|
data from MSI's OC workshop:        
|
#
?
Oct 22, 2018 20:58
|
|
|
|
| # ? Apr 24, 2024 09:32 |
|
|
evilweasel posted:yeah but that's worse in every respect than the relevant 14nm chip and seems to have been shoved out for no other purpose than to claim that it's in production
|
|
#
?
Oct 22, 2018 22:54
|
|
|
Oh wow, is Charlie predicting doom and gloom for Intel? Wow how totally unprecedented, better take this solid info and base my entire worldview around it.
|
|
#
?
Oct 22, 2018 23:06
|
|
|
Zorak of Michigan posted:Can anyone explain why Intel would just throw the entire process out? I have been part of enough computer screwups to believe that they might have to roll back six months worth of hacks and try something new to get yields up, or whatever, but to get to a point where they would need a clean sheet of paper seems really odd. My understanding is that they haven't been able to get the integrated graphics to work, like, at all. I don't know if it has additional problems, but I'm guessing so, because all they've put out is a low-speed laptop CPU without integrated graphics, which sounds pretty much worthless to me. At some point, if they internally believe that it'll take them as long to get 10nm working as it would 7nm, of they'd be better off ditching 10nm and moving to 7nm. The market was also in a pretty different place when they made their plans for 10nm.
|
|
#
?
Oct 23, 2018 00:12
|
|
|
The 8114Y has popped up a few times on benchmarking databases with a working gpu. Also their 10 nm is the equivalent of TSMC 7 nm. Their 7 nm would be all in on EUV, which makes it slated for like 2021+.
Winks fucked around with this message at 00:21 on Oct 23, 2018 |
|
#
?
Oct 23, 2018 00:18
|
|
|
MSI engineer discusses 9900K 115C Tj_max 
|
|
#
?
Oct 23, 2018 03:09
|
|
|
12. godlike is powerful. very powerful indeed.
|
|
#
?
Oct 23, 2018 03:40
|
|
|
Paul MaudDib posted:MSI engineer discusses 9900K How has this not been talked about?!!
|
|
#
?
Oct 23, 2018 03:42
|
|
|
That's awesome! Maybe that's the next thing in hardware, surviving higher temperatures. This would allow you to cool more easily, as the higher the temperature differential the more efficient the heat transfer. (This is correct physics, right?) Note: did not read tfa
|
|
#
?
Oct 23, 2018 04:01
|
|
|
It is good they released the 9900K not long before winter! Hope it’s a cold one this year!
|
|
#
?
Oct 23, 2018 04:04
|
|
|
LRADIKAL posted:That's awesome! Maybe that's the next thing in hardware, surviving higher temperatures. This would allow you to cool more easily, as the higher the temperature differential the more efficient the heat transfer. (This is correct physics, right?) High temps means you start getting inconsistent behavior, i.e. errors. We aren't going to see radical changes in this area until we have a materials breakthrough. Right now it's much easier to improve on the architecture efficiency than trying to squeeze a little more headroom out of the materials, not that they aren't trying. Plus whatever improvements they come up with still need to be commercially viable.
|
|
#
?
Oct 23, 2018 04:16
|
|
|
priznat posted:It is good they released the 9900K not long before winter! Hope it’s a cold one this year! A 9900k + 2080 Ti would make a fantastic heater. Stand in a fire in a game and feel like how it actually is to be in the fire. I also totally plan to have this combo once 9900k's are readily available.
|
|
#
?
Oct 23, 2018 04:17
|
|
|
Beautiful Ninja posted:A 9900k + 2080 Ti would make a fantastic heater. Stand in a fire in a game and feel like how it actually is to be in the fire. I also totally plan to have this combo once 9900k's are readily available. You don't need a $1500 computer to light yourself on fire
|
|
#
?
Oct 23, 2018 04:56
|
|
|
mewse posted:You don't need a $1500 computer to light yourself on fire Add about $1250 onto that and you'll be close.
|
|
#
?
Oct 23, 2018 05:39
|
|
|
So since I don't want to move up into a new board and cpu. I was thinking of getting this for my 8700k. Should I do it? Or is this just a waste of time and energy? https://www.rockitcool.com/products/copper-upgrade-kit-for-1151 Its currently on an AIO right now with the stock IHS.
|
|
#
?
