Page 3 of 4 FirstFirst 1234 LastLast
Results 33 to 48 of 49

Thread: Intel Core i9-7900X breaks several benchmark world records

  1. #33
    Gentoo Ricer
    Join Date
    Jan 2005
    Location
    Galway
    Posts
    11,048
    Thanks
    1,016
    Thanked
    944 times in 704 posts
    • aidanjt's system
      • Motherboard:
      • Asus Strix Z370-G
      • CPU:
      • Intel i7-8700K
      • Memory:
      • 2x8GB Corsiar LPX 3000C15
      • Storage:
      • 500GB Samsung 960 EVO
      • Graphics card(s):
      • EVGA GTX 970 SC ACX 2.0
      • PSU:
      • EVGA G3 750W
      • Case:
      • Fractal Design Define C Mini
      • Operating System:
      • Windows 10 Pro
      • Monitor(s):
      • Asus MG279Q
      • Internet:
      • 240mbps Virgin Cable

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by Corky34 View Post
    I think you maybe exaggerating a little.
    Not a single bit. You can measure with arbitrary precision the exact thermal point semi-conductors cross from stable to unstable in its operation. Obviously you can reach a precision threshold where quantum mechanical effects become a larger factor than temperature. But for most of the part a fraction of a degree can make all the difference between stable and unstable, and ideally you want to minimise that thermal factor as much as is reasonably possible. Using a non-crappy TIM in a multi-thousand pound CPU isn't a big ask since it'll only add pennies to their costs.

    Quote Originally Posted by Corky34 View Post
    Lets approach this from the other end, what would you consider a high quality TIM, what TIM do you think Intel should be using?
    I don't care which, as long as it's reasonably chemically stable and minimises thermal impedance. Clearly that's not mayonnaise, or the TIM Intel is using.
    Quote Originally Posted by Agent View Post
    ...every time Creative bring out a new card range their advertising makes it sound like they have discovered a way to insert a thousand Chuck Norris super dwarfs in your ears...

  2. #34
    Senior Member
    Join Date
    Dec 2013
    Posts
    3,526
    Thanks
    504
    Thanked
    468 times in 326 posts

    Re: Intel Core i9-7900X breaks several benchmark world records

    Sure you can measure it but you can't OC to that level of precision where 1 millionth of a degree would matter, so yes you're exaggerating especially if you're OC to such an extent that an increase of a millionth of a degree in temperature causes instability, and please don't throw some buzz words into you claim to make it sound credible, precision thresholds and quantum mechanics have no relationship to temperatures.

    However the question still remains, if you think Intel is using a crappy TIM what TIM would you suggest they use, you have thousands to choose from so name one, if you can't come up with a suggestion then you've rather invalidated your argument as a comparison can't be made between what you consider to be a good and bad TIM.

  3. #35
    Banhammer in peace PeterB kalniel's Avatar
    Join Date
    Aug 2005
    Posts
    30,948
    Thanks
    1,845
    Thanked
    3,357 times in 2,698 posts
    • kalniel's system
      • Motherboard:
      • Gigabyte Z390 Aorus Ultra
      • CPU:
      • Intel i9 9900k
      • Memory:
      • 32GB DDR4 3200 CL16
      • Storage:
      • 1TB Samsung 970Evo+ NVMe
      • Graphics card(s):
      • nVidia GTX 1060 6GB
      • PSU:
      • Seasonic 600W
      • Case:
      • Cooler Master HAF 912
      • Operating System:
      • Win 10 Pro x64
      • Monitor(s):
      • Dell S2721DGF
      • Internet:
      • rubbish

    Re: Intel Core i9-7900X breaks several benchmark world records

    So doesn't it just come down to the fact Intel don't have enough competition still, so don't need to push their chips, so don't need to use better TIM than they currently do?

  4. #36
    Senior Member
    Join Date
    Dec 2013
    Posts
    3,526
    Thanks
    504
    Thanked
    468 times in 326 posts

    Re: Intel Core i9-7900X breaks several benchmark world records

    IDK I'm still waiting for someone to define what a better TIM is.

    Besides, and I've not had time to verify this, from what i understand AMD use exactly the same TIM in some of their APU's so if it was a competition thing and wanting to push chips then you'd expect AMD to be using a supposedly better quality TIM than Intel and their not.

