Cryorig Announces LGA2066 Socket CPU Cooler Compatibility

Status
Not open for further replies.

derekullo

Distinguished
Didn't the last article say something to the effect, "Don't use air cooling for i9"?

http://www.tomshardware.com/reviews/intel-core-i9-7900x-skylake-x,5092-10.html

"Air cooling is out of the question for Core i9-7900X"


I don't mean to misquote or bend the meaning of what was said, but it's unclear if this statement was meant for running at stock frequencies or for overclocking.

What I'm trying to ask is, "Will my computer hit a thermal shutdown while using an i9 to render a video in handbrake with a Cooler Master Hyper 212 EVO or a similar air-cooled cpu heatsink/fan?"

Or is water cooled the only way to go for the i9 processor line?
 


That was directed at the i9's so 10 cores and up. The thermals to displace are very high if you want to use air you most likely would want t o be running a delided chip and one of the bigger air coolers the Cooler Master Hyper 212 EVO likely can't displace enough heat. Would be nice to see some testing on this matter for a few CPU coolers AIO and air.
 

the nerd 389

Honorable
Nov 20, 2012
1,573
3
12,460
357


So I should simply accept the potentially damaging temperatures and inconsistent performance because my computer turns on?

That's not really okay. If Intel expects consumers to put up with that, the reviews should reflect THAT performance.
 

Danilushka

Prominent
Feb 23, 2017
9
0
510
0
"Then the 7900x happened."
That is what happens when you get used to having no competition and a competitor takes you by surprise: hasty poorly-considered decisions.
 

bit_user

Splendid
Ambassador

First, where did anyone say you should use air cooling for an i9? Just because a cooler fits a given socket shouldn't be taken as a statement that it's suitable for all CPUs that fit in it.

Second, there's no reason the CPU needs to run at "potentially damaging temperatures", no matter what cooler you use. That's what throttling is for. You can configure where it throttles, at least on their unlocked processors.

But it would be pretty dumb to use an inadequate cooler, because you'd waste money on an expensive CPU just to have it throttle a lot and run like a cheaper one (or worse).
 


I do agree about the thermal issue and its really lame they have this and the TIM. On the flip side we are talking CPU's that are $1,000 and up range so its not like someone can't afford better cooling. I'm torn on getting a 7820x myself, would like to see how the thermals are with a delided CPU.
 

the nerd 389

Honorable
Nov 20, 2012
1,573
3
12,460
357


It's true that there are watercoolers that will keep the chip from damaging itself or throttling. Remember, though, that these are also used in video editing rigs and entry level workstations. In those applications, the pump represents a previously unnecessary single point of failure. Basically, you have to decide whether you want a reliable cooler or a reliable CPU.

That's not a decision that I'd expect Intel to force on consumers, but they did. Regarding the use of appropriate coolers, I suppose what I'm trying to say is that there are applications that this product line was well suited to where there is no longer such a thing as an appropriate cooler.

I find that upsetting, particularly as I had a number of builds lined up that will now have to wait until ThreadRipper before they're viable.
 

bit_user

Splendid
Ambassador

Yeah, Intel made a lame decision. Let's hope they course-correct and at least the Xeon version is soldered or uses something like the gel that @aldaia described. I only buy Xeons, anyway. Especially now that even i3's lack ECC support.

But that's not the only reason I'm going to skip Skylake-X. I'll wait until the Kaby or Cannon (Coffee?) Lake i9, and hope it fixes most of the performance regressions, as well as the thermals. Otherwise, it'll probably be Zen+ for me.
 

derekullo

Distinguished


Maybe we are just conditioned, no pun intended, to assume air-cooling works for stock settings due to the last 30+ years of it being the status quo?

I did a little more research on air cooling and found the Maximum TDP of the Cooler Master 212 Hyper Evo is 180 watts, assuming full blast.
https://www.newegg.com/Product/Product.aspx?Item=N82E16835103099

I know it isn't the most powerful air cooler, but it is one of the most popular.

Does a 180 watt tdp dissipated mean anything compared to the 140 watts the cpu claims to produce, obviously at stock?

Does Intel / Cooler Master just make up a number and say this is what we think it can do?

I don't want my $1000 cpu to end up like the Mars Climate Orbiter with a catastrophic failure due to both companies using a different way of measuring.
 

the nerd 389

Honorable
Nov 20, 2012
1,573
3
12,460
357


If the cooler were thermally coupled to the CPU, it could dissipate that much. The problem is that the paste insulates the CPU compared to solder. For that kind of cooling capacity, the cooler manufacturers assume that the CPU has sufficient thermal conductivity to get that heat to the air cooler in the first place.

What Intel has done is introduce a bottleneck in the flow of heat. All ratings published by coolers essentially no longer apply to Intel's product stack as a result.
 

derekullo

Distinguished
Why would Intel skimp on a processor they themself say is designed for the high end user knowing it would generate a lot of heat.

I can see doing that for a Celeron or maybe even i3.
 

bit_user

Splendid
Ambassador

We'll likely never know, because they probably won't publicly disclose the rationale for their decision and the only investigative journalism into such matters is now behind a paywall (semiaccurate.com - not all their stuff is paywalled - just the juicy bits).

So, that leaves us free to speculate and trade wild theories that no one can conclusively prove false.


Hey! Don't go trash-talking the i3! Those used to be great for small servers, until Intel nerfed them by withdrawing ECC support. You can still get it in the Skylake i3's.
 
Tin foil hat time, Intel are secretly letting AMD catch up a bit so they don't get deemed a monopoly. If this is right you will see IceLake move way up in its time table Intel had published prior.

Could be way off base here but maybe not.
 

the nerd 389

Honorable
Nov 20, 2012
1,573
3
12,460
357


It's probably worth bringing up two philosophical razors: Occam's Razor and Newton's Flaming Laser Sword.

(Occam's) The explanation is probably much simpler than that, and
(Newton's) because it can't be settled by any of us, it's not worth debating.

That said, I think we can all agree that third party component manufacturers are likely to take a hit as a result of Intel's design decisions this time around, at least until ThreadRipper is released.
 
Status
Not open for further replies.

ASK THE COMMUNITY

TRENDING THREADS