Oct 23, 2018 06:20
|
|
Have you seen my apex seals? I seem to have lost them.
|
Aeka 2.0 posted:So since I don't want to move up into a new board and cpu. I was thinking of getting this for my 8700k. Should I do it? Or is this just a waste of time and energy? Well, I'd advise against using Conductonaut on bare copper. The nickel plating is what insulates it from damage, even though it doesn't attack copper as badly as it does aluminum. Steve at GN did a video about Conductonaut's effects on different metals - and as I recall, it stains copper pretty readily, but I'd be worried more about long-term degradation. BIG HEADLINE fucked around with this message at 06:34 on Oct 23, 2018 |
|
#
?
Oct 23, 2018 06:28
|
|
|
BIG HEADLINE posted:Well, I'd advise against using Conductonaut on bare copper. The nickel plating is what insulates it from damage, even though it doesn't attack copper as badly as it does aluminum. Steve at GN did a video about Conductonaut's effects on different metals - and as I recall, it stains copper pretty readily, but I'd be worried more about long-term degradation. So using the stock IHS with LM is the best current solution for long term use?
|
|
#
?
Oct 23, 2018 06:45
|
|
|
Aeka 2.0 posted:So using the stock IHS with LM is the best current solution for long term use? To my understanding, yes.
|
|
#
?
Oct 23, 2018 06:48
|
|
|
LRADIKAL posted:That's awesome! Maybe that's the next thing in hardware, surviving higher temperatures. This would allow you to cool more easily, as the higher the temperature differential the more efficient the heat transfer. (This is correct physics, right?) It is. What is also correct physics is that 115C isn't all that hot. There are manufacturing steps which bake wafers at temps of like 1000C or more, fyi. The main reason chips fail at high temps is actually package failure, not silicon. If they engineer the package for higher temps (e.g. use only high melting point solders, make sure the organic substrate materials are fine at a higher temp than normal, pay attention to differential thermal expansion strains, etc), hotspots of 115C should be fine.
|
|
#
?
Oct 23, 2018 07:54
|
|
|
.
sincx fucked around with this message at 05:55 on Mar 23, 2021 |
|
#
?
Oct 23, 2018 08:24
|
|
|
Temperature also increases leakage, so you end up having to dissipate more power.
|
|
#
?
Oct 23, 2018 10:13
|
|
|
Paul MaudDib posted:MSI engineer discusses 9900K I mean, he pretty much sums up the entire 9900k's state of existence right there. "Be rich and buy a binned one, accept that it's gonna melt your poo poo, or gtfo you poor."
|
|
#
?
Oct 23, 2018 14:25
|
|
|
I for one am glad I get to re-live the dying days of Netburst all over again
|
|
#
?
Oct 23, 2018 14:49
|
|
|
BangersInMyKnickers posted:I for one am glad I get to re-live the dying days of Netburst all over again  it's a wild ride
|
|
#
?
Oct 23, 2018 16:16
|
|
I CANNOT EJACULATE WITHOUT SEEING NATIVE AMERICANS BRUTALISED!
|
suck my woke dick posted:
Hell yes. Where’s my 9900k amazon
|
|
#
?
Oct 23, 2018 16:43
|
|
|
rage-saq posted:Hell yes. Silicon Lottery is looking into services to provide for the 9900k, and they seem to be thinking that at least delidding will still be on the menu, and that results in the greatest temp drop (something like 7 C as opposed to 2 C from lapping). So, worst case I do without a gaming CPU for a week or two while I send it off and have them do the thing. As an aside, though - I thought I saw someone say in this thread (edit: lol like a few posts above this one, good work brain) that if you delid you need to re-up the TIM every two or three years? Did I hallucinate that, and if I didn't, is that the case? If the latter, I think I'd just rather keep a lower OC than have to dick around with that every couple years. Gyrotica fucked around with this message at 17:19 on Oct 23, 2018 |
|
#
?
Oct 23, 2018 16:50
|
|
|
I've seen 20+ year old TIM, it still works, but works WAY better when it's fresher. Like saving 10 degrees from something newer. It's a very gradual process though so it's not going to fail one day, just like gain a degree over a year.
|
|
#
?