  5. #37
    Gentoo Ricer
    Join Date
    Jan 2005
    Location
    Galway
    Posts
    11,048
    Thanks
    1,016
    Thanked
    944 times in 704 posts
    • aidanjt's system
      • Motherboard:
      • Asus Strix Z370-G
      • CPU:
      • Intel i7-8700K
      • Memory:
      • 2x8GB Corsiar LPX 3000C15
      • Storage:
      • 500GB Samsung 960 EVO
      • Graphics card(s):
      • EVGA GTX 970 SC ACX 2.0
      • PSU:
      • EVGA G3 750W
      • Case:
      • Fractal Design Define C Mini
      • Operating System:
      • Windows 10 Pro
      • Monitor(s):
      • Asus MG279Q
      • Internet:
      • 240mbps Virgin Cable

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by Corky34 View Post
    Sure you can measure it but you can't OC to that level of precision where 1 millionth of a degree would matter, so yes you're exaggerating especially if you're OC to such an extent that an increase of a millionth of a degree in temperature causes instability
    What on earth are you talking about? There's a discrete temperature boundary where on one side is stable, and one is unstable, that is, the CPU starts producing errors. You do get that, right?

    Quote Originally Posted by Corky34 View Post
    and please don't throw some buzz words into you claim to make it sound credible, precision thresholds and quantum mechanics have no relationship to temperatures.
    Again, what? You know that semi-conduction is a quantum mechanical effect, right? And that there's a direct correlation between temperature and semi-conductor stability. Yes? They're not just buzzwords, they're actual scientific terms relevant to the subject matter.

    Quote Originally Posted by Corky34 View Post
    However the question still remains, if you think Intel is using a crappy TIM what TIM would you suggest they use, you have thousands to choose from so name one, if you can't come up with a suggestion then you've rather invalidated your argument as a comparison can't be made between what you consider to be a good and bad TIM.
    Yeah, no. There's no point bickering over specific brands of thermal interface material, that's completely pointless. You have my answer.
    Quote Originally Posted by Agent View Post
    ...every time Creative bring out a new card range their advertising makes it sound like they have discovered a way to insert a thousand Chuck Norris super dwarfs in your ears...

  6. Received thanks from:

    Pleiades (03-06-2017)

  7. #38
    Senior Member
    Join Date
    Dec 2013
    Posts
    3,526
    Thanks
    504
    Thanked
    468 times in 326 posts

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by aidanjt View Post
    What on earth are you talking about? There's a discrete temperature boundary where on one side is stable, and one is unstable, that is, the CPU starts producing errors. You do get that, right?
    Yes and if you're that close to that boundary then you've got more to worry about than what TIM you're using, are you seriously saying that you'd OC a CPU to the extent that a 1 millionth degree change in temperature would cause your system to crash, because if you are how do you deal with someone farting on it an making it a millionth of a degree warmer, how do you deal with someone walking into the room and heating things up with their body temperature?

    Quote Originally Posted by aidanjt View Post
    Again, what? You know that semi-conduction is a quantum mechanical effect, right? And that there's a direct correlation between temperature and semi-conductor stability. Yes? They're not just buzzwords, they're actual scientific terms relevant to the subject matter.
    No it's not, a semiconductor is a material that falls between being a conductor and an insulator, and can be any size you want like 10 micrometers in the 1970 or 800 nanometer in the early 90's, quantum mechanics is the nature at small scales and low energy levels of atoms and subatomic particles, small scales as in under 7 nanometers, quantum tunneling, the thing that stops a semiconductor being both a conductor and insulator has been largely negated by engineering around the problem for now with design and layout changes.

    They're buzzwords because you don't appear to know what either of them are and how they don't relates to what your talking about, as in neither of those things have any bearing on temperatures.

    Quote Originally Posted by aidanjt View Post
    Yeah, no. There's no point bickering over specific brands of thermal interface material, that's completely pointless. You have my answer.
    So in other words despite claiming Intel's using a cheap poor quality TIM you can't come up with a single suggestion for what they should use instead, i think you've proven how fallacious your original statement was.

  8. #39
    root Member DanceswithUnix's Avatar
    Join Date
    Jan 2006
    Location
    In the middle of a core dump
    Posts
    12,676
    Thanks
    749
    Thanked
    1,498 times in 1,263 posts
    • DanceswithUnix's system
      • Motherboard:
      • Asus X470-PRO
      • CPU:
      • 3700X
      • Memory:
      • 32GB 3200MHz ECC
      • Storage:
      • 1TB Linux, 2TB Games (Win 10)
      • Graphics card(s):
      • Asus Strix RX Vega 56
      • PSU:
      • 650W Corsair TX
      • Case:
      • Antec 300
      • Operating System:
      • Fedora 35 + Win 10 Pro 64 (yuk)
      • Monitor(s):
      • Benq XL2730Z 1440p + Iiyama 27" 1440p
      • Internet:
      • Zen 80Mb/20Mb VDSL

    Re: Intel Core i9-7900X breaks several benchmark world records

    Hope the people in this thread have adequate cooling, things seem to be getting rather heated

    I think part of the problem comes down to that 2K price tag, people expect the best when they pay a premium amount.

