Arctic Silver weirdness

[occamsblade]

So, I want to start by saying I have used arctic silver 5 in all of my builds for years. I only recently started playing with their ceramique, and would love feedback on which is better. I have not had issues with it other than this one time. I have a gaming laptop that I rarely use. I was working on the loop on my desktop so I fired it up and was playing warframe. I got an error screen from warframe indicating that it sensed that the computer was overheating / having issues. When I ran the montior program to look at temps my CPU was over 90 deg C! I had redone the paste and done some other work on it (I think swapped out wifi card) about 8 months ago, and it had pretty much been sitting since. It had quasi ok temps at rest, but they spiked quick when I tried testing it by gaming again. So, I ripped it back down and this is what I found when I took the heatsink off. I don't recall if this is a tube of arctic silver I bought from microcenter or off of amazon. Thoughts? I put a thicker layer and tinted the heatsink this time, the CPU will get to about 85 and level off gaming. Still high, but its a laptop with a baby cpu chip and I read that those get high. The bigger of the two is the GPU, and it stays cooler than the cpu does. Does arctic silver go bad? Too thin? Possibly fake product? Any thoughts?

 

[Witchdoctor]

Not my favorite TIM, but have used it in the past with no problems ?

Maybe just bad batch ? just about anything should do the trick, personally I use this

https://www.amazon.com/GELID-GC-Ext...05&hvtargid=aud-801381245258:pla-436965406697

 

[Johan45]

I have had a tube of AS5 go bad before so it can happen. On a laptop like that with exposed dies I wouldn't recommend AS5 since it is conductive. I have also used the ceramique and it works quite well.

 

[trents]

I haven't used AS5 in years. It's not a bad product but there are many that perform a little better and don't require a curing time. I use Arctic MX4.

 

[freeagent]

Indeed, I used to swear by AS5. My preference goes to Thermalright TF8 now. Love that stuff.

 

[SPL Tech]

AS is outdated. It may have been good at one point but its time to shine is long gone. There are more than 20 other TIMs which are superior in performance. AS was like Windows XP. It was great at the time it existed, but half the people out there dont even know what Windows XP is anymore. AS is the Windows XP of TIM.

 

[EarthDog]

It may be old... and 20th... but it is still within a couple C of the non metal pastes.. most wouldnt know.

But ueah, some bad samples happen... maybe that is it.

Edit: thread is from February, lol..

 

[freeagent]

I wouldn’t call it outdated, it’s still good stuff, it’s a little finicky on how much is enough, too much, too little, there is a technique to applying it. Even the cure time isn’t a big deal. I just heat cycle it to help it along. I swore by it for nearly a decade, while trying other popular brands.

 

[SPL Tech]

I wouldn’t call it outdated, it’s still good stuff, it’s a little finicky on how much is enough, too much, too little, there is a technique to applying it. Even the cure time isn’t a big deal. I just heat cycle it to help it along. I swore by it for nearly a decade, while trying other popular brands.

But there are so many brands that are better and they are all more or less the same price. The very best stuff on the market is only $8, so what's the point in having the 20th best product when you can have the absolute best for the same price? There is no reason to use any TIM other than the best non-conductive TIM out there as anything less is just worse performance for the same amount of money.

 

[EarthDog]

The very best stuff on the market is only $8...when you can have the absolute best for the same price?

Kryonaut is $8 /g. You can get 3.5x the paste and 1C difference if you choose nht-2 for $1 more...total mind you, not per gram. The difference of 1C can be lost in a poor application or mount. But.. it depends in how often you use it too. I can go through a gram in a month reviewing motherboards... a gram will last a few/several applications.

Conductonaut, what a couple reviews consider as the best, is $14 /g...
https://www.guru3d.com/articles_pages/guru3d_thermal_paste_roundup_2019,9.html

 

[freeagent]

I hear you. It’s still a very good paste if you know how to use it, I won’t knock it. I would put it up against other popular brands. A lot of these new reviewers are just that.. new.. but anyways.. I like Thermalright TF8, slightly better than Kryonaut on paper, and easier to apply than as5

 

[Johan45]

Back to the OP, did you check the exhaust area of the laptop? I find that is usually the culprit for overheating as it gets clogged and can't vent properly

 

[SPL Tech]

Kryonaut is $8 /g. You can get 3.5x the paste and 1C difference if you choose nht-2 for $1 more...total mind you, not per gram. The difference of 1C can be lost in a poor application or mount. But.. it depends in how often you use it too. I can go through a gram in a month reviewing motherboards... a gram will last a few/several applications.

