AiO Water Cooling

Soldato
Joined
1 Jul 2011
Posts
8,632
The test was sound and reliable and when he repeated it on a more demanding system the delta increased! I.e. the air coolers performed even worse.

All you do is repeat the same thing over & over based on your own tests and claim "logic and intelligence" when anyone disagrees with you but when presented with evidence to the contrary, and when excuses run out you just scuttle away from the discussion.

I despair thinking how many people have made purchasing decisions based on your advice :(
You are making statements you cannot support with facts. Test procedure does not tell us the things we need to know / have in order to know if testing is in fact accurate.
Where and when is the air temperature being taken? If it's to the air temp into cooler/radiator and taken at same time as CPU temp the delta is likely to be way off.

OC3D data does not tell use when, where and at what time it was taken relative to CPU temp reading.

So what is his delta based? Also what was the noise level of each cooler when the delta temps were taken?

All of the above info I don't see in OC3D testing is given in Anandtech testing data.
 
Associate
Joined
15 Aug 2018
Posts
1,133
If you watch the video I've linked he says its the max temp when running benchmarks. Why would he use the same test bench, same cpu, same cpu clock, same cpu volt then not take the measurements at the same point... ie max temps. You are flogging a dead horse big time and it looks ridiculous. I'm not going to reply again cos its pointless.
 
Soldato
Joined
1 Jul 2011
Posts
8,632
If you watch the video I've linked he says its the max temp when running benchmarks. Why would he use the same test bench, same cpu, same cpu clock, same cpu volt then not take the measurements at the same point... ie max temps. You are flogging a dead horse big time and it looks ridiculous. I'm not going to reply again cos its pointless.
Fair point. Problem is he does not tell use where and when he is recording his '20c ambient' air temp, and I'm not the kind to give any credence to data I don't have all the details of how it is collected. For example I see every single cooler has a 20c ambient temperature, and from years of testing I know the air temp in to a bunch of different coolers is not always the same .. like all his recorded CPU temps are different and recorded to 0.25c accuracy, but the air temp is always 20c, never even off 0.25c and being 20.25, or 19.75c. That alone throws ups red flashing lights and sirens blaring about credibility of testing.

I have talked to a lot of reviewers who test coolers and several have changed their test procedure to monitor air temp into cooler, but some actually told me it was too hard to keep track of air temp into cooler because it was always changing so they continue to use room ambient.

The same applies to testing done in a complete built system claiming it 'gives reader real world testing results'. Sorry mate, the only 'real world' is the reviewer's test system .. or one setup exactly like it.

Anyone who has ever done any scientific testing knows the key to a good test is that someone else can setup the same procedure and get the same results. That is the reason for detailed test procedures being written up for reader to see.

It's all about repeatably, and saying it is repeatable because reviewer uses the same data over and over with new data added for new cooler being reviewed is not 'repeatable' results, it's copying the graph and adding one new test to it. I'll add to that that any test results with temps recorded to 0.25c accuracy and testing the same cooler a week later or a month later, even re-mounting re-testing the same day will have at least 0.25c different results. Temps can easily vary +/-0.5c just because every TIM seat is not the same or air humidity difference from cool, damp early morning to hot, dry afternoon .. even though the room and air into coolers is same temp both morning and afternoon. Heck, even barometric pressure can effect test results. Every time I run a test session I always run a start and finish test using a cooler that has been tested before to make sure the tests done during that session are accurate. But many reveiwers do not bother to do accurate testing, so beware of them and stick to the ones that are detailed .. because their results will be much more accurate. ;)
 
Soldato
Joined
19 Nov 2009
Posts
4,387
Location
Baa
You are making statements you cannot support with facts. Test procedure does not tell us the things we need to know / have in order to know if testing is in fact accurate.
Where and when is the air temperature being taken? If it's to the air temp into cooler/radiator and taken at same time as CPU temp the delta is likely to be way off.

OC3D data does not tell use when, where and at what time it was taken relative to CPU temp reading.

So what is his delta based? Also what was the noise level of each cooler when the delta temps were taken?

All of the above info I don't see in OC3D testing is given in Anandtech testing data.

You are not interested in facts. All you're interested in is your own supposed test results which back-up your faith-like worship of air coolers.

I'm not talking about OC3D btw. Someone presented you with testing data somewhere else which disproves your theory. You tried to dismiss it with the usual waffle but the guy came back to you with more results and blew your theory clean out of the water. Strangely, you left the thread after that.
 
Soldato
Joined
1 Jul 2011
Posts
8,632
You are not interested in facts. All you're interested in is your own supposed test results which back-up your faith-like worship of air coolers.

