Sign in with
Sign up | Sign in
Your question

A thread for Arctic Silver 5 lovers.

Last response: in Overclocking
Share
January 10, 2007 9:55:43 AM

I just put some of this lovely grey gunk in between my e6400 and Scythe Ninja and it's already shaved off 3-4C from full load temps, this is before it's even had a chance to set properly. Wonderful.

So anyone else who loves this stuff, reply here!
January 10, 2007 10:33:02 AM

Geesh! You must be a big fan of AS5. Yes, I use it with my Storm block and gpus and chipsets. :|
January 10, 2007 11:13:18 PM

yea I like it, lowered cpu temp by 5 degrees after settling
Related resources
January 10, 2007 11:45:22 PM

XD

Yes, definitely some fine paste. =D
January 11, 2007 8:53:19 AM

Yea, just used some on mumz p4 today, aswell as cleaning the fan, h/s and heatspreader on the cpu, and using AS5, overall at idle is running 15 deg celsius cooler
January 11, 2007 9:54:39 AM

Quote:
Yea, just used some on mumz p4 today, aswell as cleaning the fan, h/s and heatspreader on the cpu, and using AS5, overall at idle is running 15 deg celsius cooler


8O

15 degrees cooler?? I guess the HSF wasn't mounted correctly before, then.
January 11, 2007 10:04:16 AM

This seems a bit over the top.
January 11, 2007 11:41:18 AM

Yeah, it's wonderful. After replacing the stock transfer paste on every sink in my PC, the average temps were around 3~8ºC lower.

Gained another 2-odd degrees by upgrading ALL fans in my system to higher CFM models.
January 11, 2007 12:12:30 PM

For under $10 a tube AS5 is worth it. It is also readily available. And a tube will last quite a while doing 10CPUs or more. What bugs me is that every company seems to be selling high end paste now which digs into artic silvers business. They are copying the originals.

I hate thermaltake cause they copy everyone else. I hate all motherboard manufacturers cause they copy Abit now with overclocking, japanese capacitors, etc... I hate gigabyt, Asus, etc... heatsinks cause they are copying Swiftech and Zalman and trying to dig into the heatsink manufacturer's business. Asus is trying to dig into Antecs case business. The list goes on and on.

Zalman and Antec make their own paste now too. I think Zalman should ship with a single use pack of Artic Silver 5. That would increase the value, ratings, online reviews, benchmarks, etc... of their products. the more people talk, the more you sell.

So despite the major competition in todays computer worlds I will always be a fan of Zalman for heatsinks, Antec and LianLi for cases, Abit for motherboards, etc... and of course Artic Silver 5 for paste.

The 15' drop was probably because his heatsink had 3 inches of dust caked in it. Its always a good idea to clean and reapply the paste every 6 months or so. I've tried those fan filters and they work wonders. I can go for over a year with barely any dust in the system.
January 11, 2007 12:40:24 PM

Quote:
Yea, just used some on mumz p4 today, aswell as cleaning the fan, h/s and heatspreader on the cpu, and using AS5, overall at idle is running 15 deg celsius cooler


So you are telling everyone here that the cpu paste alone cooled the cpu by 59 degree's fahrenheit :lol:  :lol:  :lol: 
January 11, 2007 12:49:23 PM

Quote:
Yea, just used some on mumz p4 today, aswell as cleaning the fan, h/s and heatspreader on the cpu, and using AS5, overall at idle is running 15 deg celsius cooler


So you are telling everyone here that the cpu paste alone cooled the cpu by 59 degree's fahrenheit :lol:  :lol:  :lol: 

read much? he didn't say the cpu paste alone, he said he cleaned the fans, heatsinks, spreader also. which can, all together, especially if there is significant dust buildup, most definately reduce your temp by 15c

Valis
January 11, 2007 12:57:25 PM

Yes, i read enough too know BS when i see it. Anyone who types that paste alone cools a cpu by even 5 degree's celsius(that's 41 fahrenheit) is very misleading.

