Review Arctic Liquid Freezer III AIO Review: To put it bluntly, the Liquid Freezer III is unimpressive

[thestryker]

well thats annoying i wonder if its the lfIII materials for the frame looks like plastic or rubber possible the material isnt ridged enough ? or expanding with the heat ?.

It does seem like it might be related to the retention design as opposed to the bracket itself. I think GN hypothesized this in their teardown video. The pressure looks like it might be biased to one side which I think could be due to ARL and other potential future Intel CPUs. The CPU cores will be along an edge rather than in the center so it might just be intentional.

 

[beyondlogic]

It does seem like it might be related to the retention design as opposed to the bracket itself. I think GN hypothesized this in their teardown video. The pressure looks like it might be biased to one side which I think could be due to ARL and other potential future Intel CPUs. The CPU cores will be along an edge rather than in the center so it might just be intentional.

Honestly feels like intel didn't want to do any retooling with lga 1700 I wonder how much testing they did with mounting pressure on this platform. As for the void warranty they keep spouting my answer to intel is give us a solution to the problem that exists because they dropped the ball on testing. It's either that or they knew it was a problem but just didn't care.

 

[thestryker]

Honestly feels like intel didn't want to do any retooling with lga 1700 I wonder how much testing they did with mounting pressure on this platform. As for the void warranty they keep spouting my answer to intel is give us a solution to the problem that exists because they dropped the ball on testing. It's either that or they knew it was a problem but just didn't care.

I always assume when it comes to a lot of the odd choices Intel makes it comes back to OEMs. The tooling was probably cheaper so that's what they went with. The one theyre using for LGA1700/1851 is passable on well built boards, but it's still a bad design. The best LGA ILM I've used is the one on LGA2011 and I really wish they'd have switched to that. Everything about that mounting system is great really.

 

[beyondlogic]

I always assume when it comes to a lot of the odd choices Intel makes it comes back to OEMs. The tooling was probably cheaper so that's what they went with. The one theyre using for LGA1700/1851 is passable on well built boards, but it's still a bad design. The best LGA ILM I've used is the one on LGA2011 and I really wish they'd have switched to that. Everything about that mounting system is great really.

Probly will for next gen of sockets I wouldn't be shocked if they just copied the so called warranty void ones they keep spouting

 

[-Fran-]

Probly will for next gen of sockets I wouldn't be shocked if they just copied the so called warranty void ones they keep spouting

Legally speaking, as it is built now, they can't. They'd need to build something that explicitly breaks when you take out the ILM to push "warranty void" shenanigans. They can, but it would be a very bad business move in the DYI. That's also why both AMD and Intel have fuses inside the CPUs when you truly OC them, so they know if you have or not.

Regards.

 

[Albert.Thomas]

I don't understand this since I own and run an Artic Freezer III 360 ARGB? Not saying the numbers are not correct for an Intel processor but my 7950x at 179 watts runs at 5.7 MHz on 7 cores while the rest run at 5.4 under load?

I have only two thoughts here:

1 - Different systems. Unless you're comparing results with a system with the same components at the same settings and at the same ambient temperatures, results can't just be compared like that.

2 - One result in isolation doesn't tell a story. How well does a good air cooler like DeepCool's Assassin IV or ID-Cooling's A720 perform on the same setup?

 

[Albert.Thomas]

For anyone still reading this thread:

My next review of Arctic's 36 air cooler should be posted in the next week or two. It will have a more in-depth section covering testing methodology. At least on my test system, it performed a little better than the LF3 240 - and its direct touch heatpipes make it especially efficient in a way that I haven't tested in detail before, but is going to change how I record test results for future reviews of air coolers. I have a section at the end dedicated to that interesting finding

 

[crushdepth]

Very surprised by these results. I wonder if somehow it is a platform difference? I installed a 280 on an AMD 5950x a couple of days ago and the performance is just outstanding.

 

[kazuviking]

I have an idea that is iether very stupid or "cracked". Apparently you can use the Freezer i13 x's stock and lga1700 mounting kit to mount the LF3 to LGA115X and 1200 sockets. So my idea is: Using the i13 x's mounting kit with the thermal gizzly or thermaltakes contact frame. It would be interesting to see the result of this. Imo it would decrease the temps by 1-3C but nothing gamechanging.

 

[Albert.Thomas]

We have an update, folks :

https://www.tomshardware.com/pc-com...-testing-methodology#xenforo-comments-3852191