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

[lexluthermiester]

Wow, MX-5 is truly an astounding product! Pumped out literally in hours. I'm already back to 95C.

Bullshit! IF your system is running at 95C, you have a problem other than the TIM.

You don't get it, do you?

Yes, that's got to be it...

I think it's time to break out the hip-boots, cause it's getting deep in here.

 

[Mussels]

When you start out with things like this, you immediately disqualify anything that follows. PC's, the VAST majority of the time, are operating at "room" temps. And for the record, oil "seperation" doesn't happen until well above 150C for most substances that would be used in a PC/technology application. It's a chemistry thing.


That looks interesting. Would you have a link for that screenshot?

He's not saying the room temperature the PC is in matter as part of the test, he's saying you cant test a thermal compound at 20C and make a final judgement when it could be used at 100C

Bullshit! IF your system is running at 95C, you have a problem other than the TIM.

Yes, that's got to be it...

I think it's time to break out the hip-boots, cause it's getting deep in here.

But... that sort of temperature is totally normal on a lot of modern laptops?
Hell even some of the modern desktops.

Go spend some time in TPU's throttlestop forum, look at all the users upset their brand new intel gaming laptops run at 90-100C out of the box, voided their warranty with new thermal paste only to find that nothing works short of underclocking and undervolting.

 

[lexluthermiester]

But... that sort of temperature is totally normal on a lot of modern laptops?

Only when the cooling is either poorly engineered(Apple, Dell, HP) or clogged with dust/crap restricting airflow. Otherwise, no.

Hell even some of the modern desktops.

See above, but add overclocked and poorly ventilated.

Go spend some time in TPU's throttlestop forum, look at all the users upset their brand new intel gaming laptops run at 90-100C out of the box

You assume I don't. Most of those are situations of poorly designed/implimented cooling. Fan profiles for OEM laptops can sometimes be badly configured. The TIM has little to do with those situations, demonstrated by...

only to find that nothing works short of underclocking and undervolting.

...this.

 

[cvaldes]

But... that sort of temperature is totally normal on a lot of modern laptops?
Hell even some of the modern desktops.

Yes, it is.

As I pointed out earlier, my Mac mini 2018 (Intel Core i7) hovers around 100 °C when the CPU is maxed out, like during a Handbrake encode.

And same with my Acer Swift 3 notebook. Both of them have some sort of thermal grease on the CPU. I don't know what they are using, but it's probably not MX-5 based on the fact that both of them released before MX-5 came to the market.

These systems are designed to use their CPUs to basically the maximum limits of their TDP. And there's nothing new about this.

In fact, notebook computer manufacturers normally don't refer to these systems as laptops. They have been well aware for about two decades that these systems run so hot under load that you might burn yourself if you put your computer on your lap. And Joe Consumer would rather pay $500 instead of $1500.

In any case, these systems are designed for a wide variety of users, usage cases, and budgets. Not everyone is going to own a perfectly designed, perfectly vented system that has a large buffer for thermal capacity when they just want something portable under 1.5 kg.

Let's remember that most consumer computers are notebooks. People want something that might run 12+ hours on a single charge rather than a 3 kg notebook that basically needs to be plugged into a socket.

Sure, maybe a notebook could run cooler if you put a big fat heatpipe in it with 2-3 fans but then the owner might be lugging around something an inch thick. Or the manufacturer could keep the svelte chassis and throttle the CPU (or GPU) at 80 °C hence partially nerfing the components' maximum design capabilities.

A funnier note is that many Mac minis end up in data centers and other air conditioned server rooms. Running full bore at 100 °C is what many of these units do.

The funniest thing is that with Apple taking control over their own silicon, their stuff is better from a performance-per-watt metric than anything AMD or Intel can offer.

If a thermal grease can work at ~70 °C on some custom built deskto PC but breaks down at >90 °C on a mass-market notebook, that basically limits its applicability.

Shall we guess where MX-5 is?

Let's remember that Arctic is the one who took MX-5 off the market. Not CallandorWoT. Not Mussels. Not W1zzard. Not Chomiq. Not me.

 

[TheEugeneKam]

Laptop and fans are fully cleaned a few days ago as I thought that might be the problem and not tfx. Alas it was tfx. Most modern laptops that are intel are designed to run up to 90-100c. That’s normal spec. They throttle at a 100. my desktop is using my-h2 and 5900x sits at 70 max with amd pbo2 and auto under volt thing setup.

