Anyone using AS3??

[nealh]

I saw post at amdmb.com...from a user with KG7 mobo and saw his temp drop 3C at idle and under load

Anyone here using it???

 

[MrThompson]

I have been using it for several months now (engineering sample.) Great stuff!

 

[dannybin1742]

where do you find as3?

provide a link

 

[gaidin123]

http://2cooltek.safeshopper.com/50/332.htm?573

Go 2Cooltek! I just got some from them an am waiting for a few more parts to arrive before using mine...

Gaidin

 

[Divebrake]

I just put some AS3 on my Pal 8045 Tuesday. I can't see a difference in temps. I was using Antec Reference, which is also a micronized silver compound. Arctic Alumina wasn't any worse before then. Maybe AS3 is all hype or maybe Antec Reference and Arctic Alumina are equally good products, I don't know. Read this:

http://www.hardocp.com/reviews/cooling/compound/

The hardocp guys think expensive heatsink compounds are all marketing hype. I think they could be right.

 

[nealh]

great replies thanks

 

[Mikewarrior2]

DiveBrake,

And HARDOCP heatsink/grease testing is full of crap, and pure FUD.

see this link: Void Your Warrant Internal Diode Tests

see how the socket-thermsitor is incapable of picking up temp changes reliably. 7-5C Internal Diode tests show up as 0C socket-thermsitor change. Drawing conclusions based on a crap testing methodology is pure FUD. ANd they know that their testing methodology is FUD. Even Anandtech andTomsHardware have switched away from socket-thermsitor testing to be more accurate. Why won't HardOCP do the same? Because Kyle has a little leprechaun talking in his ear telling him to yell at those of us who have e-mailed him countless times asking him to change his testing methods. Now there's clear proof that he was wrong. yet he still draws the same conclusions despite knowing how he tests is wrong.



Mike

 

[rickn]

I see that 7c change that didnt record, what they don't detail is how long that was for. If it was just for a brief instant, the bios may not just have had time to update. I am not sure what the updating interval is on most motherboard, probably 5-10 seconds I guess. Now, if they shut it off for a minute and the thermistor didnt register any temp change, that aint very good. Overall they are pretty close. The thermistor is accurate enough for most people

 

[Mikewarrior2]

the 5 and 7C diode changes with 0 Socket-thermistor change was over several hours.... He had originally been testing without reapplying grease, and noted the original 7C over optimal application temps. Then, after optimally applying arctic alumina, he noticed a 7C "drop" in internal diode temps while no drop in socket-thermistor temps.

This has nothing to do with whether the socket-thermistor had time to adjust... it has to do wtih the socket-thermistor being craptacular at measuring temp changes.

Yes, the socket-thermistor is good enough for an end user as a FIRE alarm, if and when calibrated for your particular motherbard.

However, they are worthless for testing purposes.


Mike

P.S. This also means that since socket-thermistor's can't tell a difference between a good grease application and a poor grease application, how is it possible to tell the difference between different greases while using this type of temp measurement???

 

[rickn]

In this Graph you can see the spike in temp occured at the 25min mark, but that was for a very brief moment according to the Time [minutes] listed down at the bottom row. The thermistor picked it up, but it was of course a little later and not as good. I dunno, maybe i'm on the wrong graph. Either way, they are a cheap solution that motherboard makers didnt have anything else to fall back on since AMD cpu's didnt have a internal diode up until recently. Intel has had em for a real long time

 

[Mikewarrior2]

YOU need to compare both graphs:

Glaciator Bad Paste and
Glaciator Good Paste

The Important notes between the two: Socket-thermistor temp does not change between a good grease and bad grease application. Yet, the internal diode shows massive temp change. Again, clear evidence of socket-thermistor temp compression.



Mike

 

[WarCon]

One thing I heard, and it was a very reliable source, is that the Antec Reference is rebadged Arctic Silver II anyway. So if socket measurements can't tell any difference between it and AS 3 or Arctic Alumina then its no wonder.

One thing though. When you checked to see if it was any different, did you lap your heatsink between the tests? If not then you were mostly testing the first compound anyway. It stays in the microscopic valleys even though washed in solvent.

I am running thermal compound tests right now using what I have. A socket thermistor and a sidecore thermistor. I don't expect scientific results, just a personal comparison. I am also testing for highest stability points, because that is really why I would pick a good thermal compound anyway. Also one that I know is tested for long term stability is very important to me.