    If there is a marked improvement in delidding then that implies one of:

    1/ The IHS is a poor conductor: seems very unlikely. it is just a thin cap of metal isn't it?
    2/ The TIM layer is a poor choice, not an expert but assume Intel has plenty of really good experts on this.
    3/ The assembly was poorly executed. In a company like Intel I find that really hard to believe.

    So Corky, I think the reason people are blaming the TIM is that anything else seems impossible. That would seem reasonable, when Intel would think in terms of "good enough to meet the spec sheet of the CPU" but people will think "good enough to meet my expectations for a luxury product". Intel miss-judging a market seems more likely than Intel getting a packaging issue wrong.

    OFC, it could be that the overclocker that reported this in the first place got something wrong, was fed duff pre-production hardware, or some other technical/business/social problem and this is all tripe

  9. Received thanks from:

    Pleiades (03-06-2017)

  10. #40
    Senior Member
    Join Date
    Dec 2013
    Posts
    3,526
    Thanks
    504
    Thanked
    468 times in 326 posts

    Re: Intel Core i9-7900X breaks several benchmark world records

    It's more to do with the gap between the chip and the IHS, IIRC there's a gap of around 0.06mm between them that arguably needs to be there because they need to take into account heatsinks with a high(er) clamping pressure squashing the IHS into the silicon and potentiality cracking the core, the improvements most people see from delidding isn't from changing the TIM it's from a reduction in the gap between the core and the IHS (if that's refitted).

    EDIT: This table of results shows (IMO) why it's not the TIM causing the "problem" it's the gap, in fact it shows the TIM Intel uses is better than Noctua's NT-H1.

    Last edited by Corky34; 02-06-2017 at 11:51 AM. Reason: Adding Info

  11. #41
    Not a good person scaryjim's Avatar
    Join Date
    Jan 2009
    Location
    Gateshead
    Posts
    15,196
    Thanks
    1,231
    Thanked
    2,291 times in 1,874 posts
    • scaryjim's system
      • Motherboard:
      • Dell Inspiron
      • CPU:
      • Core i5 8250U
      • Memory:
      • 2x 4GB DDR4 2666
      • Storage:
      • 128GB M.2 SSD + 1TB HDD
      • Graphics card(s):
      • Radeon R5 230
      • PSU:
      • Battery/Dell brick
      • Case:
      • Dell Inspiron 5570
      • Operating System:
      • Windows 10
      • Monitor(s):
      • 15" 1080p laptop panel

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by DanceswithUnix View Post
    ... Intel miss-judging a market seems more likely than Intel getting a packaging issue wrong. ...
    I think perhaps the issue is with Intel failing to manage an expectation, rather than anything else.

    For instance, I've not heard many complaints from the professional, enterprise or HPC crowd saying their Xeon processors are unstable due to thermal issues. That's a huge market, and if the TIM was actually incapable of doing its basic job you'd think there'd be a big kick up about it.

    OTOH, overclockers got used to being able to run chips way out of spec thanks to the use of Indium solder TIM making a really excellent thermal connection between the die and the IHS. A properly flowed solder TIM should leave basically no air gap at all - just a solid metal connection (albeit formed from multiple layers of different metals).

    If there is any kind of air gap between the die and the IHS then moving from a solder TIM to a paste TIM is going to have a huge impact in extreme and edge cases. Paste TIM works between an IHS and a heatsink because it's filling micro-gaps - it doesn't have to conduct the heat far and it doesn't have to conduct all the heat - there is some physical connection. If the IHS is designed with an air gap to allow for package flex/overclamping then all the heat is going a significant distance through a relatively poor thermal conductor.

    At that point I can't help thinking that some kind of thermal pad would've been a better idea - something slightly flexible that combined a metal foam and some kind of paste carrier. If Corky's right about the current situation it seems like Intel have chosen to put the thermal weak point inside the CPU package, which strikes me as a poor decision...

