I got a different thing I want to verify: I got an I5 2500K, no OC, been running at 100% for all cores for the past 5 minutes or so, stays stable at 60-61 °C by the looks of it.
I don't know what brand of cooler I have, but I'm checking right now
thing is, last time when i dusted off my computer, the thermal paste was dried out, so I just removed it, but didn't replace it because I didn't have any
that was about 3-4 months ago, during the summer break
@JourneymanGeek 15 competing standards - "Let's make one to unify them all" Now 16 competing standards
That is the biggest issue with all those "new and exciting ideas" that try to make our lives better: unless you can get everyone to convert and make conversion easy and painless, you're just going to divide the market even further
since my previous message 15 minutes ago, it has spiked to 62.5 °C, but it's back to 61 now. Still running at 100% on all cores. Core 2 and CPU Package (whatever that is) have spikes at 63, 1 and 3 at 60, 4 at 61. individual core temp is fluctuating between 57 and 61% right now
But there are quite some guides on how to apply thermal paste
All come down to this: 1) The paste does NOT conduct the heat as well as the heat sink. Thus use as little as possible. 2) But do use enough to fill all the tiny air holes which conducts even less heat.
And even a shiny polished piece of metal is not 100% flat and has holes
And considering how clumsy I usually am, I might just leave it like that, considering I'm still 10 °C under what Intel recommends as a safe temp (72 °C) and more than that below the temperature where it starts to throttle itself
Or does TP have another goal besides making heat transfer go better?
It depends on how well things were before and after. A very poor setup before, and a perfectly applied paste one afterwards might make a huge difference. A normal setup as start and a perfect one afterwards should differ a lot less
@NateKerkhofs That's perfectly fine, in fact better than fine.
@NateKerkhofs Now that I have difficulty believing.
@NateKerkhofs You should check your per-core clock speeds and duty cycles. It's impossible to be getting those temperatures under load without thermal paste
@Hennes I'd say err on the side of more. For the inexperienced, too little will be a major issue, too much will only cause minor degradation. Thermal paste may not conduct as well as direct contact with the heatsink, but it conducts hundreds of times better than an air gap.
@NateKerkhofs When you say "Removed the thermal paste" explain what you actually did.
@NateKerkhofs Actually it's very true. Good thermal paste can reduce temperatures up to 20'c vs. bad paste, under extreme conditions, but even crap thermal paste will reduce your temperatures by 50c compared to no thermal paste
@Hennes Yeah, this. The best paste vs. even chocolate can be less than 5'c difference. A site did also test lipstick, toothpaste, and peanut butter as well I believe.
I've used thermal paste in the paste (generic brand) and semi-recently (about 5 years ago. Artic silver on a Nehalem heatsink). Never tried switching paste. Always used the old generic one on my celeron-1's and AMD MP's. Only tested the artic silver on my i7 920. Second try with the i7 the temps dropped by about 4C at idle and 10C at load.
So i guess that unless you are experienced you might have to try several times with test runs in between.
Personally I tend to see no more than about 3-5'c difference, but really getting the "ultimate" thermal paste is primarily an exercise for extreme overclockers whose systems are otherwise on the edge.
If your system is 3-5'c away from overheating, then there's tons of more effective things you can do to counter it.
But again, last time I tried no thermal paste my 920 hit 100'c before I got to the OS
@qasdfdsaq I'll be honest, I THINK I removed the TP. I detached the heatsink and there appeared to be some kind of dry residue on the CPU, but I'm not sure if that was the thermal paste or something else
What might also be possible is that it wasn't thermal paste that I saw, but something like coalesced dust that formed strings or similar in the small crannies between the heatsink and the mobo itself. It's been like 4 months since I last cleaned for dust, so I'm not sure
I'm probably going to have another dedusting in January. Get a can of air, take it outside on a dry day and just blow away all the dust
but I haven't had any major computer issues in the past 4 months since I installed Windows 10, so Don't fix what isn't broken
OK so for anyone that's travelled beyond North America and Western Europe you know what I'm talking about.
Left: Romania, last year. Right: Turkey, last night.
They start popping up in the Balkans and of course they're everywhere by the time you get to Turkey and become totally unavoidable no ...
@JourneymanGeek Haven't used the card squeegee for a decade, but I still cover my CPU :-P
Especially with modern designs that have silicon under most of the heatspreader or are off-centre and elongated and have extremely high heat density. And what's the point of the heat-spreader if half of it can't spread heat?
Still reading that page but the technique I read and used from other sites doesn't seem to be included
Could you please advise to proper router & secondary router setup for my staff building. Our existing connection is as shown on the diagram.
However many of users are facing alternative disconnection, slowdown, IP conflict etc.
So please advise the best solution for routers IP config & WiF...
Hi, The main router and EA 1900 routers are only my control. All other routers are configured by my colleagues themselves; seems different ip address. — HarisNov 1 at 8:41
@Boris_yo From the one of the pages you are linked they suggest that it is a mistranslation and they actually mean that the photo is of a similar kind of product and not the actual product itself:
> Dave’s suggestion that it’s intended to mean “the photograph shows a similar product, but not the specific item being offered” is plausible.
@Boris_yo From your Fomec site: "All goods are in-kind shooting, but because of the display, lighting and other reasons, it is inevitable that there is color different in order to prevail in kind."
Definitely supports the idea that the photos "might not be the exact product you are looking at" due to problems with lighting and display purposes.
@Boris_yo Ѿ bad translation, they mean to say Shooting The Clone :-) on e-bay it is also known as stealing the pictures from someone elses auction (who sells the real item)
all these poor excuses (on amazon and e-bay) that it Doesnt look like the item they pictured, has nothing to do with lighting issues , it really isnt the item they pictured. instead it is something they saved 5Cents on. Bought at wholesale junk factory for super cheap, or have made in china to the wrong specs.
The whiney server: the landlord in anticipation of getting an e-mail sent , pushed the "get all messages" to often. Server sends a message back to knock it off :-) that is new.
More whiney robots telling us what to do, while they demand that we verify We're The Humans. Well We ARE the humans, do your robot job and STFU
You can have your own particular site at home, and I'll let you know precisely how! Yet, it won't not spare you much cash, and it unquestionably won't spare you time. So give it genuine thought before you continue... unless your objective is just to find out about the innovation and have some goo...
For testing purposes I need to generate high CPU load on a Windows Server 2003.
I cannot install any software but have to make do with what Windows provides.
What would be the best way to achieve that?
@A.J.Ruckman most abusive to temp of a standard desktop cpu is small FFT , if something passes that, without throttling, the cooling is completly sufficient.
@A.J.Ruckman then set for "custom" and select to use most of the ram, I find that when the ram is not tweaked perfectally (and would cause tiny corruption) custom use of most of the ram (not yet used) in prime will show a fail. Some say it could also fail anyways, i do not believe it.