Go Back   Hardware Canucks > CASES & COOLING > Water Cooling

    
Reply
 
LinkBack Thread Tools Display Modes
  #11 (permalink)  
Old August 2, 2007, 06:32 PM
Cardboard box with DSL's Avatar
Top Prospect
 
Join Date: May 2007
Location: Victoria, BC
Posts: 170
Default

Great review!
I'm thinking of getting into watercooling myself sometime soon and your reviews are providing really good insight into what parts I should get.
__________________

Reply With Quote
  #12 (permalink)  
Old August 2, 2007, 06:37 PM
3.0charlie's Avatar
3.0 "I kill SR2's" Charlie
F@H
 
Join Date: May 2007
Location: Laval, QC
Posts: 9,627

My System Specs

Default

I've been using a Koolance rig for 3 years now - actually, this kit got me into WC'ing. It's based on the Exos system, with all 3 blocks. Mind you it's the "old" CPU-300-H06 block, for both S478 and LGA 775 (with adaptor).

That block kept my 3.0C (Northwood) running at 4.0GHz cool to 46C loaded (back then with Prime95). Now it's cooling a 3.2E (Prescott) running at 4.0GHz too (less volts) - load temp are 47C using Prime95 too.

I've heard both sides of the story, being how crappy Koolance can be and how easy and leak-free their systems are. They can perform too, since with the Exos's triple 80mm fan-radiator combo it is cooling the CPU, the NB and the VC (X800XT PE) rather easily.

I'm really surprised at the results. Though with more experience I now consider Koolance an "entry level" cooling system (which I believe it still is), and they are really trying to get away from that image. Could it be a flow issue?

As for contact surface, why do people make such a fuss about how shiny it is. For crying out loud, we are all adding TIM as a faying surface seal: to close the gap between both surfaces.

From an engineering stand-point, it makes no difference.

John
Reply With Quote
  #13 (permalink)  
Old August 2, 2007, 07:27 PM
Misoprostol
Guest
 
Posts: n/a
Default

Quote:
Originally Posted by 3.0charlie View Post
As for contact surface, why do people make such a fuss about how shiny it is. For crying out loud, we are all adding TIM as a faying surface seal: to close the gap between both surfaces.

From an engineering stand-point, it makes no difference.

John
I had to find something positive to say about the block....
Reply With Quote
  #14 (permalink)  
Old August 2, 2007, 07:42 PM
3.0charlie's Avatar
3.0 "I kill SR2's" Charlie
F@H
 
Join Date: May 2007
Location: Laval, QC
Posts: 9,627

My System Specs

Talking

Quote:
Originally Posted by Misoprostol View Post
I had to find something positive to say about the block....
Reply With Quote
  #15 (permalink)  
Old August 3, 2007, 12:05 PM
Misoprostol
Guest
 
Posts: n/a
Default

I seem to have come under fire elsewhere on the interweb, and I'd just like to set the record straight on a few of the concerns being raised over my review. I'm going to methodically go through and address every single concern with my review. We're accessible here. If you wanted to see something done differently, all you needed to do was ask.

Quote:
funny review contradicts its self
You're right. My wording was bad. It's been changed :)

Quote:
Well, it'd be foolish to dismiss a component with 1 review, especially when seeing the guy's own incompetence at mounting the block and then blaming the manual. From those picts you can put the bolt perfectly fine and I have the block. Ironically, this other review of Koolance 330 shows the complete opposite. So someone did it wrong.
You're right, you CAN mount this block using the stock mounting hardware. The problem with the stock mounting hardware IF USED IN THE RECOMMENDED CONFIGURATION IN THE MANUAL is that it's very very difficult to apply the thumb screws because the springs (as shown in the picture) are well above the end of the thru-bolt. THAT is why I criticized the mounting system. Honestly that makes no difference to my test results because I did not use the Koolance mounting hardware.

As far as my criticism of the manual goes, the pictures are unclear. They are as unclear as I said. I'm not writing this review for people who have mounted 100's of water blocks. Obviously some people (including me) will be able to mount it without instructions. That doesn't mean that the included instructions should have terrible pictures, and instruct the end-user to do things that are hard. It's quite difficult to tighten two thumb screws at opposite corners like that with the springs so tight. I could have done it, but I believe a hard-mount is a better/easier mounting system in this case, so that's what I've recommended to our readers.

Quote:
And maybe that someone was Koolance themselves.....they let a block out of their production facilities that had a rather noticeable raised dimple in the center of the block. Don't think that photo of the distortion of the reflection was anything but damning on Koolance's quality control. How could one expect to have any decent heat transfer with the center punched out like that? But I don't blame him for testing what he got.....any newbie consumer that bought that Koolance setup wouldn't know any different and would have used it anyway.....so the test is appropriate.
My block is a retail sample that was bought, just like all of my blocks, except for my APOGEE GT. I had a FuZion that was supplied by D-TEK, but that one resides in my own system, and I have replaced it with a retail FuZion. These results are as real as they get, folks.

As for the dimple though. Actualy I figured it would probably HELP rather than hinder performance. If I'd taken the block off and looked at the ceramique spread on the bottom of the block only to discover that that was the only part of it making contact with the IHS, then I would have discarded it as defective. It didn't seem to be all that raised, so I left it alone. Remember that my photo EXAGGERATES the dimple. Those dots you can see are 10ft above the block (3m for my Canadian friends...).

Quote:
I don't know about that test, all others I see a GTX outperforms the Fuzion by 1-2 degrees, not the other way around.
Comments like this are best addressed to me directly here on our forum through PM, or through my discussion thread. That's the only way we're ever going to clear the water. I've tested the GTX vs. the FuZion now on 4 different CPUs... Pentium D 805, Xeon 3220, Core 2 Duo E4300, Core 2 Duo E6400(current). Every single time, the FuZion (both FuZions mind you. I got rid of my sample one when I was trying to figure out why many people were getting better results with the GTX) outperformed the APOGEE GTX by a slim margin.

I've asked Gabe from Swiftech flat out "do you think my GTX is defective", and he does not think it is.

If you'd like to see my bow, I can try to post some pictures, but I can only conclude that it is bowed very well because there is (as described by OPPAINTER when I discussed it with him) a patch in the center that looks like it has almost NO thermal compound on it whenever I take the block off.

Quote:
this reviewer really gets stuck into them doesn't he
outch!!!

those results are way worse than other ones posted here

could it be that he just didn't mount it all properly??????
The results are probably worse because my test is DESIGNED to exaggerate a difference between blocks. I'm running 1.5V through this CPU and there's NO WAY it needs that much voltage for this overclock. The point is to separate the cream from the junk. The more heat I put under a given block (by my reckoning), the more pronounced a 20% difference in efficiency is going to be. I can't just run the CPU at stock because (and I tried this) I'd end up with a 2-3C difference between the Magicool CPU block and a FuZion under load. Core 2 Duo doesn't start outputting any appreciable amount of heat until you pump voltage through it.

The block was mounted properly. You can see photos of my mounting method in my review. The Northbridge is VERY CLOSE to the hold down plate, but not interfering with it. I also mounted the block twice with exactly the same results.

Quote:
Might I also point out that this reviewer is using a rad that IMO, is crap, and he is using "weaker" fans (should probably be SM's, or better yet, SH's) for said rad.
You want to see results with a Thermochill PA, then say the word. This kind of thing is much better addressed to me directly. If you really want the truth, then a good way to seek it out is to ask for it. If all you want to do is nit-pick every little thing, then you're never going to believe anything are you? I will redo my results with a Thermochill rad on ALL BLOCKS for my upcoming review of a currently undisclosed block.

Quote:
While I didn't read the whole review, I can't really say I can 100% trust the results presented. I can't put my finger on it ATM but, there are some things about this review that just don't sit right with me. I'm not a Koolance lover or supporter by any stretch (even though I do own/use a few of their things, no blocks though). Many of the things the guy said, I wholeheartedly believe in, I just can't completely swallow his review though. I would really like to see more results using a better rad...and maybe better tubing as well.
I'm sorry you have trouble believe my review, but my position in this industry is such that I cannot afford to have any sort of bias. I would have loved to see this block perform, and perform well. I love water cooling, and I'm no fanboy. If my Swiftech APOGEE GTX outperformed either of the FuZions I have, it would be in my system, and I would say it's the best. If my Magicool CPU block kicked everything else's butt, I would use that, and I'd sing its praises from the rooftops.