@Lex not a troll account, just thought I would share my experience so far. Also who creates troll accounts to post about thermal paste…
You are welcome to speculate about shadow accounts and I’m glad mx-5 works for you. I am sharing my experience so far 2016 MacBook Pro in particular. I think you might either use laptops that have better cooling or built. We use dell latitude 14 inch at work and I’m assuming the pinpoint will be similar on that.

Most workplaces give you shitty dell or HP, we use a blend of these after 2020. Before that we were strictly dell. A lot of them do run hot but thats the particular market I’m talking about. Not msi gaming bricks. I’m sure their cooling is significantly better, they are more of a niche product rather then stained laptops most people have for work or Macs

I have no horse in this race apart from telling people what I have tried so far. I already spent the money on all these pastes which will work fine for desktops.

Taking off max-5 was terrible. It’s a gooey stringy mess

syy 157 seems softer then tfx did it was closer to “normal” paste then gummy tfx texture. So a bit easier to apply. Currently waiting another 5 min as per instructions before assembly.


UPDATE:
so the paste seems to run a bout a degree cooler then mx-5. Lets see how long it lasts.

 

[lexluthermiester]

Also who creates troll accounts to post about thermal paste…

It happens. Some people are really that petty.

@Lex not a troll account, just thought I would share my experience so far.

Fair enough. However, saying...

Wow, MX-5 is truly an astounding product! Pumped out literally in hours. I'm already back to 95C. I think SYY is the next one I'm going to try today or tomorrow.

..this. TPU is no stranger to fanboy nonsense, and there's been plenty of it in this thread.

UPDATE:
so the paste seems to run a bout a degree cooler then mx-5. Lets see how long it lasts.

Then there's this. So it seems the MX-5, nor the SYY, is the problem.

 

[TheEugeneKam]

Sorry, I should have been more clear about what I meant to say runs a degree cooler than the initial MX-5 application which dropped about 20C from TFX that pumped out. So starting point between the two is pretty much the same within a margin of error. MX-5 pumped out within hours of using it for me. This one is holding up the same so far.

The poor thermal envelope of macs is the problem for sure, i exp3ect it to run hot, just not hit 100C instantly under load, which is what my old TFX paste did and MX-5 got close to.

 

[SpittinFax]

Today I replaced the MX-5 on my RX6600 with Noctua NT-H2 and the hotspot temperature dropped from 82C to 81C. I'm calling that basically a tie. So even though MX-5 has issues with paste consistency, there isn't any indication that it affects thermal performance. In fact I'd even argue that the MX-5 performs really well considering the cost per gram is around 66% lower than NT-H2. There still is the question of application lifespan but I personally haven't seen any changes in performance over time.

 

[lexluthermiester]

MX-5 pumped out within hours of using it for me.

I have never seen this effect with ANY paste. After discussing the subject for many years with many people and trying many, many times to replicate this "effect" and failing to find anything to substantiate the many claims made, I am always extremely dubious whenever someone chimes in and claims to have witnessed it, especially when they claim it happened in a matter of hours. What is far more likely is that the very flat surface of the CPU heatplate is simply squishing all the TIM out as it is secured to the CPU die or it was not applied properly. This would happen with any TIM, which is why your temps haven't changed much. Thermal interface materials are NOT miracle compounds. They can only act as a medium to conduct heat through. If the cooling apparatice is deficient, the TIM is not to blame, the cooling system is, which means the blame falls on the manufacturer of the cooling system in question, assuming the TIM was properly applied to the surface.

 

[Mussels]

I have never seen this effect with ANY paste. After discussing the subject for many years with many people and trying many, many times to replicate this "effect" and failing to find anything to substantiate the many claims made, I am always extremely dubious whenever someone chimes in and claims to have witnessed it, especially when they claim it happened in a matter of hours. What is far more likely is that the very flat surface of the CPU heatplate is simply squishing all the TIM out as it is secured to the CPU die or it was not applied properly. This would happen with any TIM, which why your temps haven't changed much. Thermal interface materials are NOT miracle compounds. They can only act as a medium to conduct heat through. If the cooling apparatice is deficient, the TIM is not to blame, the cooling system is, which means the blame falls on the manufacturer of the cooling system in question, assuming the TIM was properly applied to the surface.