 

[Richardito]

<< I have been using it for several months now (engineering sample.) Great stuff! >>



How did you get that sample? Connections?

 

[Richardito]

<< One thing I heard, and it was a very reliable source, is that the Antec Reference is rebadged Arctic Silver II anyway. So if socket measurements can't tell any difference between it and AS 3 or Arctic Alumina then its no wonder. >>


Who is your source? Do you also have connections?

 

[Richardito]

<< DiveBrake,

And HARDOCP heatsink/grease testing is full of crap, and pure FUD.

see this link: Void Your Warrant Internal Diode Tests

see how the socket-thermsitor is incapable of picking up temp changes reliably. 7-5C Internal Diode tests show up as 0C socket-thermsitor change. Drawing conclusions based on a crap testing methodology is pure FUD. ANd they know that their testing methodology is FUD. Even Anandtech andTomsHardware have switched away from socket-thermsitor testing to be more accurate. Why won't HardOCP do the same? Because Kyle has a little leprechaun talking in his ear telling him to yell at those of us who have e-mailed him countless times asking him to change his testing methods. Now there's clear proof that he was wrong. yet he still draws the same conclusions despite knowing how he tests is wrong.



Mike >>



Hey, there is free speech in this country (did ya know?). You can't tell people to say whatever you want them to say. Intelligent people draw their own conclusions...

 

[Mikewarrior2]

yes. I do not see where I say something that is against free speech. In fact, it is my right to free speech that allows me to criticize what I believe is a clearly FUD based review site when ti comes to heatsink reviews. For over a year, I have been saying that socket-thermistor's are not accurate. NOw there are 2 sites with clear proof of this. period. Lots of intelligent people believe that socket-thermistor's are inaccurate as well.

In fact, if you are claiming otherwise, go read up on the internal diode testing. I know you're a huge of fan of HardOCPs work, adn that is fine. My post is for those who think that jsut because its published on the internet, that it must be accurate.

Also, yes it is Kyle(of HardOCP's) right to claim that greases are marketing hype. Unfortunately, he bases his conclusions on a measurement source that is highly inaccurate and not usable for comparison purposes. Again, as I have stated for over a year.


mike

 

[Nevin]

Richardito,

Quite a few people have early engineering samples of AS3 because when they saw mention of it in a forum, they contacted me and asked for some. If you had asked, I would have sent you some too.

Nevin House
Arctic Silver, Inc.

 

[sygyzy]

Nevin,

I'd like an engineering sample of Artic Silver 4. Yes it's not out yet but I want to be there when it is.

 

[Nevin]

I'll put you on the list, but don't hold your breath. It may be a while.

Nevin

 

[pillage2001]

<< I'll put you on the list, but don't hold your breath. It may be a while.

Nevin >>



I'll take some off you too. I've used all the Artic silvers you can produce.

 

[MrThompson]

I see Richardito is in the attack Arctic Silver mode again...

 

[Mikewarrior2]

<< I see Richardito is in the attack Arctic Silver mode again... >>



nah, he's just upset that socket-thermistors have now been shown to be even more inaccurate than believed initally...... and because his "site's" reviews are now shown to be inaccurate, unreliable, and erroneous.



MIke

 

[Divebrake]

http://www.vanshardware.com/articles/2001/december/011209_AS_Interview/011209_AS_Interview.htm

Here's an interview with the owner of Arctic Silver. He echoes mikewarrior2's statements about socket thermistors making Arctic Silver products look bad. Of course he could be biased, he's got grease to sell. I'm not really sure who to believe.

 

[Divebrake]

<< One thing though. When you checked to see if it was any different, did you lap your heatsink between the tests? If not then you were mostly testing the first compound anyway. It stays in the microscopic valleys even though washed in solvent. >>



Yeah, I thought of that and I did lap when changing greases for that very reason.

 

[Mikewarrior2]

if Kyle @ hardOCP would base his claims/observations on more accurate means of temperature measurement, then I wouldn't have a problem with it. HOwever, it is when people make observations using a temp source that is unreliable, inaccurate, and often times compressed beyond all belief that I utterly cannot standy and watch them improperly trash a product.

Again, look at the voidyourwarranty.net tests using the internal diode. A socket-thermistor cannot pickup the difference in temp betwee a good grease application adn a poor grease application (7C diode, 0C socket-thermistor). That's clearly a good sign of the compression you get from socket-thermsitors.


Mike