P.S. Hell if it is that good I'm gonna coat my house with the stuff, and get rid of the air conditioner :wink:
January 11, 2007 1:06:55 PM

Quote:
Anyone who types that paste alone cools a cpu by even 5 degree's Celsius (that's 41 Fahrenheit) is very misleading.
That is certainly possible... I got 7C less (at load) on a P4 celeron I had after cleaning the heatsink and putting AS5 on it. It went from 65C (149F) down to about 58C (136.4F). Notice that is actually a 12F change.

EDIT: Here is a link for you: http://www.usatoday.com/weather/wtempcf.htm
January 11, 2007 1:10:53 PM

Quote:
Yea, just used some on mumz p4 today, aswell as cleaning the fan, h/s and heatspreader on the cpu, and using AS5, overall at idle is running 15 deg celsius cooler


So you are telling everyone here that the cpu paste alone cooled the cpu by 59 degree's fahrenheit :lol:  :lol:  :lol: 

No, he's saying that by replacing the paste and cleaning the heat spreader and fan he dropped the temperature 27 deg F

By your calculations a drop of 0 deg C would still be a drop of 32 deg. F. :) 
January 11, 2007 1:13:09 PM

Quote:
Yes, i read enough too know BS when i see it. Anyone who types that paste alone cools a cpu by even 5 degree's celsius(that's 41 fahrenheit) is very misleading.

P.S. Hell if it is that good I'm gonna coat my house with the stuff, and get rid of the air conditioner :wink:


Same logic. It's actually 9 deg. F drop.
(That's 41 deg f minus the 32 deg f offset)
January 11, 2007 1:13:26 PM

Quote:
Yes, i read enough too know BS when i see it. Anyone who types that paste alone cools a cpu by even 5 degree's celsius(that's 41 fahrenheit) is very misleading.

P.S. Hell if it is that good I'm gonna coat my house with the stuff, and get rid of the air conditioner :wink:



Ummm, a 5C change in temp is equivalent to a 10.5F change in temp. At least that's what my middle school taught me. Could be wrong.

Tom
January 11, 2007 1:42:26 PM

So if my cpu which idles at 86 degree's F, and i put some paste on there it would drop it to 45 F.(no paste in the world could do that). Maybe i should reword it for some of you guys. Anyone who type's that by cpu paste alone, can lower the cpu temps up too 41 degree's f is very misleading.

P.S Sorry to the OP, as i can see this is gettin out of hand. And is quickly becomeing something of an intel Vs. amd type of thread. I do not hate Artic silver, nor do i love it. But it can not lower cpu temp's by that much.
January 11, 2007 1:47:40 PM

Quote:
So if my cpu which idles at 86 degree's F, and i put some paste on there it would drop it to 45 F.(no paste in the world could do that). Maybe i should reword it for some of you guys. Anyone who type's that by cpu paste alone, can lower the cpu temps up too 41 degree's f is very misleading.

P.S Sorry to the OP, as i can see this is gettin out of hand. And is quickly becomeing something of an intel Vs. amd type of thread. I do not hate Artic silver, nor do i love it. But it can not lower cpu temp's by that much.


Please quote the post that said they lowered their CPU temp by 41 deg F using cpu paste alone.

A CHANGE (wish I could type this louder) of 41 deg F is equivalent to a CHANGE of 19.5C. Can't find that in any post.

YOU ARE CONFUSING ABSOLUTE TEMPS WITH RELATIVE TEMPS. PLEASE STOP. IT HURTS.

Tom
January 11, 2007 2:05:12 PM

Quote:
So if my cpu which idles at 86 degree's F, and i put some paste on there it would drop it to 45 F.(no paste in the world could do that). Maybe i should reword it for some of you guys. Anyone who type's that by cpu paste alone, can lower the cpu temps up too 41 degree's f is very misleading.

P.S Sorry to the OP, as i can see this is gettin out of hand. And is quickly becomeing something of an intel Vs. amd type of thread. I do not hate Artic silver, nor do i love it. But it can not lower cpu temp's by that much.


Please quote the post that said they lowered their CPU temp by 41 deg F using cpu paste alone.

A CHANGE (wish I could type this louder) of 41 deg F is equivalent to a CHANGE of 19.5C. Can't find that in any post.

YOU ARE CONFUSING ABSOLUTE TEMPS WITH RELATIVE TEMPS. PLEASE STOP. IT HURTS.

