AMD Protective Heat Spreader Petition

[AtomSmash3r]

All those who think AMD should include an Intel Pentium IV like heat spreader on all their high end CPUs say aye.

I think AMD CPUs are far too fragile. AMD could further increase their appeal by adding a heat spreader to their CPUs. This would aid in cooling and it would prevent damage of the very delicate core. I'm sure this one addition could save a lot of headaches.

Agree or disagree, please post your thoughts in this thread. I'm hoping this thread becomes a petition in favor of including a heat spreader, but who knows?

Thanks.

 

[Spikesoldier]

A few weeks back I posted regarding this topic. I thought that OEM versions should and Retails should not. People would ultimately be removing them to overclock them and just finds the IHS in thier way of a few measley MHz. Now I wouldn't want to call anyone who has crushed a core an idiot or a moron, but I think an IHS should be optional, or left out.

 

[Danlz]

I agree that they are far to fragile and should have a cover like their K6 series. A processor without core protection is like a car without bumpers. AMD's management are not dummies but why they continue to expose the core remains an unanswered question. Another way to send the signal to AMD is to just vote with the ol' checkbook and consider Intel's or Via's superior packaging as an important factor in your next upgrade (I did!).

 

[Jeff7]

Or maybe, leave the core exposed, but use some sort of epoxy around it. There'd be the core, but around it would be a thick layer of something solid or rubbery.
Here's a quick drawing of what I mean.
The core would just be surrounded by something on the same level as the core, or maybe a tad bit lower. It would act to protect the core, but it would still leave it bare.

If they do use a spreader though, maybe using copper in it would be a help. But Spikesoldier makes a good point, that hardcore overclockers would still remove the shield. That's why I think my idea makes more sense.

 

[chemhaqr]

I agree with jeff7. They should leave the core exposed but offer a plastic or metal type shim attatchment to it so the core will get crushed no more. We have been milling over this exact same concept at Hardware central for the last 6 to 8 months. It was concluded by us that a ring type shim would be best to help their cores whil;e offereing ultimate thermal transfer which the ATHY needs anyway. Also I think AMD should make a copper HS standard with all retail chips anymore instead of the coolermaster junk that usually comes with them.

 

[Jeff7]

You mean I came up with a good idea? Cool.

Strange that us end users provide seemingly better solutions than the highly paid engineers at AMD.

 

[o1die]

A personal letter to their Sunnyvale, Ca address wouldn't hurt. I wrote them on the issue of unlocking the L1 bridges on the xp.

 

[sandorski]

Well, I'm sure it might help, but if you just install the proper heatsink in the proper manner you don't have anything to worry about.

 

[Steppy]

Boy, I can't wait to hear the response you get.

 

[Rahminator]

If you follow instructions from AMD on attaching the HSF, the chance of crushing/cracking/chipping is nil. Applying all the pressure to the clip is the only way to do it and don't even dare to touch any other part of the heatsink and you will be safe.

 

[BlvdKing]

The heatspreader will hinder heat transfer on faster Athlons. Not a good idea IMO.

 

[DAPUNISHER]

first poll I've seen where the vote was split down the middle like this. I hope this dosen't require one of our infamous Florida recounts :Q

 

[Spikesoldier]

I decided not to vote, I don't think they need to, but I still think if its going to be there, it has to be totally optional.

 

[Kell]

I think Jeff7's idea is fantastic--and it is technically feasible. All that would have to happen is AMD would manufacture an aluminum shim that leaves the core and the multiplier bridges exposed, epoxy it in place with run-of-the-mill thermal epoxy, and machine/polish both shim and core in the same stroke. Then we'd have the perfect shim, a stronger packaging, and minimal risk of core crushing.

If someone was concerned about hard-pressure core contact, this "perfect shim" could be compromised by polishing an additional micron of thickness off the shim after the core+shim polish. The danger of core crushing would still be minimal due to the flexibility of the new organic packaging.

The only real problem is that some (read crappy) HSFs will not fit with this arrangement (most notably the Chrome/Golden/Super ORBs).