  12. Received thanks from:

    Pleiades (03-06-2017)

  13. #42
    root Member DanceswithUnix's Avatar
    Join Date
    Jan 2006
    Location
    In the middle of a core dump
    Posts
    12,676
    Thanks
    749
    Thanked
    1,498 times in 1,263 posts
    • DanceswithUnix's system
      • Motherboard:
      • Asus X470-PRO
      • CPU:
      • 3700X
      • Memory:
      • 32GB 3200MHz ECC
      • Storage:
      • 1TB Linux, 2TB Games (Win 10)
      • Graphics card(s):
      • Asus Strix RX Vega 56
      • PSU:
      • 650W Corsair TX
      • Case:
      • Antec 300
      • Operating System:
      • Fedora 35 + Win 10 Pro 64 (yuk)
      • Monitor(s):
      • Benq XL2730Z 1440p + Iiyama 27" 1440p
      • Internet:
      • Zen 80Mb/20Mb VDSL

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by scaryjim View Post
    I think perhaps the issue is with Intel failing to manage an expectation, rather than anything else.

    For instance, I've not heard many complaints from the professional, enterprise or HPC crowd saying their Xeon processors are unstable due to thermal issues. That's a huge market, and if the TIM was actually incapable of doing its basic job you'd think there'd be a big kick up about it.
    That is a very different market. You buy a big heavy box from Dell, you plug it in and you use it with expectation that it is fast and it doesn't break down (ie doesn't get in your way as a professional).

    The enthusiast market is very different, people don't want fast they want the fastest, even if that isn't the brightest idea

  14. #43
    Gentoo Ricer
    Join Date
    Jan 2005
    Location
    Galway
    Posts
    11,048
    Thanks
    1,016
    Thanked
    944 times in 704 posts
    • aidanjt's system
      • Motherboard:
      • Asus Strix Z370-G
      • CPU:
      • Intel i7-8700K
      • Memory:
      • 2x8GB Corsiar LPX 3000C15
      • Storage:
      • 500GB Samsung 960 EVO
      • Graphics card(s):
      • EVGA GTX 970 SC ACX 2.0
      • PSU:
      • EVGA G3 750W
      • Case:
      • Fractal Design Define C Mini
      • Operating System:
      • Windows 10 Pro
      • Monitor(s):
      • Asus MG279Q
      • Internet:
      • 240mbps Virgin Cable

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by Corky34 View Post
    Yes and if you're that close to that boundary then you've got more to worry about than what TIM you're using
    No, you don't. If the TIM is too insular and you swap it for something less insular and it drops a few degrees, that brings you far back away from that boundary and potentially gives you more headroom. The point being, if a millionth of a degree can be the difference between stable and crashing, then a few whole degrees can make a significantly greater difference and ensure much better stability.

    Quote Originally Posted by Corky34 View Post
    are you seriously saying that you'd OC a CPU to the extent that a 1 millionth degree change in temperature would cause your system to crash
    It doesn't have to crash your OS outright to cause instability/dysfunction.

    Quote Originally Posted by Corky34 View Post
    because if you are how do you deal with someone farting on it an making it a millionth of a degree warmer, how do you deal with someone walking into the room and heating things up with their body temperature?
    Now you're just being grossly facetious.

    Quote Originally Posted by Corky34 View Post
    No it's not, a semiconductor is a material that falls between being a conductor and an insulator
    That required QM and solid-state physics to develop. I can't take you seriously if you disregard QM's role in semi-conductors.

    Quote Originally Posted by Corky34 View Post
    quantum mechanics is the nature at small scales and low energy levels of atoms and subatomic particles, small scales as in under 7 nanometers, quantum tunneling
    QM and its applications is much more complex and diverse than you obviously understand.

    Quote Originally Posted by Corky34 View Post
    the thing that stops a semiconductor being both a conductor and insulator has been largely negated by engineering around the problem for now with design and layout changes.
    Solving electrical engineering problems doesn't make QM go away. Factoring in QM in engineering solid state physical systems is an absolute prerequisite, and you can be absolutely certain that Intel's engineers have to incorporate QM knowledge to avoid pitfalls in their designs and testing them.

    Quote Originally Posted by Corky34 View Post
    They're buzzwords because you don't appear to know what either of them are and how they don't relates to what your talking about, as in neither of those things have any bearing on temperatures.
    And you apparently don't understand the history of the development of semi-conductors.

