Maximum PC

I have been looking for (and not seeing) the safe operating temperature range
for the core i7-3960X CPU. The Intel site only states 66.8C for "Tcase".
What is "Tcase" ? (something else I looked for and did not find a clear answer.)

I'm using the Intel RTS2011 LC liquid cooler, which I attached using Arctic Silver 5.
MoBo: Asus P9X79 Deluxe.

My CPU has been running between 115-120 F (about 45-50 C) -- via Asus AI temperature recorder -- playing Deus Ex: HR.
I also have two Sapphire Radeon HD 7950 cards in crossfire running.

I'm not interested in seeing how far I can push this rig, so no over clocking is planned; I just want it to run easily for a long time.
Part of my reason for getting monster hardware is so that I don't have to over clock it to get good performance.
(I know: What a Wuss )

(JBaz--- same reason I never put after market pipes, timing wheels, hot cams, jet kits, etc. etc. on my Fat Lady )

Having no aftermarket pipes on a motorcycle is essentially riding around on a scooter... Sounds pittiful... Although, you could be like me and just run MotoGP pipes (aka none) and be right around the 141 dB, waking up the dead noise. MMMM love the loudness. Hear me rolling 3 miles away and breaking all kinds of sound ordinance at 3 in the morning. lol


Anyhow...

TCase = temperature at center of Heat spreader or center of surface of core. Both Intel and AMD suggest TCase temperature reading for max allowable operating temp for their processors.

It's a kinda of a big woosy topic since there isn't a sensor on your head spreader. For your processor, it should be fine to around 91C for your core temps according to xbitlabs, before the cpu will enable thermal throttling.

Problem with monkeying around with pipes is you screw up the air flow dynamics thru the whole motor,
& can lose power and gas mileage, foul your plugs and burn your valves.

Guy I know put Cobra pipes on his Valkyrie, looked & sounded REEEELLLY KOOOL but he lost a lot of power.
Another guy put aftermarket pipes on his Suzuki 1100 (forget the exact model, but pretty much the standard UJM motor),
and immediately had power and mileage problems, also trouble starting the bike. So then he did a jet kit, and had more problems.
So then he did surgery on this air box, and things just got worse. Kept following bad advice.
Bike was loud as hell (not a cool-sounding loud either), got lousy mileage and had piss-poor
power. In the end he put the bike back into stock config, kissed about $1000 good-bye, & it ran fine again.
At Valkyrie rallies I've been to, there is a minor industry of gimmicks to make the bikes sound cooler or "get more power" (yeah, right.).
Too many shade-tree Bubba mechanics think they can second-guess R&D teams of professional engineers.
My take is: I find a stock bike that looks, sounds and runs the way I want; and it will come optimized by people a lot smarter than me, too.

Sure... if your stupid enough to not run a tuning device on your bike...

That's just idiotic. Even on my stock R6, I run a tuner because its a nice and cheap $180 usb device that evens out your power, specially in the low band once you get your mapping down right. I'm lucky since I know a few owners in my area that have a dyno so I can make sure I'm running the right fuel/air ratios across the board.

The downside is that with either of my r6, the power band is very aggressive since it has 5 valves per cylinder; the top end acts like a turbo and if I'm not careful (or if I was and just wanted to...) I can fish tale the rear right around 10k if I just give it gas in corners. Sometimes I can hold the drift, but most of the time, I just back it off so I can get grip again; not risk coming out of the turn with a tank slapper. But again, on the track, I'm never below 9k rpm's or the very start of the top end powerband.

A properly tuned sports bike is what you want. Although I do want to get a better tunning package that features more sensors and customization (specially the ones with quick shifting packages), but I can't justify spending $900+ for such a system when I can throw that money into better suspension, but then again, my riding skill isn't good enough to warrant such a suspension. And if I had that kind of money, I'd throw it away on some magnesium wheels.

Honestly, I just I'd use that 900 to afford more track days cause they both are just sitting in my garage taking up space.



And honestly, putting a jet kit on any carb bike is a worthless adventure. Did it on my F2 and took it off immediately. This is why GOD made FI (aka Fuel Injection)... lol screw all of that old crap. Give me a usb tuner and a laptop and I'm all good.

Carburetors are soooo 20th century

BTW, do you happen to know if/how well the Antec "TwoCool" stock case fans (Antec P280 case) perform with
Asus AI Fan Xpert+ ? I moved one from its stock connection (case fan power hub) to a fan
header on the Asus P9X79 mobo, and I noticed that it runs a couple hundred rpm slower than the CPU fan (Intel LC) and
the new Cougar fan I just installed (both of which are presumably responding the Fan Xpert, but it's not clear
if the TwoCool fan is, or if it is, how well. I have not tried to manually tweak individual fans yet). This particular fan (TwoCool) also has
a "H/L" switch, which appears mis-wired -- it runs fast when switched to "L".