 

[ST4RCUTTER]

I'm in disagreement with some on the heatspreader issue. While I do think it will hinder heat transfer slightly, I think the benefits would outweigh the drawbacks. The percentages of CPU's that are crushed are far higher on the AMD side simply because the idiots who don't follow directions will crush their cores and scream for a replacement, all the while blaming AMD. These same idiots will apply crushing pressures to the Intel die's but fail because of the heat spreader.

In my eyes AMD needs to do two things to finally gain parity with Intel chips: lower core heat dissipation and include a heat spreader. SOI should help with the first problem. Hopefully AMD will resolve the second. In the end though, if the heat spreader costs too much to keep their CPU's cheap to make it will not happen.

 

[Spikesoldier]

Wouldn't we need new heatsink/fan retention clips if they decide to put a heat spreader on their chips? The height changes a little bit.

 

[MadRat]

The heat spreader seems like a good idea at first, because it does provide protection to the CPU core from physical harm. Its horrible from an overclockers standpoint, because it becomes just another poor transition from core to the heatsink. It also does nothing for the thermal protection of the CPU core.

 

[tpetre4322]

i use a mc462a swiftech, and a copper spacer, with foam feet removed,iam glad that they gave me this option, and if they had put a cover over everything, all cores would be multiplyer locked,except the ones that arent now,with no way to change it.

 

[chemhaqr]

<< I think Jeff7's idea is fantastic--and it is technically feasible. All that would have to happen is AMD would manufacture an aluminum shim that leaves the core and the multiplier bridges exposed, epoxy it in place with run-of-the-mill thermal epoxy, and machine/polish both shim and core in the same stroke. Then we'd have the perfect shim, a stronger packaging, and minimal risk of core crushing.

If someone was concerned about hard-pressure core contact, this "perfect shim" could be compromised by polishing an additional micron of thickness off the shim after the core+shim polish. The danger of core crushing would still be minimal due to the flexibility of the new organic packaging.

The only real problem is that some (read crappy) HSFs will not fit with this arrangement (most notably the Chrome/Golden/Super ORBs). >>



I put the orbs on the same shelf as the PIV and winXP. All three utterly worthless. Let them die!

 

[Jerboy]

<< All those who think AMD should include an Intel Pentium IV like heat spreader on all their high end CPUs say aye.

I think AMD CPUs are far too fragile. AMD could further increase their appeal by adding a heat spreader to their CPUs. This would aid in cooling and it would prevent damage of the very delicate core. I'm sure this one addition could save a lot of headaches.

Agree or disagree, please post your thoughts in this thread. I'm hoping this thread becomes a petition in favor of including a heat spreader, but who knows?

Thanks. >>



Difinitely.


Yes, my opinion on fragileness of AMD is much stronger after my independent survey taken here.

You'll understand why...
Jerboi's survey 1
Jerboi's survey 2

Among those responded, 15.5% of the participant on the average has damaged their Athlon in such a way that it resulted in total-loss during installation one way or the other. Quite a few people also chipped it, but still works.

 

[AtomSmash3r]

Perhaps a heat spreader would be unessecary once motherboards are in compliance with the Palomino and Morgan thermal diode -- at least from a "frying the chip" standpoint. This still doesn't help the fact that the cores are far too easy to damage.

 

[Jeff7]

I pitiful bump to encourage my idea (see further up in the thread).

 

[Bovinicus]

Despite what Orcish says, if AMD hypotehetically could use a heat spreader, I think they should. It is not their job to cater to hardcore overclockers. Why would they not use one just because hardcore overclockers might remove it to get a few extra MHz... That is one of the dumbest comments I have ever heard, I am sorry to say.

 

[Jerboy]

<< Despite what Orcish says, if AMD hypotehetically could use a heat spreader, I think they should. It is not their job to cater to hardcore overclockers. Why would they not use one just because hardcore overclockers might remove it to get a few extra MHz... That is one of the dumbest comments I have ever heard, I am sorry to say. >>



Because, heat spreader means less incidence of core breakage, which in turn means less revenue from individual users having to make multiple purchases to replace the wrecked processor(s).

 

[BlvdKing]

A heatspreader is a bad idea. It not only covers up the L1's, but it would hinder heat transfer on faster Athlons, and might even effect yeilds. I do like the idea of a small copper/aluminum shim that surrounds the core, thus giving the heatsink a bit more surface area to sit on.