The temp issue only affects us enthusiasts that actually bought a K type processor for the sole purpose of overclocking, not so for the common user who only use their PC for Facebook or Angry Birds. Since there are more who don't dabble with overclocking, you could almost say that the temp issue is almost nonexistent hence swept under the rug. Without a competitive product from AMD, there is no urgency for Intel to make their product more receptive to the mainstream users.Possibly? The fact that IB bets hot is now known even to non-geeks, so addressing that might make people more receptive to their product?
Admit it, Intel could just solder the die to the IHS, release it as the 3870K and we'd wet ourselvesWhy doesn't intel resolve this issue already when they could shave off temps in the double digits just by changing the TIM with something better?
Well, it's interesting that our attentions focused on "cooling solutions" during the last decade, with Rube Goldberg contraptions, water-cooling, lapping, TIMs and other aspects. All that time, each generation of processors would first appear as "toaster-ovens," and then Intel would find a way to reduce the TDP. Now, we're at 77W -- close to the 65W level of earlier processors barely half as powerful.Admit it, Intel could just solder the die to the IHS, release it as the 3870K and we'd wet ourselves![]()
I think it would with the K series because enthusiasts/overclockers buy them. With everything else? Probably not.Would shaving off temps in the double digit with solder = more money in Intel's coffers?
Exactly, that was my point. Intel caters to the bigger crowd, which happens to be mainstream users who don't overclock or don't know how to overclock.I think it would with the K series because enthusiasts/overclockers buy them. With everything else? Probably not.
Well, remember -- there's a widely held opinion that they had reason to depart from that practice. I think I saw a white paper posted at the Intel web-site about "thermal cycling." Someone else could offer more concrete adjustments to the sense of risk we have about metal-based TIM concoctions.So then they should use the soldering method for their K series...
Updated now that I have push/pull configuration on the ZT-10D now. Delidded 3570K w/ IC Diamond between die and IHS, then Arctic Silver Ceramique between the IHS and ZT-10D. Hopefully will be trying CL Liquid Pro on all surfaces soon.Under Prime load:
4.5Ghz - 1.216v
4.6Ghz - 1.264v
4.7Ghz - 1.312v
SuperPi runs:
4.8Ghz - 1.312v
4.9Ghz - 1.352v
*updated*
3 weeks in temps are back to pre-mod level with IC diamond. Hopefully they stop there and I don't have to reapply like Ferzerp, but it's not looking good.I delidded mine on 8/12/12, posted in a thread on Hardocp. Used IC diamond under the IHS.
Temps are the same 2 weeks later after testing. Not long enough to make any conlcusion, but I'll update again in a couple more weeks to see how it is doing.
I wonder if anyone has tried this with the IHS walls trimmed away? It seems to me that a fixed height of the IHS and thermal cycling would explain this.3 weeks in temps are back to pre-mod level with IC diamond. Hopefully they stop there and I don't have to reapply like Ferzerp, but it's not looking good.
I wonder if anyone has tried this with the IHS walls trimmed away? It seems to me that a fixed height of the IHS and thermal cycling would explain this.
If the gap between the IHS and die were close enough, it's hard for me to imagine that there is any "pumping out." I can't say. The ICD is very thick stuff. With an ample application, you'd think the paste would "stay put."pwoz said:3 weeks in temps are back to pre-mod level with IC diamond. Hopefully they stop there and I don't have to reapply like Ferzerp, but it's not looking good.
After all the work I've done, the IHS doesn't sit on the PCB, it sits down on to the die.I wonder if anyone has tried this with the IHS walls trimmed away? It seems to me that a fixed height of the IHS and thermal cycling would explain this.
Last I saw . . . you were using the IC Diamond? I can't recall what FerZerp was using.I will if I don't get Liquid Pro/Ultra before that time. I'll do a Prime Blend for 30 minutes at a known frequency and voltage today and post the maximum temperature results, and then I'll come back to it two Mondays from now (and maybe three, then four, etc.) and do the same thing if I don't upgrade to the liquid metal TIM before then.
Yuriman: Reapplying a TIM between the IHS and the CPU die every few weeks is a totally unacceptable state of affairs. We're trying to find a solution that is permanent, and there is nothing certain yet. The two best possibilities are the diamond paste and the liquid metal or metal-pad products, and some people are starting to test them.Yuriman said:Which TIM should I use for between the die and the IHS, and which for the IHS-> heatsink if I'm concerned about longevity? I'm not keen on the idea of having to reapply every few weeks. Has anyone had any longevity issues with liquid metal pro, or will it last a really long time? I currently have a tube of HeGrease Extreme which was reviewed pretty well by a few sites but I realize some of these other pastes are probably significantly better.
Yes, I am using IC Diamond. I just sent an email to Xoxide letting them know the testing that I'm doing in the hope that I can get some Liquid Pro and/or Liquid Ultra and post results using those. I figure if they are gracious enough to send it to me for free or at least give me some free shipping then I'll have results for everybody to see, and possible customers to send their way. Never hurts to ask, I guess?Last I saw . . . you were using the IC Diamond? I can't recall what FerZerp was using.
Definitely doesn't. I asked Arctic Silver way back in the day, and they ended up sending me 50 tubes of AS to use for our HSF testingYes, I am using IC Diamond. I just sent an email to Xoxide letting them know the testing that I'm doing in the hope that I can get some Liquid Pro and/or Liquid Ultra and post results using those. I figure if they are gracious enough to send it to me for free or at least give me some free shipping then I'll have results for everybody to see, and possible customers to send their way. Never hurts to ask, I guess?
I'm going to keep mine rolling for now. I'll just post an update every week until it stabilizes or I have to redo it.Would anyone be willing to leave it for longer than 2-3 weeks instead of reapplying to see if it will degrade even more?
Thermal Cycling/TIM Failure Test
System Info:
3570K Delidded with IC Diamond between the die and IHS
ZT-10D HSF with Arctic Silver Ceramique between the IHS and the HSF
4.5Ghz @ 1.232v (after vdroop)
Delidded and IC Diamond applied 8/29/2012
After one week:
(9/4/2012)
Ambient temp - 25C
3570K max temps after Prime "blend" for 30 minutes - 67C, 76C, 74C, 73C
More results to come...
I understand that, but I'm wondering if there is a any "slow degrade" pattern to it if the transfer ability is observed over a long period. I get that once it's reached the "failure" point it will take just moments for the temperatures to spike.I'm not sure you really need a long term test. When mine has issues, I see it within 5 seconds of starting up IBT even on the default settings. It's not an issue of cooling system heat soak (which long runs will show), but instead is a heat transfer issue, and so is pretty immediate when it happens.
Thread starter | Similar threads | Forum | Replies | Date |
---|---|---|---|---|
K | Discussion Power (TDP) ZEN4, AM4 vs AM5, Power stage Vcore load 80A - 90A - 105A | CPUs and Overclocking | 0 |