    Quote Originally Posted by Corky34 View Post
    So in other words despite claiming Intel's using a cheap poor quality TIM you can't come up with a single suggestion for what they should use instead, i think you've proven how fallacious your original statement was.
    I'm just not going to let you bate me into a completely pointless pissing match over specific brands of TIM. It doesn't matter which one specifically is used as long as it performs well. It's like claiming Lada's aren't crap, and if you don't answer whether it should be replaced by a Ferrari, Porsche, Lamborghini, Koenigsegg, or a long list of high performance car brands, then Lada's aren't crap! The argument is as nonsensical as it is invalid, stop trying to make it a thing.
    Quote Originally Posted by Agent View Post
    ...every time Creative bring out a new card range their advertising makes it sound like they have discovered a way to insert a thousand Chuck Norris super dwarfs in your ears...

  15. Received thanks from:

    Pleiades (03-06-2017)

  16. #44
    Senior Member
    Join Date
    Dec 2013
    Posts
    3,526
    Thanks
    504
    Thanked
    468 times in 326 posts

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by aidanjt View Post
    No, you don't. If the TIM is too insular and you swap it for something less insular and it drops a few degrees, that brings you far back away from that boundary and potentially gives you more headroom. The point being, if a millionth of a degree can be the difference between stable and crashing, then a few whole degrees can make a significantly greater difference and ensure much better stability.
    You were not talking about a few degrees, you said "1,000,000th of a degree can be the difference between a stable and unstable overclock."

    Quote Originally Posted by aidanjt View Post
    It doesn't have to crash your OS outright to cause instability/dysfunction.
    Semantics.

    Quote Originally Posted by aidanjt View Post
    Now you're just being grossly facetious.
    Yes, yes i am, because while some people are trying to have a sensible discussion you're just posting aggressive posts with the intent of being argumentative,

    Quote Originally Posted by aidanjt View Post
    That required QM and solid-state physics to develop. I can't take you seriously if you disregard QM's role in semi-conductors.
    No it didn't, semiconductors proceeded the discover of quantum mechanics be almost 100 years depending on what part of QM you consider as discovered.

    Quote Originally Posted by aidanjt View Post
    QM and its applications is much more complex and diverse than you obviously understand.
    Do tell me then how QM relates to heat generation.

    Quote Originally Posted by aidanjt View Post
    Solving electrical engineering problems doesn't make QM go away. Factoring in QM in engineering solid state physical systems is an absolute prerequisite, and you can be absolutely certain that Intel's engineers have to incorporate QM knowledge to avoid pitfalls in their designs and testing them.
    No it doesn't and that's not what i said, i said it has been largely negated by engineering around the problem for now with design and layout changes, maybe you need to look up the definition of 'negated'

    And no, factoring in QM in engineering solid state physical systems is not an absolute prerequisite as QM only effect the world of the very small, very small as in less than 10-7nm, but then again if you knew what you were talking about you'd know that.

    Quote Originally Posted by aidanjt View Post
    And you apparently don't understand the history of the development of semi-conductors.
    Says the guy who doesn't know the first semiconductors (discovered in 1833) proceeded the discover of quantum mechanics (1859 – Kirchhoff introduces the concept of a blackbody, to 1942 – J. Robert Oppenheimer predicts quantum tunneling) by almost 100 years, at least do people the favor of a minimal amount of research before spouting spurious claims.

    Quote Originally Posted by aidanjt View Post
    I'm just not going to let you bate me into a completely pointless pissing match over specific brands of TIM. It doesn't matter which one specifically is used as long as it performs well. It's like claiming Lada's aren't crap, and if you don't answer whether it should be replaced by a Ferrari, Porsche, Lamborghini, Koenigsegg, or a long list of high performance car brands, then Lada's aren't crap! The argument is as nonsensical as it is invalid, stop trying to make it a thing.
    I'm not baiting you into anything, you started this by making a spurious claim and you're continuing to dig a hole because it's impossible for you to backup your claim with any evidence, instead you're choosing to sow discord rather than letting people have an adult discussion about a subject, the only person making this a thing is you with your insistence that Intel is using a crappy cheap TIM.

