Cool looking new CPU block from MIPS Germany

[century child]

Sorry if this has already been posted and I overlooked it. I tried a search but didn't find anything.

AT added a product to thier shop today called the Iceforce from MIPS:

http://www.aquatuning.de/product_inf...ickel-POM.html

http://www.aquatuning.de/product_inf...ickel-POM.html

Evidently it's so new that it's not even up on the MIPS website yet.

In any case, I just hope it performs as good as it looks.

[Waterlogged]

Check the last page of the parts gallery.

[century child]

D'oh! Sorry.

[Waterlogged]

I wouldn't worry about it. The gallery pics from Six are nice but they don't really say "available for sale now" (even if it is from a questionable e-tailer), do they.

[scamps]

but it is available - at least @mips german shop http://www.mips-computer.de/sg_MIPSP...56.htm?ID=3525
and it is already on it´s way to me

though (or let´s better say because?) I already have the beta-sample

[Church]

I hope they send it to Martin for a testspin. They look rather OK on bling side, i wonder if performance is on par.

[Shteud]

churchy
german testers allready test it. nothing special.
http://www.hardwaremax.net/wasserkue...eforce-hf.html
good flow, midle temps (like hk 3.0)

[Church]

Hmm, if final retail product won't get improved thermals wise, then it doesn't look like member of very top blocks, bling and low pressure drop alone doesn't cut it. Temps imho should be within 1-1.5C from top, for other factors like flow resistance/bling and alike to overcome that.

[Waterlogged]

I hope they send it to Martin for a testspin. They look rather OK on bling side, i wonder if performance is on par.

Won't get tested for a while (if ever) if they do as he's on sabbatical.

churchy
german testers allready test it. nothing special.
http://www.hardwaremax.net/wasserkue...eforce-hf.html
good flow, midle temps (like hk 3.0)

:facepalm:. . .this is my general reply when I see "German tester" as I've not seen a good test from a German tester since Radical_53 last posted test results here. Far too many variables to make any kind of conclusive results. 1 pump, 2 rads (1 3x120 and 1 3x140) with highly questionable fans, GPU as part of the loop and Prime95 will all make me put this test, like every other German test I've seen over the past 3½-4½ years, in the round file. If a Euro tester wants to get anything other than a :facepalm: from me, all they have to simply do is duplicate Martin's or Skinnee's testing methodology and remove as many variables as humanly possible.

[Shteud]

Waterlogged
he-he. they too close with producers...

[AVulgarPicture]

Exactly what waterlogged said.

Also, this block is hot sexy sex.

[scamps]

[COLOR="#000000"]...If a Euro tester wants to get anything other than a :facepalm: from me ...

Fortunately no "Euro tester" worries about your facepalm ...

[AVulgarPicture]

Yeh, but they should.

[Church]

scamps: +1 to WL. It's just that sometimes for few LC components about the only reviews tests that can be found are from those testers, so it would benefit community a lot for them to get more pedantic as to how testing gets done.
Yes, using proper testing methodology multiplies needed work/time for tests that much more (eg. many remounts, retests at different possible configurations of several variables at play), but it 1) eliminates lot chances of introducing some error affecting results (eg. bad or best TIM mount, ambient temp fluctuation during test time), so that results are MUCH more trustable in product choice, 2) gives much more thorough review of product (eg. how it performs at different pumping power or at different rad efficiency or different heat load or at different deltas or with different fans/noise level or with different mount pressure and so on and on ..).
I don't think that there is much of 'purchased reviews', imho german testers also have their pride .. but their testing in my eyes is way less actually usable for at least me in evaluating/comparing/products products and so far situation is as it is - i have only limited and not very trustable info on plethora of products, that i'm not willing much to buy myself or suggest to others, instead limiting only to products that are "safe choice" and reviewed in trusty and thorough enough way, even if in real life some of those german LC products might be competitive enough or even best available choice - i just don't actually know about their capabilities for sure .. yet? (with hope of status quo change in future).
P.S.
Even more important to lessen impact of luck/errors is at current LC state, eg. when top blocks are within 1-2 degrees, and simple bad luck in TIM application with range much more then that might turn waterblock relative performance to turn out fully different.

[chaotic]

Any internal pics?

[scamps]

@churchy
I agree 100%. But this is no question of nationality but only of the accuracy and reliability of a tester. Perhaps somebody reminds "Hesmelaugh" who used heating equipment already before martin (as far as I remember). From Switzerland ...
I really enjoy every test of martin and I think we all can trust in him absolutely. But even martin can´t avoid getting test samples that are better than real shop quality. And that is one of the biggest faults in his methods. To test real life you would have to buy at least two samples on your own somewhere - but who yould be able and willing to give all the money for that?
I had to learn on my own that a preview sample of a big (non-german!) manufacturer was almost 2° better than the same cooler bought in a local shop. And I am sure that is no exception.

At the end the differences in background of the tester, methods, used parts and equipment teached me to read at least three independent reviews before forming my opinion about a product. And after that I often verify this with my own testing.

By the way and to get back to topic: My own testing shows less than 1 K between Mips Iceforce and EK Supreme HF with jetplate #6. And a big increase of flow inside my testing loop (EK: 158 l/h, Mips: 194 l/h). In the meantime 7 runs each ...

@chaotic
There was some trouble regarding an Alphacool patent ...


€dit 532c: hardwaremax.net and Maexi are from Austria not friends of Germany and german products in every case.