I'm not talking about OC3D btw. Someone presented you with testing data somewhere else which disproves your theory. You tried to dismiss it with the usual waffle but the guy came back to you with more results and blew your theory clean out of the water. Strangely, you left the thread after that.
If I did I'm sorry. Who posted different test data? What post is it in?
 
Soldato
Joined
19 Nov 2009
Posts
4,387
Location
Baa
If I did I'm sorry. Who posted different test data? What post is it in?

It's on another forum I'm a member of which you used to pester anyone asking for CPU cooling advice. I'll post a link if it doesn't break any rules.

Anyone interested can find it by googling "doyll aio clc". A warning though, there are a lot of search results - he's been posting this stuff all over the shop for years.
 
Soldato
Joined
1 Jul 2011
Posts
8,632
It's on another forum I'm a member of which you used to pester anyone asking for CPU cooling advice. I'll post a link if it doesn't break any rules.

Anyone interested can find it by googling "doyll aio clc". A warning though, there are a lot of search results - he's been posting this stuff all over the shop for years.
I assume you realize just because I post about what CLCs are and how they compare to air (and sometime custom H20 loops) in no way indicates that they are not in fact truthful posts usually supported by data like here.

I don't make statements like 'Someone presented you with testing data somewhere else which disproves your theory.' I post data or links to data that supports what I say.

Here's first one I find googling 'AIO vs CLC, what's the difference'
https://www.overclock.net/forum/10-amd-cpus/1623479-clc-aio-cooling.html

Or how about one talking about the differences between CLC and custom loop saying:
"Saying that watercooling loops and closed loop liquid coolers are both watercooling would be similar to saying a high-end Mercedes SLK and an entry level Kia Sonata are both cars."
http://www.tomshardware.co.uk/forum/281201-29-liquid-cooling

Here is link to the difference between Asetek and CoolIT OEM CLCs sold under many different names like 'Corsair' who does not make any CLCS at all, but only puts there labeling on other OEM CLCs.
https://www.reddit.com/r/buildapc/comments/4b6xtk/discussion_a_word_of_warning_about_corsair_clc/

Just for info, most air coolers are made by the company who's name is on them. Same applies to custom loop components.

Also for info, Asetek has the US patent on the pump on waterblock in sealed loop concept, meaning in USA or any country honoring US patents the only CLCs with pump on waterblock you will see are either made by Asetek ro paying royalties to Asetek.
 
Last edited:
Soldato
Joined
19 Nov 2009
Posts
4,387
Location
Baa
*snip*

I don't make statements like 'Someone presented you with testing data somewhere else which disproves your theory.' I post data or links to data that supports what I say.

You've been posting the same table of data for years.

Don't you think it's time to start listening to other people when they tell you their results are vastly different to yours?
 
Soldato
Joined
18 Oct 2002
Posts
4,093
Location
Devon
I thought I'd try a CLC when I recently spotted a good deal on a warranty replacement (unused) BeQuiet Silent Loop 240 on the members market, this coming from a Thermaltake Macho HR-02.

Pro's
Cooling - Temperatures dropped by ~10% idle and load with similar low noise fan profiles, though this did not equate to a higher stable overclock.
Weight - I'm sure it's not an issue but effective air coolers tend to be tall and heavy putting stress on the motherboard.
Aesthetics - Not a huge deal considering I don't have a windowed case.

Con's
Noise - The pump started off silent but within 2 weeks became audible at idle, I don't mind noise when gaming but when idle I want silence.
Reliability - Anything with a moving part will eventually break.

CLCs by their nature are a compromise and will never be able to offer the exceptional cooling and noise profile of a full fat custom loop. Many people are happy with that compromise given the pro's and I'm glad I got to try one out, but when I upgrade my rig in the next 12-24 months I will be going back to air cooling, just like the person that sold me the BeQuiet ;)
 
Soldato
Joined
1 Jul 2011
Posts
8,632
I think you mean Thermalright HR-02, not Thermaltake. :p

If temps went down 10% it is most likely because your case airflow was not suppling HR-02 with air at or less than 5c warmer than room. This is a common problem when builder does not know that most cases don't come with enough fans if they are good or their stock case fans are almost worthless.

Silent Loop pump making noise is most likely because yo have air in it. Where is radiator mounted? Radiator needs to be mounted higher than pump waterblock unit. If it is not then at least the hose ends into radiator need to be higher than pump waterblock unit.

The be quiet! Silent Loop is not a CLC, it is an AIO. The differences are copper radiator with threaded fittings, a much better pump that flows almost twice what most CLC pumps flow, and a fill port/plug so user can add coolant and/or service as needed.

If you want to get it working correctly start a thread and I'm sure we can get it running cool and quiet as it should be.
 
Associate
Joined
30 Jun 2006
Posts
1,218
Location
Cardiff
To the OP> I have used a mixture of Aio's and air coolers over my past few builds. I also have the components for a proper custom water cooling setup but never ever got round to it for whatever reason. This includes Phanteks and XSPC Raystorm cpu blocks, about 2-3 EK rads, 2 Black ice rads, xspc rads, D5 pump, tubing and distilled water most of which have never been used.

For Air coolers the most recent experience is with the Noctua D15 and D14 although in the past I have used Thermalright. For aio's I have h115i pro, x62 and Ek predator 360.

As for air coolers go the D15 met my needs more than the other air coolers. While the D14 is similar it doesn't play nice with tall memory sticks. I am about to try and cram a D15 into an Inwin A1 when it arrives.

SO while the Noctua looks big and is heavy which it is it securing method is really good and as good as any I have used so far. It's performance pretty much speaks for itself you can google it and I am sure it will turn up many results. Up to a certain RPM it certainly pretty quiet although I got to warn you the fans colour seems to upset people personally it doesn't bother me. For the aio's I use a H115i most of the time although from time to time I switch it out for the EK predator. The x62 is a dead duck unfortunately after I used it with a 5960k. It wasn't that old either.

Between the H115i and x62 I prefer the H115i. And that was after ditching the nzxt fans straight away from out of the box to ML140 pros. But the thing I dislike most about the two is the software you have to use. I don't want to use nzxt or deffo not corsairs. Performance wise they are more or less the same although the pump noise seems to be a smidgen quieter on the Corsair unit. I don't know about the H115i but I do know an x62 cannot handle a 5960x;)

Which brings me to the EK predator... So many members around here are more than familiar with the EK predator or should I say it issues. Having said that mine has been faultless. Although there is more significant pump noise (maybe just because it older) and I am not a fan of the EK vardars. I prefer my ML pros or Noctua fans. So it certainly noisier than my h115i and compared in context to the D15 it way noisier.

But it the only cooler I have (not including the watercooling parts I have never used) which can beat the D15 and not by a tiny bit but by a bit.
 
Associate
Joined
19 Mar 2014
Posts
483
If i was to buy a new cpu cooler i would hands down get the new Corsair platium AIO if you look at reviews from OCUK or https://m.youtube.com/watch?v=05EadRvVceg#menu you will see its a great AIO. Im all for custom water cooling but if you want to cool a cpu and have a nice sexy AIO you cant really go wrong with it. I have the h100i v2 with an 8700k thats delidded, running 1.4v at 5ghz and it won't go over 62 degrees even when its 20 degrees ambient. Some air coolers are great and half the price but look **** and weigh a ton. Its down to personal choice


*** Fully Star Swearing ***

That’s great advice, amazing! Corsair are one of the few to use the latest Gen 6 Asetek pumps! Also their iCue software knocks all the competition out of the park. Don’t let old reviews of their software put you off.
 
Associate
Joined
19 Mar 2014
Posts
483
To the OP> I have used a mixture of Aio's and air coolers over my past few builds. I also have the components for a proper custom water cooling setup but never ever got round to it for whatever reason. This includes Phanteks and XSPC Raystorm cpu blocks, about 2-3 EK rads, 2 Black ice rads, xspc rads, D5 pump, tubing and distilled water most of which have never been used.

For Air coolers the most recent experience is with the Noctua D15 and D14 although in the past I have used Thermalright. For aio's I have h115i pro, x62 and Ek predator 360.

As for air coolers go the D15 met my needs more than the other air coolers. While the D14 is similar it doesn't play nice with tall memory sticks. I am about to try and cram a D15 into an Inwin A1 when it arrives.

SO while the Noctua looks big and is heavy which it is it securing method is really good and as good as any I have used so far. It's performance pretty much speaks for itself you can google it and I am sure it will turn up many results. Up to a certain RPM it certainly pretty quiet although I got to warn you the fans colour seems to upset people personally it doesn't bother me. For the aio's I use a H115i most of the time although from time to time I switch it out for the EK predator. The x62 is a dead duck unfortunately after I used it with a 5960k. It wasn't that old either.

Between the H115i and x62 I prefer the H115i. And that was after ditching the nzxt fans straight away from out of the box to ML140 pros. But the thing I dislike most about the two is the software you have to use. I don't want to use nzxt or deffo not corsairs. Performance wise they are more or less the same although the pump noise seems to be a smidgen quieter on the Corsair unit. I don't know about the H115i but I do know an x62 cannot handle a 5960x;)

Which brings me to the EK predator... So many members around here are more than familiar with the EK predator or should I say it issues. Having said that mine has been faultless. Although there is more significant pump noise (maybe just because it older) and I am not a fan of the EK vardars. I prefer my ML pros or Noctua fans. So it certainly noisier than my h115i and compared in context to the D15 it way noisier.

But it the only cooler I have (not including the watercooling parts I have never used) which can beat the D15 and not by a tiny bit but by a bit.

Corsair iCue is finally a great product, in a way that Link never was!
 
Associate
Joined
19 Mar 2014
Posts
483
I assume you realize just because I post about what CLCs are and how they compare to air (and sometime custom H20 loops) in no way indicates that they are not in fact truthful posts usually supported by data like here.

I don't make statements like 'Someone presented you with testing data somewhere else which disproves your theory.' I post data or links to data that supports what I say.

Here's first one I find googling 'AIO vs CLC, what's the difference'
https://www.overclock.net/forum/10-amd-cpus/1623479-clc-aio-cooling.html

Or how about one talking about the differences between CLC and custom loop saying:
"Saying that watercooling loops and closed loop liquid coolers are both watercooling would be similar to saying a high-end Mercedes SLK and an entry level Kia Sonata are both cars."
http://www.tomshardware.co.uk/forum/281201-29-liquid-cooling

Here is link to the difference between Asetek and CoolIT OEM CLCs sold under many different names like 'Corsair' who does not make any CLCS at all, but only puts there labeling on other OEM CLCs.
https://www.reddit.com/r/buildapc/comments/4b6xtk/discussion_a_word_of_warning_about_corsair_clc/

Just for info, most air coolers are made by the company who's name is on them. Same applies to custom loop components.

Also for info, Asetek has the US patent on the pump on waterblock in sealed loop concept, meaning in USA or any country honoring US patents the only CLCs with pump on waterblock you will see are either made by Asetek ro paying royalties to Asetek.

For reviews trust in Tech Jesus, on GamerNexus YouTube channel. Corsair use Asetek pumps and most Corsair AIOs now use the latest Gen 6 pump from Asetek. Corsair H1XX Platinum AIOs also use ML fans, probably second only to Noctua IMHO.
 
Soldato
Joined
1 Jul 2011
Posts
8,632
What is the flow rate and head pressure rate of Asetek pumps?

Several years ago I found the specs for:
Thermaltake water 2.0 performer Rebranded Asetek 545LC 12v 3.1w
Open flow rate = 61 L/h, flow rate through 240mm XSPC RS249 radiator = 56.7 L/h​
Link was:
User Cakewalk_S on OCN tested H50 pump 12v 2.6w
Open flow rate 40 L/h, flow rate through 240 radiator 30 L/h
For some reason none of the CLCs have any publish pump specifications like we see on custom loop pums giving use their flowrate and lift. CLCs give us fans cfm and static pressure .. but nothing for the pump. Enermax is the only exception wiht 450 L/h for the pump in their Liqtech TR4, but that CLC has lots of problems with some owners on 3rd warranty replacement so not one worth even looking at until they get it sorted.
 
Soldato
Joined
1 Jul 2011
Posts
8,632
I wonder how developed the human race would be if our ancestors thing like you do about CLCs.

But your logic why should we develop better means of transportation when we can walk?
 
Soldato
Joined
18 Oct 2002
Posts
4,093
Location
Devon
the corsair platinum AIO's are not asetek. wooo flow rate argument again. It is irrelevant given they still cool the CPU and do a good job of it.
I think that's a bit silly. I recently wanted to upgrade my monitor, did I look at the first one I saw and think it will display a picture and do a good job of it? No I spent some time researching what panel size I wanted, what pixel density I wanted, what panel type best suited my needs, what was available that met my requirements and was in budget and then read reviews. I looked at AIO / CLC cooling the same way, quality of radiator (copper vs aluminium), totally closed loop or refillable (CLC vs AIO), ability to extend the loop, pump rating, included fan spec, software, noise. These technical specs may be irrelevant to you but not to people like myself want and need to make sure they are making the right purchasing decision.
 
Soldato
Joined
18 Oct 2002
Posts
4,093
Location
Devon
I think you mean Thermalright HR-02, not Thermaltake. :p
- Correct
If temps went down 10% it is most likely because your case airflow was not suppling HR-02 with air at or less than 5c warmer than room.
- Nope I have read and followed your guide and bought a digital thermometer to identify the best options for cooling my case which has been stripped out of all non essentials to improve airflow. I saw ~2c drop in idle temps and ~8c drop in load temps some of which may be down to fresh paste but my gut feeling is the SL140 is better at cooling than the Macho.
If you want to get it working correctly start a thread and I'm sure we can get it running cool and quiet as it should be.
- Done and thanks for your reply.
 
Back
Top Bottom