  17. #45
    root Member DanceswithUnix's Avatar
    Join Date
    Jan 2006
    Location
    In the middle of a core dump
    Posts
    12,676
    Thanks
    749
    Thanked
    1,498 times in 1,263 posts
    • DanceswithUnix's system
      • Motherboard:
      • Asus X470-PRO
      • CPU:
      • 3700X
      • Memory:
      • 32GB 3200MHz ECC
      • Storage:
      • 1TB Linux, 2TB Games (Win 10)
      • Graphics card(s):
      • Asus Strix RX Vega 56
      • PSU:
      • 650W Corsair TX
      • Case:
      • Antec 300
      • Operating System:
      • Fedora 35 + Win 10 Pro 64 (yuk)
      • Monitor(s):
      • Benq XL2730Z 1440p + Iiyama 27" 1440p
      • Internet:
      • Zen 80Mb/20Mb VDSL

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by scaryjim View Post
    If Corky's right about the current situation it seems like Intel have chosen to put the thermal weak point inside the CPU package, which strikes me as a poor decision...
    Indeed, that seems an almost laughable choice if true.

    Could this be a consequence of using LGA rather than pins? Another technique widely thought to be just for profit maximising by pushing costs onto motherboard manufacturers.
    Last edited by DanceswithUnix; 02-06-2017 at 12:50 PM. Reason: remove accidental text paste

  18. #46
    Not a good person scaryjim's Avatar
    Join Date
    Jan 2009
    Location
    Gateshead
    Posts
    15,196
    Thanks
    1,231
    Thanked
    2,291 times in 1,874 posts
    • scaryjim's system
      • Motherboard:
      • Dell Inspiron
      • CPU:
      • Core i5 8250U
      • Memory:
      • 2x 4GB DDR4 2666
      • Storage:
      • 128GB M.2 SSD + 1TB HDD
      • Graphics card(s):
      • Radeon R5 230
      • PSU:
      • Battery/Dell brick
      • Case:
      • Dell Inspiron 5570
      • Operating System:
      • Windows 10
      • Monitor(s):
      • 15" 1080p laptop panel

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by DanceswithUnix View Post
    ... Could this be a consequence of using LGA rather than pins? ...
    I wondered about that too. Of course, AMD have used LGA on their larger server chips for a considerable number of years, so I assume it must provide some benefit as packages get larger (can you use finer-pitched contacts for LGA?), but they've stuck to pins with AM4. I guess with pins your processor sits in a plastic block that must provide reasonable mechanical support, while LGA almost floats on a bed of pins - I can see how it would deform easier...

  19. #47
    Senior Member
    Join Date
    May 2009
    Location
    Where you are not
    Posts
    1,279
    Thanks
    569
    Thanked
    94 times in 83 posts
    • Iota's system
      • Motherboard:
      • Asus Maximus Hero XI
      • CPU:
      • Intel Core i9 9900KF
      • Memory:
      • CMD32GX4M2C3200C16
      • Storage:
      • 1 x 250GB / 1 x 1TB / 1 x 2TB Samsung 970 Evo Plus NVMe
      • Graphics card(s):
      • Nvidia RTX 3090 Founders Edition
      • PSU:
      • Corsair HX1200i
      • Case:
      • Corsair Obsidian 500D
      • Operating System:
      • Windows 10 Pro 64-bit
      • Monitor(s):
      • Samsung Odyssey G9
      • Internet:
      • 40Mbps SKY Fibre

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by Corky34 View Post
    [URL="http://forums.hexus.net/hexus-news/374932-intel-core-i9-7900x-breaks-several-benchmark-world-records-2.html#post3815872"]Also Delta T = differential temperature
    I'm well aware of what Delta T means Corky, perhaps you should learn to differentiate between the parts that pertain to the issue with the TIM applied beneath the IHS?


    Quote Originally Posted by Corky34 View Post
    If you had bothered to read the research paper on Indium Solder failure then you'd perhaps know that there is no "mounting pressure to the IHS" and that thermal cycling is exactly what causes Indium Solder to fail in the first place, kind of obvious when you think about it.
    If you had bothered to read the original Intel research and not the third party research, you would have realised that mounting pressure has a part to play in failure of the TIM applied underneath the IHS, particularly any shearing effects applied (which happens when coolers move over time).


    Quote Originally Posted by Corky34 View Post
    As was pointed out above, that's exactly what you said, here let me quote you...
    ...poorly applied.... Wait, you did read that part didn't you? Which part of the post did you have difficulty understanding? Perhaps I can clarify it for you?

    Quote Originally Posted by Corky34 View Post
    Is that not you questioning why Intel moved to thermal grease? because it sure seem that way from where I'm sitting.
    Hey, what was my original post?

    Quote Originally Posted by Iota View Post
    For the prices they are charging, you would have thought they would be soldered.
    I'll go around in circles, just to clarify. I'm assuming you can follow what my intention was here, without trying to twist it to suit your own purpose, perhaps I should type things out in full in future just to satisfy you so you can't read into things to suit your own arguements (moot as they are)?