I assume the TwoCool fan is not as sophisticated as the Cougar, but as Sgt. Schultz would say:

"Nossink. I. Know. Nossink."

The Antec TwoCool 140's run at a crappy 33.6 CFM @ 1200rpm. Well they aren't bad for stock coolers honestly, but my 10yr old Antec 80mm fans push more CFM even though they are a bit louder.

Edit, nvm, they do 58.9 CFM @ 1200rpm

I'm thinking about getting another Cougar to replace that TwoCool.

Ok... I'm like out of my mind right now, can't even search correctly... I donno what I'm thinking. You have 120's. It does 21.3 CFM at 600rpm and 42.6 CFM at 1200rpm. Still, my 80mm Antec's put around 40CFM.


Honestly, you should just do away with fans and cool the whole thing with a super charger... lol

I believe the maximum operating temperature allowed is 100C, or 115C. I forget. If thermal safeguards are enabled though, the processor (at least, on the C2D and Ci3 I used) seemed to start doing things at 90C. I kind of want to say for a high-end part, Intel designed this thing to run to its MTFR at about 40C.

What is MTFR? You mean MTF rate?

Anyhow, Intel kept the thermal limit of SB-E chips a bit ambiguous compared to the normal SB chips which are suppose to have a max thermal temp of 100C. But it doesn't matter because Intel sets the thermal throttling at 91C. Even running through their manual, they don't say it besides the TCase max.

Can't you force throttling off? Forget Intel. They just designed it, we overclock it.

Their manual says that you can, but they put so much warnings through their documentation in every instance that relates to anything about "thermal" or "protection" as a sticky note. Who knew the owners manual for a 3930k or 3960k manual would be 80+ pages!?!? lol

Even then, if you are running liquid, you won't reach that kind of temps anyhow. You'll run into stability issues with voltage first.

What about just shearing off the sides of the case and sinking the rig in a vat of oil ?

You'd still need to have appropriate cooling apparatus' installed, but that would then mean that you'd need to upgrade to a proper liquid cooling loop using a dedicated pump with cooling blocks on the cpu, gpu's and chipset. If not, the oil around those area's will get too hot and because of the viscosity of the oil (still fluid dynamics is a problem with air or oil), it will stay and hover so you need to move the hot oil out and replace it with cooled oil.

Hydrocarbons like oil expand considerably as they heat up, so convective circulation of the oil could be effective without a pump.

They still don't radiate or move out fast enough and its still in a very random direction unless you have channels built in to help guild the flow of the oil in a particular direction without a pump. But even then, the first CRAY super computer that was oil cooled used massive pumps to circulate the oil around... it's also the reason why you have oil pumps in motor engines. (yes yes, they have water to cool them down, but a number of 911 engines are still oil cooled).

You could in theory use normal heatsinks, but you still need a way to move oil in and out to cool them. A normal fan that is used to cool in air gets very inefficient at pushing a heavy liquid like oil around; it can stress it so much that it will probably die faster. If you did this, then you also have no way of removing the heat from the vat; still need some sort of radiator to expel the heat generated by the components.

I mean the guys who built those reactor gaming rigs (Hardcore Computers) went through the same challenges and basically said in an interview that they had to use a liquid cooling loop to move oil around efficiently and remove heat from the entire system, while at the same time limit the exposure of heat on heat sensitive components surrounding the CPU and GPU's. The whole case becomes the reservoir.

And then there are the oil leaks, which would be inevitable.
In my experience, when you are dealing with liquids, it's not a matter
of "IF" there will be leaks/spills, but "WHEN".

If you use quality parts and build it correctly, you won't have any issues. Leaks may occur if you are dealing with higher than normal pressure, but in a cooling loop, most good quality parts can handle it. Blocks can be rated to 50 or 70+ psi, compression fittings could probably take a beating too, but even then, a simple water pump for a computer won't even hit 5-10 psi of force... unless something went wrong and there's more resistance than normal like a blockage or a kink in a hose.

Honestly, most of the time, any kind of leak is going to be associated to a bad gasket, which tends to be a result from non-standard/inferior parts.

Rule # 1: There is no such thing as a "leak/spill-proof" system.
Rule # 2: If you have a leak/spill-proof system, see Rule # 1.

Rule #3: Read directions, mitigate your risk...

And I did not say anything about "leak/spill-proof", I only attribute that lower quality components will have a higher probability to cause issues or problems.