Tom

I stand corrected said the bowlegged man, I see your point, but i still do not belive that by adding artic silver alone, it will drop temps by 7 to 12 degress F.
January 11, 2007 2:12:59 PM

Quote:

I stand corrected said the bowlegged man, I see your point, but i still do not belive that by adding artic silver alone, it will drop temps by 7 to 12 degress F.


Me neither. Mine dropped, well I'm not sure it dropped at all when I switched from the stock goo from Intel to the AS5. Biggest difference I have made in CPU temps has been case airflow management. I got 8C just by re-arranging how my case fans sucked and blowed.

Tom
January 11, 2007 3:50:32 PM

Hehehe, this is really quite amusing. I was just showing my appreciation for a good product, didn't think it would really make a lot of difference.

Unit conversion will always always be a tremendous pain in the arse and create all sorts of problems until the yanks start using SI units like everyone else (I think I may have just opened a can of worms here....) and even then, I expect.
January 11, 2007 4:42:31 PM

Arctic Silver is some pretty good stuff alright. There are several others on the market that aren't bad either. Shin-Etsu, Zalman paste (although I've seen reviews where it beat and was handily beaten by the AS5 odd).

There might be a new King of paste/grease if this review is anything to go by. The new Thermalright paste sure looks good on the few sites that have reviewed it thus far.
January 11, 2007 4:45:06 PM

Quote:
So if my cpu which idles at 86 degree's F, and i put some paste on there it would drop it to 45 F.(no paste in the world could do that). Maybe i should reword it for some of you guys. Anyone who type's that by cpu paste alone, can lower the cpu temps up too 41 degree's f is very misleading.

P.S Sorry to the OP, as i can see this is gettin out of hand. And is quickly becomeing something of an intel Vs. amd type of thread. I do not hate Artic silver, nor do i love it. But it can not lower cpu temp's by that much.


Please quote the post that said they lowered their CPU temp by 41 deg F using cpu paste alone.

A CHANGE (wish I could type this louder) of 41 deg F is equivalent to a CHANGE of 19.5C. Can't find that in any post.

YOU ARE CONFUSING ABSOLUTE TEMPS WITH RELATIVE TEMPS. PLEASE STOP. IT HURTS.

Tom

I stand corrected said the bowlegged man, I see your point, but i still do not belive that by adding artic silver alone, it will drop temps by 7 to 12 degress F. Believe what you want... although I should say that my temps were at full load, the AS didn't change my idle temps much.
January 11, 2007 5:13:50 PM

Quote:
I just put some of this lovely grey gunk in between my e6400 and Scythe Ninja and it's already shaved off 3-4C from full load temps, this is before it's even had a chance to set properly. Wonderful.

So anyone else who loves this stuff, reply here!


Uhh, wouldn't the color be SILVER and not GREY? I mean, it is called Artic Silver 5, right? :) 
January 11, 2007 5:54:26 PM

That's what I thought, too... but it's called silver because it contains silver, not because of its colour.
January 11, 2007 6:28:27 PM

Excuse me, gentlemen, as I remove my foot from my keyboard! :) 

I hate to burst some people's bubble, but I did read in CPU magazine some time back about this whole thermal paste bit. The author tested the high-end products (AS5 amongst them), and generic goo from an automotive store and (I think), a Radio Shack.

As it turned out, all of the thermal pastes performed to within a degree (maybe two), of each other.

Personally, I am still using the stock HSF/thermal pads on my E6300 (OC'd to 3.15GHz), and the temps seem just fine.
January 11, 2007 7:36:32 PM

Quote:
I hate to burst some people's bubble, but I did read in CPU magazine some time back about this whole thermal paste bit. The author tested the high-end products (AS5 amongst them), and generic goo from an automotive store and (I think), a Radio Shack.
I have seen many reviews of AS online. Are you telling me they are all wrong? That my own results are also wrong? Do you have any links or sources to that information?
January 11, 2007 7:39:18 PM

Quote:
So you are telling everyone here that the cpu paste alone cooled the cpu by 59 degree's fahrenheit :lol:  :lol:  :lol: 


You might want to check your math on that one. 15 degrees C = 59 degrees F, but here you are talking about a 15 degree C *difference* (~27 F.)