Oct 23, 2018 16:59
|
|
|
BIG HEADLINE posted:Well, I'd advise against using Conductonaut on bare copper. The nickel plating is what insulates it from damage, even though it doesn't attack copper as badly as it does aluminum. Steve at GN did a video about Conductonaut's effects on different metals - and as I recall, it stains copper pretty readily, but I'd be worried more about long-term degradation. Gallium (or galinstan) will alloy with copper, but it shouldn't invade and destroy it like aluminum. You'll end up with an alloy layer on the copper that won't scrub off, but it shouldn't cause any problems. seriously though, keep it the gently caress away from aluminum. one drop will completely destroy any aluminum part. gallium don't give a poo poo. gallium nasty.
|
|
#
?
Oct 23, 2018 17:02
|
|
|
Dr. Fishopolis posted:seriously though, keep it the gently caress away from aluminum. one drop will completely destroy any aluminum part. gallium don't give a poo poo. gallium nasty. Yup. If anyone hasn't seen the effects, here's a relevant clip...
|
|
#
?
Oct 23, 2018 17:27
|
|
|
I did direct die cooling using a copper waterblock and Conductonaut. The block has a silver, die-shaped, stain on the contact plate but it is otherwise perfectly fine. It was absolutely impossible to clean, it is clearly permeated into the copper and not just on the surface.
|
|
#
?
Oct 23, 2018 17:49
|
|
|
You can get automotive grade chips that are designed to run at 125C but those are typically a lot lower power and cost chips than a desktop CPU so I'm not sure if it could work there as well.
MaxxBot fucked around with this message at 20:40 on Oct 23, 2018 |
|
#
?
Oct 23, 2018 18:40
|
|
|
Intels 9900K page says max temp is 100c. What the gently caress is going on here.
|
|
#
?
Oct 23, 2018 20:04
|
|
|
redeyes posted:Intels 9900K page says max temp is 100c. What the gently caress is going on here. Based on context ("core temperatures may vary by 10C under normal conditions, and up to 20C is OK") maybe the cores themselves can run at up to 115C if the package doesn't exceed 100C? MSI Guy may be wrong, but his screen does show his chip running at 109C. edit: TweakTown says it's an option in the BIOS Paul MaudDib fucked around with this message at 20:12 on Oct 23, 2018 |
|
#
?
Oct 23, 2018 20:08
|
|
|
The part I don't get is isn't TJmax set on the processor itself? Can you somehow set this on the motherboard?
|
|
#
?
Oct 23, 2018 20:12
|
|
|
You could override it even way back in the Conroe days. I remember setting mine badly with my Q6800 and needing to look up what the default was because the motherboard only had you key in values, it didn't have an obvious auto setting to use the CPUs recommendation.
|
|
#
?
Oct 23, 2018 20:23
|
|
|
ohhh, I never noticed a setting. I'm going to look and up my tjmax to like 110c on my 8700k. It it melts so be it.
|
|
#
?
Oct 23, 2018 20:29
|
|
|
I remember Postin’ Paul saying that the 9900K was to be the fastest gaming chip for the next 5 or so years. Something about having to switch to CCX architecture and latency getting worse over the next few node shrinks. If that’s still true given the benchmarks we now have, does that make the price tag a bit easier of a pill to swallow?
|
|
#
?
Oct 24, 2018 01:59
|
|
|
I'll tell you in 5 years.
|
|
#
?
Oct 24, 2018 02:03
|
|
|
The 9900k is back to $499 on Amazon. Perhaps this says something about stock levels? Like when the stock actually starts showing up I mean. This also means for anyone that preordered above $499, this should be the 'pre-order guarantee' price.
|
|
#
?
Oct 24, 2018 02:15
|
|
|
|
| # ? Apr 24, 2024 09:32 |
|
|
Siets posted:I remember Postin’ Paul saying that the 9900K was to be the fastest gaming chip for the next 5 or so years. Something about having to switch to CCX architecture and latency getting worse over the next few node shrinks. If that’s still true given the benchmarks we now have, does that make the price tag a bit easier of a pill to swallow? Pure self indulgent speculation.
|
|
#
?
Oct 24, 2018 02:19
|
|