Wouldnt the multiple people in this thread stating otherwise, be an obvious clue that there are things outside your experience?

It's a lot more likely than your claims of trolls, fake accounts and bizarre accusations you've made out of fear that you may either be wrong, or not know something

You're in complete denial that MX5 is a faulty product, it's been EOL'd due to it. No shit, the people with the bad batches are going to experience these problems and your denial and accusations to the people experiencing it are just fucking weird.

 

[lexluthermiester]

Wouldnt the multiple people in this thread stating otherwise, be an obvious clue that there are things outside your experience?

It's a lot more likely than your claims of trolls, fake accounts and bizarre accusations you've made out of fear that you may either be wrong, or not know something

It's an example of uninformed people making assumptions without evidence and possibly even the Mandela effect. And when I smell moose-muffins, I call moose-muffins.

You're in complete denial that MX5 is a faulty product

Yes, that must be it, denial. Couldn't be something else could it? Arctic had a faulty batch, 1(one), which they attempted to correct before being blasted and ridiculed by an ignorant cancel-culture public, no doubt spurred on by competitors.

Yes, I'm standing my ground and not willing to accept nonsense. This not denial, it is confidence in the scientific method. If there are users who have a bad batch they should be contacting Arctic about it, not coming here whining and badmouthing Arctic.

 

[Chomiq]

First of all, you are conducting your experiment at room temperature. That is not the typical operating environment of a CPU.

The oil separation is probably happening at higher temperatures and might be exacerbated by temperature deltas between high operating temperature and when the system is turned off.

The Core i7 in my Mac mini 2018 will reach almost 100 °C, the maximum operating temperature of the CPU. Apple designed this system to use the CPU's full TDP. What is the ambient temperature in the room at night? Maybe 20 °C thus an eighty degree range. And Apple quotes a lower temperature for storage.

According to Arctic's note, this oil separation is causing premature drying of the thermal compound. This likely happens after extended usage, not something that happens immediately after application otherwise they would have noted it in their lab tests with prototype pastes.

Remember that thermal grease isn't the only thing that suffers from oil separation.

A slab of bacon has been hanging in room for months and months without oil separation. Put that slab on your kitchen counter overnight. Nothing will happen. Put that slab in roasting pan and throw into a 90 °C (195 °F) oven overnight. There will be a puddle of oil. That's oil separation. Throw a chocolate bar into the refrigerator for a few weeks; that white bloom on the outside is cocoa butter that has migrated to the exterior.

If you cook or bake regularly, you'll be quite familiar with oil separation and how materials vary in behavior based on temperature and age. Heavy cream whips better when very cold but egg whites are better when warmed up. In fact, older egg whites are better for whipping up than whites from freshly laid eggs. There's a lot of chemistry and physics involved in cooking; a good cook is a keen observer even if they don't have a chemistry or physics degree.

You're missing the part where people reported oil separation on MX-5 immediately after coming out of the tube.

My tube was fine and I see no problems when running with my 5800X. If I see any issues I'll just switch over to a different paste.

Arctic couldn't keep the consistency and they did the best they could - EOL it because word around the block was that MX-5 can be runny, gummy sh*tshow.

 

[freeagent]

If TFX pumped out there is something else going on. TFX is very thick, it doesn’t run. It’s way thicker than SYY.

 

[95Viper]

Stop the insults.
Stay on topic.
Report the problem and do not become the problem!

 

[Shrek]

I have never seen this effect with ANY paste. After discussing the subject for many years with many people and trying many, many times to replicate this "effect" and failing to find anything to substantiate the many claims made, I am always extremely dubious whenever someone chimes in and claims to have witnessed it, especially when they claim it happened in a matter of hours.

I tend to agree; as I understand it, it is a very slow process that results from differential thermal expansion. Not something that can be resisted as the forces would be extreme. That is where I found the DOWSIL data so intriguing as it is suggestive that around 10,000 cycles may be involved in such a mechanism.

I am still wondering if wetting plays an important role whereby the thermal paste can make its way back in by capillary action; this is all conjecture, but one has to start someplace.

We seem to concentrate excessively on thermal conductivity, perhaps because that make things easy; TC-5026 has a thermal conductivity of 2.87 W/m.K. If a thermal conductivity of 20 W/m.K resulted in a 0.2°C drop, 2 W/m.K would result in a 2°C drop; well worth it in my opinion if the result is longevity.

 

[TheEugeneKam]

@Shrek in Canada, to ship this item and import fees costs more then the paste itself. SYY is holding up fine for now for me. I haven’t replaced mx-5 on my dell yet as it works fine in that application.

I know pump out takes a bunch of cycles to affect the machine. It wasn’t the case with mc-5 which is odd. I wonder if it’s stringy consistency has something to do with that as I have never seen a paste be that stringy before. It was still wet when I took it off my Mac so drying was defenetly not the case. Eh who knows, the product is EOL anyways so this will become non issue fairly soon.

 

[Shrek]

For me it is a learning process, so of utility whether or not it becomes obsolete.

I have 120g of GD900 so may never need to get any thermal paste again.

 

[P4-630]

For me it is a learning process, so of utility whether or not it becomes obsolete.

I have 120g of GD900 so may never need to get any thermal paste again.

Had no idea what it was, did a quick google search, aliexpress on top, so guess it's some chinese brew..
Each their own..

 

[Shrek]

 

[Boombastik]

We seem to concentrate excessively on thermal conductivity, perhaps because that make things easy; TC-5026 has a thermal conductivity of 2.87 W/m.K. If a thermal conductivity of 20 W/m.K resulted in a 0.2°C drop, 2 W/m.K would result in a 2°C drop; well worth it in my opinion if the result is longevity.

The conductivity is 2.87 but the thermal resistance is only 0.03 from 20 psi to 80psi and for that reason beats thermal paste that have 5-6 w/m. K. Is the only paste that the thermal resistance don't change with pressure. For example akasa t5 pro grade plus have 5.5 conductivity and thermal resistance 0.04.

 

[Shrek]

While it is true

• Thermal paste is about lowering temperature

what use is that if

• The lowered temperature only endures a few months

 

[Pictus]

Sorry, I should have been more clear about what I meant to say runs a degree cooler than the initial MX-5 application which dropped about 20C from TFX that pumped out. So starting point between the two is pretty much the same within a margin of error. MX-5 pumped out within hours of using it for me. This one is holding up the same so far.

I would try something different like Shin-Etsu 7921

6.49US $ 25% OFF|1g 2g 3g 5g 10g Shin-etsu 7921 7868 Syringe Thermal Conductive Grease Paste Plaster Cpu Pc Notebook Cooling Original Authentic - Pc Components Cooling & Tools - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

It is super dense and will not pump out, but very hard to apply.
The guy in the video used a hair dryer, but I would keep the hair dryer on...

 

[GerKNG]

Thermal Paste pump out is generally a heatsink issues (some paste can handle it better but all will pump out if the heatsink is the problem which it is in 95% of the cases)
i have dealt with it so often over the years with tons of GPUs (i repasted around 20-30 GPUs with almost all branded thermal pastes and it was almost the same.)
if the heatsink is rather thin or the GPU gets very! hot the expansion of the silicon and copper under a heat load causes the pump out effect.
one of the worst cards that i had to deal with was the 2080 Ti AMP Extreme.

repaste with MX4, completely gone within a week, Kryonaut 2 weeks, NT H2 2 weeks, the brown thermalright paste 2 weeks, GC Extreme a week... Same goes for silicone free pastes like hydronaut which will just turn into soupy sand.
if a heatsink is heavy and thick it's no problem even with the cheapest stuff (for example my 3090 Gaming X Trio) Arctic MX4 lasted almost a full year with no problems.
same with my 6900XT Nitro which was my first repaste with Arctic MX5. and guess what! a year later the temps are basically identical.

even the highest end stuff like kryonaut extreme will pump out within a week on bad heatsinks.
from my experience pump out is a bit less bad if you don't tighten the heatsink as much as it was from the factory. (just 1/8 rotation less can help sometimes)

 

[TheEugeneKam]

Good tips! I remember shin-etsu, I think they were rumoured to be OeM for Lenovo a while back. Could be wrong though.
So far SYY is holding up solid in my end. Mx-5 is working fine in dell as well.

 

[Ruru]

Wait, what? EOL already? What I've missed?