After all, 30C is what, 86F? And if 15 C is 59 F then.... [/quote]86 - 59 = 27.....
January 11, 2007 7:59:13 PM

Quote:
Yes, i read enough too know BS when i see it. Anyone who types that paste alone cools a cpu by even 5 degree's celsius(that's 41 fahrenheit) is very misleading.

P.S. Hell if it is that good I'm gonna coat my house with the stuff, and get rid of the air conditioner :wink:



Again, I think you need to learn the difference between actual temperature and temperature difference. 5 degrees C would be 9 F.
January 11, 2007 9:50:57 PM

Nope, I read it in an actual magazine- remember those? I didn't clip and save the article because, well, it just wasn't that important to me.

However, I did try to do a quick search on CPU and MaximumPC (I think the article was in CPU, but I really don't remember); I didn't find anything.

Quite frankly, this whole "only use AS5" crap is akin to the shit I hear about AMD vs. Intel; ATI vs. nVidia; MS vs. Apple.

As long as you use the right amount of thermal paste and properly seat your CPUs HSF, then it's all good.
January 12, 2007 1:42:49 AM

yeah, I know what a real magazine is... I used to have a subscription to Maximum PC and PC Magazine but I canceled them because they were always about a month behind tech websites... whats the point in reading about something I already know about? And where did I say anything about AS5 only? It improved my systems load temps, thats all.
January 12, 2007 2:00:58 AM

Quote:
Excuse me, gentlemen, as I remove my foot from my keyboard! :) 

I hate to burst some people's bubble, but I did read in CPU magazine some time back about this whole thermal paste bit. The author tested the high-end products (AS5 amongst them), and generic goo from an automotive store and (I think), a Radio Shack.

As it turned out, all of the thermal pastes performed to within a degree (maybe two), of each other.

Personally, I am still using the stock HSF/thermal pads on my E6300 (OC'd to 3.15GHz), and the temps seem just fine.


agreed i have used them all, and yes the radioshack stuff.
which im using/testing now.

0, nada, zilch. difference in temps.
January 12, 2007 3:24:48 AM

Quote:
For under $10 a tube AS5 is worth it. It is also readily available. And a tube will last quite a while doing 10CPUs or more. What bugs me is that every company seems to be selling high end paste now which digs into artic silvers business. They are copying the originals.

I hate thermaltake cause they copy everyone else. ...blah blah blah... Arctic Silver 5 is cool... blah blah blah... I can go for over a year with barely any dust in the system.


Ok, for starters, Arctic Silver is a material. Technically a blend of materials, but just a dollop of goop, really. Highly engineered goop, mind you, but goop nonetheless. And assigning "intellectual property rights" to materials annoys me in the first place, but whatever. But herein lies a lesson about generic products.

There are protections in our legal system for "copying" products. Such as patent law. And under patent law, if your idea is original enough, you can patent it to prevent other people from copying you. So if somebody is selling a similar product as somebody else, either somebody didn't take proper steps to protect their intellectual property rights, or somebody else is just taking advantage of the fact that somebody merely used a general engineering solution. While I recognize the desire to reap the benefits of putting in the work, generic products keep consumers from getting gouged. So if somebody comes up with an idea that's worth patenting, either for a cooling solution (like paste or heat spreaders) or for prescription drugs, they should do so and market their product while it is under patent. If the idea isn't all that original, then why should they be able to reap the spoils when somebody else could potentially design something else just as good?

With that said, Arctic Silver is, indeed, a patented product. Mostly due to the fact that AS has optimized the heat conductivity against the paste's insulating properties. (if you ever feel the urge, stick some probes in the paste and test for conductivity - it doesn't have any, which is pretty amazing considering it's a paste with a ton of silver particles in suspension) This is mostly so retards who get paste everywhere won't short out anything, since who cares if the heat sink and processor share a common static voltage?

Regardless, while all of what has been mentioned in this post is all well and good, Arctic Silver isn't exactly the end-all be-all of thermal conductivity. As a matter of fact, Crest has them beat pretty good. For that matter, so does Vegemite. See here:
http://www.dansdata.com/goop.htm

My hunch is all that ceramic in suspension along with the oil isn't exactly helping the silver move those joules around. But that's just a hunch.