Conductonaut, what a couple reviews consider as the best, is $14 /g...
https://www.guru3d.com/articles_pages/guru3d_thermal_paste_roundup_2019,9.html

It's a lot more than 1C. More like 5C in most of the reviews I have seen and the difference between the best stuff and something toward the bottom is more like 10C. When I swapped my TIM for 'the good stuff' I dropped 5C instantly.

 

[EarthDog]

It's a lot more than 1C. More like 5C in most of the reviews I have seen and the difference between the best stuff and something toward the bottom is more like 10C. When I swapped my TIM for 'the good stuff' I dropped 5C instantly.

I provided a link showing the 1C difference between those pastes. I can drop another link showing the same... wheres your 5C link?

Toothpaste vs liquid metal doesnt count for 10C...

Again, a few/several C is about the difference between non metal pastes. There are multiple roundups showing that. Here, its better.. https://www.vortez.net/articles_pages/noctua_nt_h2_review,5.html

 

[SPL Tech]

I provided a link showing the 1C difference between those pastes. I can drop another link showing the same... wheres your 5C link?

Toothpaste vs liquid metal doesnt count for 10C...

Depends on how you mount it. Mount it like most people do and it can be as high as 23C actually. If you completely nail it and have a crap ton of mount pressure, the gap closes but it's still large enough to warrant buying one brand over another when they cost the same anyway.

So there is my answer. In most mounting scenarios, 23C.


https://www.tomshardware.com/reviews/thermal-paste-comparison,5108-11.html

 

[freeagent]

Overclockers.com should do a modern TIM roundup with all the current big names, not so much the little ones unless they are special.

 

[EarthDog]

Toothpaste vs liquid metal doesnt count for 10C...

Depends on how you mount it. Mount it like most people do and it can be as high as 23C actually. If you completely nail it and have a crap ton of mount pressure, the gap closes but it's still large enough to warrant buying one brand over another when they cost the same anyway.

So there is my answer. In most mounting scenarios, 23C.


https://www.tomshardware.com/reviews/thermal-paste-comparison,5108-11.html

Like most mount? Did you read what the test methodology was?

For example, we're using popular pre-assembled water coolers that should ensure heat sink temperatures below 60 °C (140 °F), premium aftermarket air coolers with back plates that should demonstrate high mounting pressure, and a run-of-the-mill budget cooler with push-pin mounting (that'll give us limited pressure). Stock coolers like that let the CPU get above 60 °C/140 °F (AMD) and 80 °C/176 °F (Intel).

Depending on viscosity and composition, not all pastes are a good fit for every application, nor are they all well-suited for novices.

So... low pressure (pushpin mount), and high pressure (aios and heatsinks with backplates) shows exactly what I said. Whatever his medium pressure is show a big difference. He doesnt mention what medium is really. Why do you think, with what he already defined, medium is what 'most use'? To me, seems like most use an aio or aftermarket heatsinks with a backplate or pushpin cooler...

In normal enthusiast situations, aftermarket coolers with higher mounting pressure dont show it's worth paying 3.5x for the same amount of product. Nor pushpin air...just this mythical medium mount...whatever that is That doesnt mean one cant.. but just saying it typically isnt worth it...that link helps..thanks!

 

[Silver Surfer]

Artic Silver 5 was designed to form a layer between the contacting surfaces, that's why the long cure time so the layer could be seated, which was great in the old days when heat sinks were not finished to a mirror finish but today many are finished to excellent contact surfaces so you no longer need a layer building type of TIM. All you need is to fill the microscopic imperfections between the 2 surfaces, and that is it.

The OPs pictures almost look like the surfaces were not completely clean when put together in the first place, looks like oil or sweat was left behind before the surfaces were put together, however the picture is also a great visual example of the surfaces not needing a layer building TIM. GC Extreme or Artic MX4 would be great on that setup, at less than half of what was used. SS

 

[freeagent]

So every time I posted in this thread it was from my phone, and for some reason I couldn't see that pic. That AS5 is separated. It is not supposed to look like that.

 

[trents]

Personally, I think the person who applied that AS5 in post #1 had a runny nose.