    Intel, which has spent upwards of $10 Billion on R&D each year since 2012 and has a gross margin percentage averaging around the 60% mark could have opted to solve the issue of poorly applied TIM underneath the IHS (resulting in issues from Ivy Bridge onwards with temperatures for processors that are marketed to be overclocked (K designation)). Regardless of the type of TIM utilised by Intel, be it some variation of solder or some variation of epoxy, they could have chosen to do this. They reserved using solder in prior generations of CPU (-E designations) for the exact same architectures as they chose to use epoxy in the K designation CPUs for the very same prior generations.

    If you are suggesting that Intel didn't have the resources to resolve the issues of Indium Solder TIM not scaling well with a process shrink from SB onwards without simply resorting to a poorly applied epoxy based TIM? You only have to Google the results users have had delidding and re-applying a different TIM properly. Yes, absolutely you can argue with the smaller die sizes that you'll gain an increase in temperatures (fairly relative considering less power is used), if the epoxy TIM alone wasn't providing adequate heat dissipation to the IHS, perhaps change the material of the IHS to compliment better heat dissipation. There are so many options that have been available and yet they haven't been taken.

    Quote Originally Posted by Corky34 View Post
    Also I'm not disagreeing with you, I'm questioning your understanding of the subject you're talking about, I'm attempting to get to the bottom of why you think the TIM Intel (and AMD BTW) use is not the best option available to them and is cheap.
    If users have delidded and reapplied using a different TIM and got much lower temps, clearly there is an issue with the processor package thermal solution. Be that the TIM being used, how that TIM is being applied or the other materials used in parts of the processor packing.

    Quote Originally Posted by Corky34 View Post
    And purely technical limitations take a little more than just willpower to overcome.
    R&D budget? They have one, I'm pretty sure it's a decent sum available to them https://s21.q4cdn.com/600692695/file...ual_Report.pdf

    Quote Originally Posted by Corky34 View Post
    You're assuming it's the supposedly poorly performing TIM that the cause of the problem when that's yet to be proven in any meaningful way.
    Of course it could be the process itself that is the root cause of the problem.

    Quote Originally Posted by Corky34 View Post
    Like i said if you think it's such a cheap poorly performing TIM what would you have them use instead?
    A processor packaging that is suited to the task at hand?

    Quote Originally Posted by Corky34 View Post
    Don't you think it's a little ironic to be referencing me back to the very same Intel technology journal that i posted?
    No, not when you clearly haven't read it.

    Quote Originally Posted by Corky34 View Post
    Perhaps it would be best if you just referenced the particular passage you think supports your opinion.
    Perhaps. Or perhaps, we could end this discussion because it's quite clear you have your own opinion that you are unable to support with evidence (which works both ways before you bother to say it...).

  20. Received thanks from:

    Pleiades (03-06-2017)

  21. #48
    Senior Member
    Join Date
    Dec 2013
    Posts
    3,526
    Thanks
    504
    Thanked
    468 times in 326 posts

    Re: Intel Core i9-7900X breaks several benchmark world records

    Quote Originally Posted by Iota View Post
    I'm well aware of what Delta T means Corky, perhaps you should learn to differentiate between the parts that pertain to the issue with the TIM applied beneath the IHS?
    Then why did you question my use of the term right here when you said "also I see you banging on happily about Delta T. So, Ts (underside) or Tc (topside)?" if that's not you questioning what Delta T refers to? Perhaps you should have clarified WTH you mean when referring to Ts & Tc and not have posed that as a question by ending with a "?"

    Quote Originally Posted by Iota View Post
    If you had bothered to read the original Intel research and not the third party research, you would have realised that mounting pressure has a part to play in failure of the TIM applied underneath the IHS, particularly any shearing effects applied (which happens when coolers move over time).
    As i said after posting the Intel tech journal i was having problems finding the original research so i posted an Intel tech journal that referenced the original JOM (The Journal of The Minerals, Metals & Materials Society) article that i read over a decade ago, also as the article in JOM talks specifically about indium solder there's no "mounting pressure" involved, it's solder, in fact the words "mounting" and "pressure" isn't even mentioned in the article, neither is "shearing".

    Quote Originally Posted by Iota View Post
    ...poorly applied.... Wait, you did read that part didn't you? Which part of the post did you have difficulty understanding? Perhaps I can clarify it for you?
    No you don't need to clarify it for me, however you do need to define WTH you mean by "poorly applied" because that's an entirely subjective statement that you've failed to quantify.