Also, Everett, please please PLEASE learn how to convert to the metric system. It is way better, and I say that as an American.

Now, as far as your comment about
Quote:
So if my cpu which idles at 86 degree's F, and i put some paste on there it would drop it to 45 F.(no paste in the world could do that).

You're right. Actually, no paste in the world can even force a processor to drop even 1°F. That's all about ambient temperature, albeit in conjunction with the paste, heat spreader, fans, etc. If you really wanted, you could get under 45°F with no paste whatsoever. Or even a heat sink. You'd just have to keep the room cold enough. Like, say, if you submerged your comp in liquid helium, I bet you'd have some real difficulty getting the temp even above 100K. Granted, that would probably be because you'd have broken your processor by freezing it to death, but you know, details.... Regardless, the only way your processor is getting down to 45°F when your room is near 70°F ambient is if you attach a refrigeration unit. No matter what paste you use. (unless it's an endothermic reactive paste - let me know if you come up with that one, and I'll help you market it)

By the by, I strongly suggest a few of you check this out:
http://www.youtube.com/watch?v=B9q2jNjOPdk

Might be worth a quick watch.
a c 197 K Overclocking
January 12, 2007 4:24:29 AM

Anyone have any idea of the storage life of an open tube?
January 12, 2007 7:50:13 AM

Interesting review, that, especially as the opinion on most review sites seems to be that silver grease > white grease.

And on the colour thing, it's a bit like bubbles: they're made of see through stuff, but a bath full of them is white....
January 12, 2007 10:15:10 AM

Quote:
By the by, I strongly suggest a few of you check this out:
http://www.youtube.com/watch?v=B9q2jNjOPdk

Might be worth a quick watch.


That's great. They should make one for Tom's Forumz, and make it a condition that you have to watch it when you sign up. Then, if anyone breaks the rules, they get BANNED FOREVER. OMFG!
January 12, 2007 11:36:45 AM

I am an electronics engineer so I'll give little theory on thermal snot for those who don't know.

The ideal setup is for a heatsink to sit directly on top of the CPU. Copper to die, the heat swims up the little fins (large surface area) and the fan blows air onto the fins to take heat away. I always learned in heat sink design class that they should be flat black? I don't know if this is true or not, would be nice to test out. Maybe copper painted flat black would perform better. In my class some people painted their heatsinks gloss black and the instructor made them repaint flat black.

Well in a perfect world the heatsink does not sit flat on the surface. Everything is not laser lapped to be perfectly flat to a millionth of a millimeter. So we apply some thermal goop to fill the gaps. A little thin layer just enough to squeeze and fill the voids so everything sits flat. The compound actually conducts heat less than having the heat sink sit directly on the die, so having too much compound is no good.

Is there a difference between cheap compound and good compound. YES. The good ones lst longer without drying up, and they have better thermal properties. Will switching to Artic Silver make a 5'C difference. I don't think so. Maybe 1or2'C at best. Will switching from an inch thick of oozing el cheapo dried out compound to a nice fresh thin layer of AS make a 5'C diference. Of course.

I use AS5 cause I'm guranteed its good, its readily available, its not expensive, and they make a complete lineup of epoxys too. In my store I like having the AS5 paste and Epoxy right up on the shelf for sale. nice to have both from the same manufacturer.

But in all honesty in the end, the quality of lapping on your heatsink will make all the difference in the world. The flatter it sits, the less compound you depend on and the better your cooling. Get a good heatsink firstly, it usually comes with a tube of grease and your good to use that.

As far as OEM heatsinks go, I always just peel the pink pad off and clean it down with a razor and alcohol or goo gone. I will then use some white thermal compound I have hanging around from another heatsink or some AS5. I always have AS5 hanging around in the shop and exclusively use that, I usually give my tubes of white stuff away to people.

If I get a tube of white stuff with a heatsink, I keep it but use AS5 instead. If somebody is going to use and OEM heatsink I give them one of those tubes of white stuff on the house and tell them to use that instead of the pink pad.

