth the potential damage to your expensive chip? or even worse yet, is this worth throwing your $350 or $500 chip's warranty out the window?
Click to expand...
Hellz yeah, and by someone even asking that question I can tell I'm not at XtremeSystems or XtremeResources anymore :rolleyes:
 
Wow, i'm surprised at the vehement reactions to this post. If done carefully there's not a lot of risk in removing the IHS.. let it be.. stpo crapping on this guys party..
 
ozziegn said:
cuz I damn sure cant afford to throw my $400 CPU's warranty out the window.
Click to expand...
you're missing the point. just cause you feel one way doesn't mean everyone does. ;)
(also, i bought an oem cpu so no warranty. it clocked like shit. i was planning on getting another cpu anyways, so there was no incentive for me not to fuck around with it)

HiJon89, well stated. there are those of us who don't care, those who like to mess with stuff, and those who seem to object to the people who mess with stuff. i'm sure people know which i am :p
 
messing with stuff is always fun but most temp readings people use are probably only accurate to within 10 deg C. So untill someone a signifigant sample of readings are taken with a rarther accurate thermal probe or at least callibrated temp reporting (linky). People should stop handing out this 5-7 deg C temp drop figure.

I agree this forum is quite different from XS, but removing the HS is not free; risk has a cost.
 
DryFire said:
messing with stuff is always fun but most temp readings people use are probably only accurate to within 10 deg C. So untill someone a signifigant sample of readings are taken with a rarther accurate thermal probe or at least callibrated temp reporting (linky). People should stop handing out this 5-7 deg C temp drop figure.

I agree this forum is quite different from XS, but removing the HS is not free; risk has a cost.
Click to expand...
well, given that the temp reading are taken on die by the processor, even if they are "inaccurate" given the actual temp, I'd imagine them to be at least uniformly off, thus, if you recieved a certain temp before the IHS removal, and a lower temp after, I'd say the temps got lower. These aren't exactly the little blue diodes of the AXP days.
 
I'm not saying temps aren't going down (I assume they would) but it may interpret a 2 degree drop as a 5 degree drop or more.
 
DryFire said:
I'm not saying temps aren't going down (I assume they would) but it may interpret a 2 degree drop as a 5 degree drop or more.
Click to expand...
and in that vein, it could interpret a 10 degree drop as 5 ;)
 
DryFire said:
Yes, so at best you're hazarding a guess.
Click to expand...
no, I still say it is accurate, eclipse was just poking a hole in your arguement that temp drops were overblown
 
Wouldn't watercooling be better than throwing a $500 chip's warranty out the window and risk destroying it?
 
ozziegn said:
now, you have to ask yourself: is this really worth the potential damage to your expensive chip? or even worse yet, is this worth throwing your $350 or $500 chip's warranty out the window?

NOT.
Click to expand...

Noob statement/question but...I always thought overclocking the CPU would void your warranty... :rolleyes: I guess I'm wrong?
 
ozziegn said:
but then again, taking off that useless piece of aluminum thus leaving the core very vunerable to getting damaged VERY EASILY is even worse.
Click to expand...


I’m not entirely sure what the “IHS” is made from; I do know that if you lap it as the silver color goes away you get copper. Never tried lapping past the copper.

Fact, if people didn’t experiment and “break the rules” once and a while this thread would be about as boring as the “rules” thread (Sorry Kyle).

Only by pushing the limits and once and a while breaking something do we learn. The more we learn and the faster and cooler we run, the less it costs many of us in the end.

The only time I’ll worry is when we get to sacrificing virgins (if you can find any) to the over clocking gods and no one takes pictures. :rolleyes: ;)
 
You must log in or register to reply here.
Back
Top