LinkBack URL
About LinkBacksRead more.But the off-the-shelf unmodified processor has a 'low-quality' TIM.
Read more.But the off-the-shelf unmodified processor has a 'low-quality' TIM.
For the prices they are charging, you would have thought they would be soldered. Another reason for me to buy AMD.Sadly Intel's most premium consumer processors still use a 'conventional paste' thermal interface material (TIM). They aren't soldered
Every time someone accuses Intel of using a poor quality TIM i want to scream, do people really think a company that spends $12.7 billion on R&D is going to penny pinch when it comes to TIM for nothing but financial reasons.
Jonj1611 (31-05-2017)
I agree but solder costs money i mean they would have to add at least another $100 to solder the lids on te cpus. (sarcasm) They are bareley scracping by charging what they do, i mean look at all the innovation they have brough us in the last few years (/sarcasm).Originally Posted by Iota
I have a feeling in a couple of years we will start to see the same anti competative practises admissions from some OEMs (not by the OEMs but by Intel) which we heard in the past when thunderbird etc could nbot get a foothold in the market despite being better. I mean look at whats already happening with the forcing of optane on motherboard manufacturers...
Why not, it happens everywhere else. Motor manufacturers spend vast amounts of R&D yet still have to issue recalls because they cheap out on a component.
Last edited by MonkFish; 31-05-2017 at 02:44 PM. Reason: Grammar
Because CPUs and material sciences isn't comparable to motor manufacturers where cheaping out on a component isn't financially, let alone scientifically, provable to be detrimental to the final product.
And before anyone jumps down my throat I'm not saying Intel uses the best quality TIM full stop, I'm saying they use a TIM best suited for their specific needs and those needs do not include cutting costs, a TIM needs to do more than just provide low thermal resistance.
Jonj1611 (31-05-2017)
Exactly it also has to be good value when you are applying it to million of processors.
I'm not sure contextomy works when people can read what i said directly above your reply.
I'd be intrigued to know what other need you think using TIM instead of solder meets? Perhaps Intel's silicon process is more fragile than GloFo's and they can't heat them up sufficiently to melt the solder...?
While not impossible, the fact that all previous HEDT chips including Broadwell-E (also on 14nm) were soldered makes that a bit unlikely though. And as the Haswell Refresh showed, even Intel acknowledge that their previous TIM wasn't very good and went on about their next-generation polymer thermal interface material (NGPTIM). So I think the far more likely explanation is that Intel feel that their TIM is good enough and that equally applies to a $42 Celeron as it does to a $1999 18-core i9.
Out of interest, does anyone know if their subsequent CPUs (Skylake, Kabylake) use the old TIM or NGPTIM?
Not being a material scientist myself I'm only going on research papers I've read but from what i read about it indium solder started causing problems when die sizes started shrinking, OTTOMH i can't remember when Intel made the switch and what the dies size was, however they started detecting cracks and voiding within the solder joint, and i assume things like pump out and drying out is accepted as possible problems when using thermal grease.
Overclocking guide covered some of the problems with solder joints in this article back in 2015, if you really want to geek out though Intel published a more technical look at the problem in their technology journal back in 2008 (PDF).
"OMG! I missed the bus and will now have to wait 5 minutes for the next one... I'm buying AMD! (As it's the latest cool thing to say)"
Re: Intel Core i9-7900X breaks several benchmark world records
You're not making sense
------------------
Valar Morghulis
People scream when the TIM thing is brought up but there is a great example here that proves Intel need delidding. If they didn't, why would there be so many tools to do it for Intel CPUs but barely any for AMD. It was said when people were reviewing the 1800x and 1700 when overclocking they saw no discernible difference (delidded if they managed) or not. I think it was der8auer or another top overclocker who uses sub-zero cooling, I will need to dredge them up.
Edit: Found it without having to watch der8auers video: https://www.digitaltrends.com/comput...g-unnecessary/
Last edited by Tabbykatze; 31-05-2017 at 04:53 PM.
Thanks, interesting stuff.
That said, the 2008 paper was about Intel's move to Indium based STIM, not away from (it was part of the lead-free program). Yes, they discussed issues with the material, but in terms of issues they were able to overcome/design around.
The 2015 article touches on increasing failure rates for very small dies*, so I guess you can understand why they'd go for a paste for KBL-X where you've got the same small die problem as with standard SKL/KBL. I guess that also potentially explains why AMD are OK soldering Ryzen (which I believe is somewhere between 180mm2 and 220mm2).
Skylake-X, OTOH, is still a pretty big die. And it's derived from server/workstation chips which are often used in thermally challenging environments. I find it hard to believe that there's a genuine technical issue with soldering those, particularly when AMD are soldering similar size (if not smaller) dies without issue.
Having now read up on how complex the process of soldering an IHS is, I suspect that at least part of the decision was financial. I reckon there's at least three production processes they can save money and time on if they go for paste rather than solder. The move away from solder may have been predicated by technical issues with small chips, but the decision to apply it to the entire product range...? I see $$$$ and hear ka-chings....
* which, as a cheeky aside, is actually a rather serious cost-related issue; if chips are failing due to the soldering process that costs Intel money
Yea sorry about that, the Intel technology journal sort of waffles on a bit and used parts of the original research that i was having problems finding at the time, this was is the research paper i wanted/should have referenced (PDF).
Yea i get that feeling too, being generous I'd say they realised soldering smaller dies was becoming problematic and instead of having two separate processes they just made the switch to grease across the whole lineup, i guess that could be seen as financially driven or just accepting the inevitable.
Last edited by Corky34; 31-05-2017 at 05:59 PM.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
Reply With Quote