Artic Silver 5 is more of a marketing gimmick, they are a name brand that I can advertise and it looks good. Are they the best...probably damn near it. Are they better than the rest, well technically yes if a 1/2'C makes any difference. I jsut use it cause its what I order from the wholesalers because its available and I use their epoxy too for ram sinks.

I also read once that Shin Etsu was superior to Artic Silver. But its not as readily available as Artic Silver.
January 12, 2007 1:09:14 PM

For years now, whenever I build a new machine, the first thing I do with the heatsink is lap it.

I stick the TV on and sit down with a beer and an old, small, engineer's flat and several grades of wet'n'dry going down to crocus paper. After two or three episodes worth of time carefully draw-filing the heatsink on the abrasive papers the bottom of the heatsink is dead flat and polished.

You would be amazed at the lumps and dips that show in the boxed coolers from both of the biggie chip makers when I start the first, coarsest, paper. I usually start with a 120 carborundum then a 160 then 300 and finally crocus for the polish. Naturally, I sand at right angles for each new, finer, paper until all the previous lines are all gone.

By doing this lap and polish I have seen at least 4*C and sometimes as high as 7*C difference compared with no lapping and normal paste. I usually use a fine layer of boring white paste applied with a razor blade as that is all that is needed after this kind of preparation.

I was given a tube of AS5 once and so I did a compare. I found a temp difference of less than 1*C so I just use the white goo since the AS5 ran out. I waste most of it when I scape it around with the blade. (you can't get it back in the tube - otherwise I'd never run out)

Q
January 12, 2007 2:43:35 PM

What does the metric system have to do with fahrenheit or celsius? :? Anyway, now that you brought that up. I am well aware of how to use the metric system. Seeing how i have more than one set of metric and standard tool's at the shop. These are not the cheap sets mind you, one set of metric wrench's cost as much as an 8800gtx(and thats just for an 8mm to 19mm set). :wink:
January 12, 2007 3:03:35 PM

Everett,

I believe NovemberWind was referring to the fact that the Celcius scale for temperature is part of the 'Systeme Internationale' (SI) units otherwise referred to as the Metric system or the 'MKS system' (Metre, Kilogramme, Second)

As for tool sizes, most of the automotive industry now uses metric sizes, however, the aviation industry still uses the Imperial system for fastener sizing, more specifically AF sizes. It is unlikely they will ever change unless Boeing, AirBus and McDonnellDouglas cease building aircraft and all aircraft get made in Russia and China.

Q
January 12, 2007 3:12:03 PM

LOL... endothermic reactive paste! Actually phase change cooling is kind of that idea, except it's not a paste.

btw for all the geniuses.... 5C drop = 9F drop. My car is canadian, which sucks, because I have to adjust temperature in celsius. Any passengers are like, wth?, why are you setting the temp to 21?

I put arctic silver on my cpu, and it did take a couple of degrees off. I think the advantage of AS5 is you can put down a really thin layer.
January 12, 2007 4:43:16 PM

OMG dude, that was great! :) 

I loved that YouTube link; as I was reading some of the posts here, I was going to post a link to a site I saw last night where a guy did a test with toothpaste and vegamite (amongst other things) as a thermal paste, and as I was doing the search, I saw the link to it in your post! Ya beat me to the punch!
January 12, 2007 5:06:39 PM

Everything Gondo said is pretty much correct. I'm just going to add a little bit to his "black heatsink" bit.

Quote:
I am an electronics engineer so I'll give little theory on thermal snot for those who don't know.

The ideal setup is for a heatsink to sit directly on top of the CPU. Copper to die, the heat swims up the little fins (large surface area) and the fan blows air onto the fins to take heat away. I always learned in heat sink design class that they should be flat black? I don't know if this is true or not, would be nice to test out. Maybe copper painted flat black would perform better. In my class some people painted their heatsinks gloss black and the instructor made them repaint flat black.


Flat black has to do with the thermal emissivity of a material. In all honesty, it isn't going to make a bit of difference for computing applications, but for those curious few, here we go:

Heat is transfered by conduction (heat moving from the bottom of your heatsink to the fins), convection (air moving heat away from your heatsink), and radiation. It's radiation where black comes into play. But thermal radiation isn't a really large factor at low temperatures - mostly dominated by conduction/convection. Still, if you want your heat sink to give off as much heat as possible, it needs to be "black."