[Church]

I'm sorry to all for dragging (and with this post as well) deep in offtopic this thread.
Scamps: if you liked such combination of properties, imho DT 5Noz block may be potentially interesting for you to try out as well. Not as fine on bling side though.

[Waterlogged]

Fortunately no "Euro tester" worries about your facepalm ...

...because the majority of the ppl that read their tests have decided that massive quantities of sub par mediocrity reviews are better than fewer done properly, so why would they worry about 1 outspoken person.

@churchy
I agree 100%. But this is no question of nationality but only of the accuracy and reliability of a tester. Perhaps somebody reminds "Hesmelaugh" who used heating equipment already before martin (as far as I remember). From Switzerland ...
I really enjoy every test of martin and I think we all can trust in him absolutely. But even martin can´t avoid getting test samples that are better than real shop quality. And that is one of the biggest faults in his methods. To test real life you would have to buy at least two samples on your own somewhere - but who yould be able and willing to give all the money for that?
I had to learn on my own that a preview sample of a big (non-german!) manufacturer was almost 2° better than the same cooler bought in a local shop. And I am sure that is no exception.

At the end the differences in background of the tester, methods, used parts and equipment teached me to read at least three independent reviews before forming my opinion about a product. And after that I often verify this with my own testing.

By the way and to get back to topic: My own testing shows less than 1 K between Mips Iceforce and EK Supreme HF with jetplate #6. And a big increase of flow inside my testing loop (EK: 158 l/h, Mips: 194 l/h). In the meantime 7 runs each ...

I remember HES and he was starting to tweak the reviews into shape where they would stand up to scrutiny but then he just disappeared.

I don't know how many times we have to remind ppl that you should never expect temps similar to someone else, tester or not. You do realize that 2°C is well within mounting variations (why do you think Martin, Vapor and Skinnee do 5+ mounts?) and could also be partly cause by just a little too much TIM?

Sounds like you do better testing than the Yahoo's that call themselves testers over there. Maybe you should step up and get some good testing gear and put them to shame?...and maybe shut me up in the process.

[scamps]

...Scamps: if you liked such combination of properties, imho DT 5Noz block may be potentially interesting for you to try out as well. Not as fine on bling side though.

Thanks for this hint! Did not see this nice one before. But as you wrote: not yet as fine on bling side (I would prefer very much smaller typo on it), but -wow- great performance. But yet not available here afaik.

...but then he just disappeared....

SHANE, even Waterlogged wants you back! Come on!!!

[Waterlogged]

SHANE, even Waterlogged wants you back! Come on!!!

He still had a little ways to go but he was definitely heading in the right direction.

I heard he had started taking some pretty bad abuse because he was changing his testing method from the "Euro sub standard" to something closer to what Martin and Skinnee had back then and felt he couldn't win, so he quit.

[Utnorris]

So is this block strickly Intel?

[Church]

Why abuse someone for doing more thorough work in proper way? i fail to get it.

Utnorris: you can check among new products at least @AT.de. Three variations. One for AMD, another - for lga2011, and third - for all the intel sockets but lga2011.

[Utnorris]

They have one:

http://www.aquatuning.de/product_inf...km6q16eic80r1r

Too bad they don't have it on the US AT site. I would get that, looks like something out of Star Trek.

[AVulgarPicture]

i just realised. this block totally doubles as a ninja star.

[Martinm210]

But even martin can´t avoid getting test samples that are better than real shop quality. And that is one of the biggest faults in his methods. To test real life you would have to buy at least two samples on your own somewhere - but who yould be able and willing to give all the money for that?
I had to learn on my own that a preview sample of a big (non-german!) manufacturer was almost 2° better than the same cooler bought in a local shop. And I am sure that is no exception.

At the end the differences in background of the tester, methods, used parts and equipment teached me to read at least three independent reviews before forming my opinion about a product. And after that I often verify this with my own testing.

By the way and to get back to topic: My own testing shows less than 1 K between Mips Iceforce and EK Supreme HF with jetplate #6. And a big increase of flow inside my testing loop (EK: 158 l/h, Mips: 194 l/h). In the meantime 7 runs each ...

While I completely agree sample variance is always possible (I have seen it before..yet most samples I see are clearly unopened retail packages) and you should always look at several reviews since it provides more conditions to consider, I don't know any 3rd party tester who has the luxury testing multiple block samples on multiple processor samples, so what you are calling a "fault" is really the only practical option? I would take it several steps further and say you don't know for sure unless you test on multiple processor samples and across a range of processor types. We all know that the IHS is not perfectly level on the processor and that varies from sample to sample. I can also confirm what you get on LGA1155 will be completely different on LGA2011, etc. So.... you do what you can and hopefully people appreciate what you can contribute and not complain. If they do..here is a quarter...call someone who cares..

I would also like to learn more about your test methods such as:

1)How many retail samples did you buy out of pocket to come up with your 1K result?
2)Care to share some details about the testing methodology used?
3)Where are the 7 runs to show mounting error?
4)What sort of logging tools did you use?
5)What sort of thermal sensors?
6)What loading program and for how long?
7)What did you use to measure mounting force?
8)What processor?
9)What thermal paste?

I didn't follow your testing methodology details or results perhaps because there were NONE that you were willing to share?....


Back on topic.

The MIPS block looks nice, I like the nozzle plate setup and ninja hold down..