    Quote Originally Posted by Iota View Post
    Hey, what was my original post?
    Are you having problems remembering, do keep up, it was this...
    Quote Originally Posted by Iota View Post
    For the prices they are charging, you would have thought they would be soldered. Another reason for me to buy AMD.
    The same AMD who, BTW, have used exactly the same thermal grease as Intel in some of their APU's

    So yes IMO that is you questioning why Intel moved to thermal grease, call me gullible if you like but i tend to take what people say as what they actually mean.

    Quote Originally Posted by Iota View Post
    I'll go around in circles, just to clarify. I'm assuming you can follow what my intention was here, without trying to twist it to suit your own purpose, perhaps I should type things out in full in future just to satisfy you so you can't read into things to suit your own arguements (moot as they are)?

    Intel, which has spent upwards of $10 Billion on R&D each year since 2012 and has a gross margin percentage averaging around the 60% mark could have opted to solve the issue of poorly applied TIM underneath the IHS (resulting in issues from Ivy Bridge onwards with temperatures for processors that are marketed to be overclocked (K designation)). Regardless of the type of TIM utilised by Intel, be it some variation of solder or some variation of epoxy, they could have chosen to do this. They reserved using solder in prior generations of CPU (-E designations) for the exact same architectures as they chose to use epoxy in the K designation CPUs for the very same prior generations.

    If you are suggesting that Intel didn't have the resources to resolve the issues of Indium Solder TIM not scaling well with a process shrink from SB onwards without simply resorting to a poorly applied epoxy based TIM? You only have to Google the results users have had delidding and re-applying a different TIM properly. Yes, absolutely you can argue with the smaller die sizes that you'll gain an increase in temperatures (fairly relative considering less power is used), if the epoxy TIM alone wasn't providing adequate heat dissipation to the IHS, perhaps change the material of the IHS to compliment better heat dissipation. There are so many options that have been available and yet they haven't been taken.
    No please don't go around in circles, you'll just more confused than you already are.

    And perhaps it would be better for everyone if you did type things out in full in future as expecting people to read your mind and guess what your intentions are isn't a very sensible approach to life.

    Also everything you've said assumes that a poorly applied TIM has a detrimental effect and while that's true when there's not enough TIM it's not when there's to much, like I've been saying the TIM and however you want to define "poorly" has very little effect on the overall effectiveness of the packages ability to transfer heat away from the cores, if anything it's the gap between the silicon and the IHS.

    The removal of that gap is why people who delid their CPUs see an improvement in temperatures, not the replacing of "poorly" applied "cheap" TIM for a better quality one and if you can think of a way for Intel to reduce that gap without the potential that a high clamping force HS is going to crush the IHS into the silicon causing it to crack then I'm sure they'd like to hear from you because they've been spending billions of dollars trying to figure that out.

    Quote Originally Posted by Iota View Post
    If users have delidded and reapplied using a different TIM and got much lower temps, clearly there is an issue with the processor package thermal solution. Be that the TIM being used, how that TIM is being applied or the other materials used in parts of the processor packing.
    And I'm not saying there isn't, I am however saying that issue isn't the TIM and how it's applied.

    Quote Originally Posted by Iota View Post
    R&D budget? They have one, I'm pretty sure it's a decent sum available to them https://s21.q4cdn.com/600692695/file...ual_Report.pdf
    An R&D budget is more than willpower is it not?

    Quote Originally Posted by Iota View Post
    A processor packaging that is suited to the task at hand?
    Like i said if you can come up with one I'm sure Intel would like to hear from you, you'd be rolling in it if you could figure out how to reduce the tolerances between the die's and the IHS without a) increasing the density of the IHS and potentially raising temperatures, and b) making sure the IHS can't be squashed into the rather delicate silicon that makes up the cores and cracking it.

    Quote Originally Posted by Iota View Post
    No, not when you clearly haven't read it.
    It may have been over a decade ago but rest assured i read it.

    Quote Originally Posted by Iota View Post
    Perhaps. Or perhaps, we could end this discussion because it's quite clear you have your own opinion that you are unable to support with evidence (which works both ways before you bother to say it...).
    And yet this thread is replete with posts from me providing the evidence that debunks what seems to be a commonly held belief amongst conspiracy theorists that Intel's only doing this to save a few bucks.
    Last edited by Corky34; 03-06-2017 at 09:25 AM.

Page 3 of 4 FirstFirst 1234 LastLast

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •