[EOL] Arctic MX-5 is here!!Tests incoming! Completed. Now its MX-6 testing time!

[LFaWolf]

Isn’t arctic silver 5 conductive and you can’t use it on gpu and could also possibly short out your motherboard if you get the paste on the board?

 

[P4-630]

Isn’t arctic silver 5 conductive and you can’t use it on gpu and could also possibly short out your motherboard if you get the paste on the board?

Google says:
Not Electrically Conductive: Arctic Silver 5 was formulated to conduct heat, not electricity. (While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads.

I'd say yes.

 

[Vario]

Isn’t arctic silver 5 conductive and you can’t use it on gpu and could also possibly short out your motherboard if you get the paste on the board?

People make that argument, but other people use liquid metals. Also, it's not conductive.

Google says:
Not Electrically Conductive: Arctic Silver 5 was formulated to conduct heat, not electricity. (While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads.

I'd say yes.

Why would you say yes when the source you quote says no?

If your still rockin a quadcore from 2008 and using that $60 DD tdx block (I've got an AMD version in my junk box) those results look saweet!
Today? Mmmmnah, nice try.

is this a better try?

Best Thermal Paste for CPUs 2024: Over 50 Compounds Tested and Ranked

The right thermal paste will improve performance.

www.tomshardware.com

 

[LFaWolf]

I looked it up since I am back on my PC -
"Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.
(While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)"

Source- http://www.arcticsilver.com/as5.htm

So I would still avoid it, since MX-4 is equivalent, if not better, and not thermally capacitive.

 

[Vario]

I looked it up since I am back on my PC -
"Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.
(While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)"

Source- http://www.arcticsilver.com/as5.htm

So I would still avoid it, since MX-4 is equivalent, if not better, and not thermally conductive.

Stuff has been around forever, has it EVER happened to anyone?

 

[Mussels]

Pump out effect has yet to be proven in all use case scenarios. However, they do dry out and lose some of their effective cooling ability.


I wouldn't go that far as my own testing showed it was a good performer. It's not the worse, but it's not the best either and as mentioned above, it does dry out.



Greatly doubt. Highly subjective situation. I've been working in this industry for decades and have worked on tens of thousands of machines and have never seen definitive evidence of it. Not once.

You missed the point:

I never said pump out is an issue, a concern, or a problem for PC builders.
What i said is: Pump out is a real, physical phenomenon that's already been dealt with at a design level.

Go back to CPU's without an IHS and try the runnier pastes, theres a reason OEM's used the chewing gum crap - so it wouldnt simply melt off the side from our poorly mounted coolers of that era
And yes they were 100% poorly mounted, when you could chip a die by putting it on slightly off center and needed the force of an angry rhinoceros and the precision of a dentist to add and remove

as coolers and mounting have changed over the years, so have thermal compounds. Havent we seen that in this thread, where certain compounds just absolutely suck for laptops and naked dies despite being great in other situations?

 

[LFaWolf]

Stuff has been around forever, has it EVER happened to anyone?

Maybe I don't know. But if it has metal (silver?) traces in it I am going to avoid it and stick with MX-4. Not going to take the chance.

 

[freeagent]

Stuff has been around forever, has it EVER happened to anyone?

Never had a problem. Even when I was sloppy when I was new. S462 had a bare die and pins lol.. no problems ever.

 

[maxfly]

People make that argument, but other people use liquid metals. Also, it's not conductive.

Why would you say yes when the source you quote says no?


is this a better try?

Best Thermal Paste for CPUs 2024: Over 50 Compounds Tested and Ranked

The right thermal paste will improve performance.

www.tomshardware.com

Ab so loot lee, lol. Budget pastes have always had a place in every toolbox. I simply found it odd to use such an old block and test as an example of AS5s efficacy. You obviously love your tim and that's cool man, you do you! I like many others, didn't and don't like their wishy washy conductive/capacitive statement with my bare cpus, gpus and chipsets. Fortunately there was/is an easy solution back then and much more so now. Switch, to Ceramique a guaranteed non conductive tim then(that performed a bit better and didn't freeze). Or to something like MX-4 now. Thermal paste is an easy cheap temp improvement and nearly all of it is non conductive nowadays, or its not. Its generally not hard to find out as its a big marketing point.
As far as, has anyone ever damaged anything with AS5? Does it matter? Why worry about it when there's bucket loads of better, guaranteed non conductive options to choose from? If there's even the slightest chance that AS5 might damage their gear most people will use something else that doesn't purposefully try to confuse them with the conductivity, or capacitance silliness. Its a complete non issue now. LFaWolfs choice to use MX-4 speaks directly to that point.

 

[lexluthermiester]

What i said is: Pump out is a real, physical phenomenon that's already been dealt with at a design level.

Fair enough.

Go back to CPU's without an IHS and try the runnier pastes, theres a reason OEM's used the chewing gum crap

I always wondered about that. Was serious in saying I've never seen pumpout, even back in the socket A & socket 370 bare die days.

Isn’t arctic silver 5 conductive

No it isn't. BITD it had a capacitive effect but it was uncommon and only happened if excess AS5 got on electrically conductive parts and bridged them. Using only enough AS5 to do the job was the best solution. The current formulation is absent that capacitance effect and AS5 is still an excellent performing TIM.

 

[Pictus]

I am using Alseye T9+ Platinum, at the time it was the only potentially good
paste quick available "at hand" here in Brazil, I prefer it over the Halnziye Hy-p13.


Composition: Silicone 10%, Carbon 45% and Metal Oxide 45%
Temperature Resistance: -30°C ~ 280°C
Thermal Conductivity: 13.5 w/m-k
Viscosity at Low Strain Rate 1,200 Pa-s
Viscosity at High Strain Rate 100 Pa-s
Specific Gravity 2.6
Volatile Content, 48 hours at 125°C 0.02%
Thermal Resistance at 25 N/cm² 0.04 °C-cm²/W
Bond Line Thickness at 25 N/cm² 0.02mm (0.0008 in)

Here is a Ryzen 5600X + Thermalright FC140, the foot/plate of the FC140 is a bit
convex, so I lapped just bit to make it more flat to better fit with the 5600X.
After almost 6 months, I decided to check visually(the TMP was OK) if the lapping
was correct and the state of the paste.
Well, the paste was +- ok, but at the highest pressure points it was drier.
So I decided to make an experience...
Removed the paste in the CPU/Cooler and mixed it with a bit of Krytox grease, it
is the grease/oil we use to lubricate keyboard switches

https://www.krytox.com/en/-/media/files/krytox/krytox-product-overview.pdf

The mix result is more viscous/dense than the original, lets hope Krytox will
last longer than silicone...

 

[lexluthermiester]

I am using Alseye T9+ Platinum, at the time it was the only potentially good
paste quick available "at hand" here in Brazil, I prefer it over the Halnziye Hy-p13.


Composition: Silicone 10%, Carbon 45% and Metal Oxide 45%
Temperature Resistance: -30°C ~ 280°C
Thermal Conductivity: 13.5 w/m-k
Viscosity at Low Strain Rate 1,200 Pa-s
Viscosity at High Strain Rate 100 Pa-s
Specific Gravity 2.6
Volatile Content, 48 hours at 125°C 0.02%
Thermal Resistance at 25 N/cm² 0.04 °C-cm²/W
Bond Line Thickness at 25 N/cm² 0.02mm (0.0008 in)

Here is a Ryzen 5600X + Thermalright FC140, the foot/plate of the FC140 is a bit
convex, so I lapped just bit to make it more flat to better fit with the 5600X.
After almost 6 months, I decided to check visually(the TMP was OK) if the lapping
was correct and the state of the paste.
Well, the paste was +- ok, but at the highest pressure points it was drier.
So I decided to make an experience...
Removed the paste in the CPU/Cooler and mixed it with a bit of Krytox grease, it
is the grease/oil we use to lubricate keyboard switches

https://www.krytox.com/en/-/media/files/krytox/krytox-product-overview.pdf

The mix result is more viscous/dense than the original, lets hope Krytox will
last longer than silicone...

Those numbers look good. Would be interesting to test it.

EDIT:
Very pricey though. $20 for 2G is not very competitive.

NeweggBusiness - 9siv2yudcu2132

Shop NeweggBusiness.com for all your business tech needs! From workstations to servers, plus flexible payments & rewards—it's the perfect solution!

www.neweggbusiness.com

$9 is better, but still not a great price.

https://www.amazon.com/Compound-ALSEYE-High-Performance-Insulation-Processor/dp/B085DP1ZZT

 

[Vario]

Ab so loot lee, lol. Budget pastes have always had a place in every toolbox. I simply found it odd to use such an old block and test as an example of AS5s efficacy. You obviously love your tim and that's cool man, you do you! I like many others, didn't and don't like their wishy washy conductive/capacitive statement with my bare cpus, gpus and chipsets. Fortunately there was/is an easy solution back then and much more so now. Switch, to Ceramique a guaranteed non conductive tim then(that performed a bit better and didn't freeze). Or to something like MX-4 now. Thermal paste is an easy cheap temp improvement and nearly all of it is non conductive nowadays, or its not. Its generally not hard to find out as its a big marketing point.
As far as, has anyone ever damaged anything with AS5? Does it matter? Why worry about it when there's bucket loads of better, guaranteed non conductive options to choose from? If there's even the slightest chance that AS5 might damage their gear most people will use something else that doesn't purposefully try to confuse them with the conductivity, or capacitance silliness. Its a complete non issue now. LFaWolfs choice to use MX-4 speaks directly to that point.

I chose that particular benchmark for two reasons 1) it was a techpowerup review and 2) it compared directly with a standard "good brand product". The result is even closer on the Tom's Hardware benchmark. The reason I use AS5 is simple, I still have a couple tubes of it from 2004, and it still performs well, so I keep using it. It's a very stable product.

 

[lexluthermiester]

The reason I use AS5 is simple, I still have a couple tubes of it from 2004, and it still performs well, so I keep using it. It's a very stable product.

Keep using it, no reason not too.

 

[ThrashZone]

Why doesn't anyone mention AS-5? I've never had a single issue with it. Been using it for probably a decade now. It's my go-to when I don't have any liquid metal around, which I only get the Thermalright stuff since it's almost half the price of Thermal Grizzly and is probably the exact same shit anyways

edit: Go figure... as I'm writing this post, someone else finally posts about AS-5 lmao

Hi,
Not trendy
Think thermal grizzly hydronaut is the only good version they have "water blocks" or at least never read or had it scratch my blocks or cpu's

Think application matters more than the latest and supposed to be greatest paste, hell mx-4 2019 is okay paste still mx-5 was about the same just green

I've personally had best results with nt-h1 than nt-h2
Bad thing about nt-h2 is inflated price for worthless alcohol wipes which are very weak and take for ever to dry must be 50% or something.

 

[Pictus]

Those numbers look good. Would be interesting to test it.

EDIT:
Very pricey though. $20 for 2G is not very competitive.

NeweggBusiness - 9siv2yudcu2132

Shop NeweggBusiness.com for all your business tech needs! From workstations to servers, plus flexible payments & rewards—it's the perfect solution!

www.neweggbusiness.com

$9 is better, but still not a great price.

https://www.amazon.com/Compound-ALSEYE-High-Performance-Insulation-Processor/dp/B085DP1ZZT

Look what I found...

1.95US $ 50% OFF|17w/mk Lk-17 2g 5g Super High Performance Thermal Grease For Gpu Graphics Cpu Cooler, Ic, Lcd Led, Electronic Product Graphics - Fans & Cooling - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

 

[Shrek]

I still don't get 'more thermal conductivity is better'

See attachment: with 20 W/m.k I get 0.3C drop, with 5 W/m.k would give just over 1C drop

I would much rather have longevity than one degree saved.

Bond Line Thickness at 25 N/cm² 0.02mm (0.0008 in)

I actually used a lot thicker in my calculations.

 

[Pictus]

I would much rather have longevity than one degree saved.

I also prefer more longevity, but any improvement in temperature is always welcome.

 

[LFaWolf]

Look what I found...

1.95US $ 50% OFF|17w/mk Lk-17 2g 5g Super High Performance Thermal Grease For Gpu Graphics Cpu Cooler, Ic, Lcd Led, Electronic Product Graphics - Fans & Cooling - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

Are you sure that is not a counterfeit?

 

[Shrek]

I also prefer more longevity, but any improvement in temperature is always welcome.

True, but for me cost is also a factor; 2g does not seem a lot.

 

[lexluthermiester]

True, but for me cost is also a factor; 2g does not seem a lot.

It isn't. I don't generally buy anything unless at least 5 or 6 grams is included.

 

[Pictus]

Are you sure that is not a counterfeit?

The Alseye T9+ Platinum is not counterfeit.
The LK-17 we will only know by testing...

 

[Borc]

Has anyone used Halnziye paste? It looks like a generic brand but some claims it is performs very well.

I tested several of these Halnziye pastes on my notebook and to be honest I don't like them. They are on the low viscous side, they all struggled on my notebook CPU. Some of them are even worse than the MX-5. The best from them (Halnziye HY-P13) is in the average. On a desktop CPU HY-P13 should be fine, it should work much better there. Halnziye HY-P13 is used from many other brands by the way, for example IONZ IZP14. You can see in the albestech test this paste worked quite good on a desktop CPU.

As for notebooks I believe Honeywell PTM7950 might be the best solution in the market at the moment. I haven't tried this yet because it is hard to buy. However, from the results and comparisons I have seen I believe it can beat the best other pastes in the market. There is a paste and pad version, most of the users on reddit did use the pad version with good results. It's unclear which version works better if they are not performing the same.

 

[lexluthermiester]

Some of them are even worse than the MX-5.

You just had throw that in didn't you?

Once again, MX-5 is an excellent performer. Your experience was not the same as everyone else.

 

[Chomiq]

So is it discontinued or not? Should I not bother with it and return my tube and use mx-4 instead?