The reason I put black in quotes is that in this sense, "black" means something slightly different (though related) to what you see visibly. A perfect black body has maximum emissivity, meaning that the body is capable of absorbing and emitting all wavelengths of light and by "light" I mean everything on the E&M spectrum. And the light that is emitted at different temperatures is different colors. For example, when your toaster oven glows red-hot, because the temperature is in the red color range. Your processor never (I hope) glows red-hot. It's color range is going to be pretty far into the infrared range. That means that the color of your heat sink needs to be "black" in the infrared range - not just regular black, which might not really be "black" once you get outside the visible spectrum.

Cheers.
January 12, 2007 5:27:10 PM

So how did AS change my temps? The heatsink was properly seated to begin with. It only took me a few minutes to change and clean the heatsink... there was no time for the ambient temp of my room to change. I ran SETI to bring the CPU temps up for about 10 minutes before and after I changed it.

EDIT: My reply was supposed to be to NovemberWind
January 12, 2007 5:36:36 PM

C'mon gondo, don't be a hater.

Competition is great, because we can see innovation in products and lower prices. :D 
January 12, 2007 5:42:49 PM

Quote:
Regardless, while all of what has been mentioned in this post is all well and good, Arctic Silver isn't exactly the end-all be-all of thermal conductivity. As a matter of fact, Crest has them beat pretty good. For that matter, so does Vegemite. See here:
http://www.dansdata.com/goop.htm
I already read that... I really like the "Mislead-O-Tron", lol. lots of tech sites do that, thats why its good to read the numbers on the graphs.
January 12, 2007 5:45:49 PM

Yeah, but nobody on the fourms uses kelvin to keep track of there temp's, thats why i was confused as to why he brought up the metric system. :wink:
January 12, 2007 5:46:53 PM

what else can you say about as5, the only better solution would be to fuse the processor to the heatsink, it does not get any better than as5. :D 
January 12, 2007 5:57:22 PM

Quote:
Everything Gondo said is pretty much correct. I'm just going to add a little bit to his "black heatsink" bit.

I am an electronics engineer so I'll give little theory on thermal snot for those who don't know.

The ideal setup is for a heatsink to sit directly on top of the CPU. Copper to die, the heat swims up the little fins (large surface area) and the fan blows air onto the fins to take heat away. I always learned in heat sink design class that they should be flat black? I don't know if this is true or not, would be nice to test out. Maybe copper painted flat black would perform better. In my class some people painted their heatsinks gloss black and the instructor made them repaint flat black.



Flat black has to do with the thermal emissivity of a material. In all honesty, it isn't going to make a bit of difference for computing applications, but for those curious few, here we go:

Heat is transfered by conduction (heat moving from the bottom of your heatsink to the fins), convection (air moving heat away from your heatsink), and radiation. It's radiation where black comes into play. But thermal radiation isn't a really large factor at low temperatures - mostly dominated by conduction/convection. Still, if you want your heat sink to give off as much heat as possible, it needs to be "black."

The reason I put black in quotes is that in this sense, "black" means something slightly different (though related) to what you see visibly. A perfect black body has maximum emissivity, meaning that the body is capable of absorbing and emitting all wavelengths of light and by "light" I mean everything on the E&M spectrum. And the light that is emitted at different temperatures is different colors. For example, when your toaster oven glows red-hot, because the temperature is in the red color range. Your processor never (I hope) glows red-hot. It's color range is going to be pretty far into the infrared range. That means that the color of your heat sink needs to be "black" in the infrared range - not just regular black, which might not really be "black" once you get outside the visible spectrum.

Cheers.


How would the color affect radiation? I must be missing something.

One would think that painting the heatsink would serve as an insulator, which is why you don't see heat sinks painted.
January 12, 2007 6:07:50 PM

mm i love it cause it dropped from 50 to 47 45 which isn't that bad in my case with a stock fan for my socket 940. is it okay also to use it when watercooling and how often would i have to change it in order to keep the temperature low?
!