Honestly though... Tubing? LOL @ that. Tubing does not affect performance. I'm not going to waste expensive Tygon on a test bed because something thinks there is a chance that there's a performance difference between 1/2" ID Clearflex and 1/2" ID Tygon...

Quote:
OK, a couple more things about this reviewer.

1. The "Product Wizard" he said was up to date, well apparently, it's not complete. There are a whole mess of "older" boards missing.

2. If he had such a big problem reading the paper version of the install instructions, maybe he should have read the ones on the web site.

I know I sound like I'm defending Koolance but I'm not, honest. I think someone who truly knows what they're doing should do this same comparison (with the better parts of course), then we all can be certain of the results....I nominate Radical 53.

Methylphenidate, I love my Apogee Extremes (3x). I also have a Fuzion thats going to go on a E6300 in the not too distant future.
1. You're right. I did not check older boards, but this would have been a GREAT thing to put in our discussion thread, or PM me about. I will update my statement about their Product Wizard as soon as I'm done with this post. The reality of it is that I only checked a couple current boards. I did not think to check for old boards to find out if something was "up to date" because usually that's not what's missing...

2. I have had a look at the instructions on their web site (which I don't believe that a novice user should have to look up) and they are exactly the same as the ones that are included with the block. The colour photos help me see what they are doing, but this still doesn't change the problem I have with their mounting system.

There is no problem here with me not knowing what I'm doing. Mounting a water block is not rocket science. The fact that I'm able to generate such consistent results between mounts indicates to me that I've developed a very consistent and reliable way to mount water blocks. Once again though, this would have been better directed to me, rather than posted on a forum elsewhere on the web where I'm unlikely to see it.

Since I started using the same mounting system for everything, I've yet to have a difference of more than 1C under load between the same block over two different mounts. That includes slight variations in room temp. It's not hard.

Quote:
It doesn't matter how good the rad is as long as it's decent. The basis of these tests can be validated by the fact that all blocks were done with the exact same setup/conditions.
Honestly I don't know exactly how this rad stacks up. I have a radiator round up coming in the next couple weeks that will compare the PA120.2, the Swiftech MCR220-QP, the Blacki Ice Pro, Black Ice GTS 240, and the Magicool dual rad. I think the results will be very interesting.

One thing I do know for sure is that the radiator is not the bottleneck in this situation. My rad is cool to the touch even when the CPU has been under load for 15 minutes. My water is not warm. If it was, I would have switched the rad for something else.


Quote:
According to the testing done by Radical 53, that rad is middle of the pack at best when SL's are spinning ~1200 RPMs. So, is it decent? IMO, no, it isn't. I also don't think many other ppl here would either as most of the rigs in the gallery will attest to. If you think it's a decent rad, find 5 rigs in the gallery that use it and I'll accept the results.
It doesn't have to be even a decent dual 120mm rad to cool an overclocked Core 2 Duo. Like I said, if my water was hot, I would have changed it. It shouldn't make any difference whatsoever though because all blocks were run with the same radiator. If it's an advantage, they all have it, if it's a handicap, then they all have it.

Quote:
Also, the radiator wont play that big of a role as long as its not stressed. If the radiator can bring the coolant levels back to near ambient temps b4 it gets recycled, what would be the performance benfits of having a larger radiator? IF you guys want to get nick picky about components, then the reviewer should of droped water temp probes in and out of his radiator to messure how much heat the water picked up after going though a full loop, and back out.
I could have dropped water probes in the water, but it's really not necessary. I've followed the scientific method. That's all we should really be asking for, is it not? I could generate tons of graphs and numbers if you wanted, but at the end of the day what do we care about? Our CPU temps.

Quote:
Friedchicken, It's not my only reason for dismissing this review, just the only one that has physical substance to it. The other is the way the article was written. It just sounds like he went into this with the attitude of "I'm going to show it sucks, even if I have to force it". It's a gut feeling, something I've slowly (and painfully) learned to trust over the years.
Now that's just offensive, and should have definitely been sent to me privately if you have a problem like that. I have no vested interest in Koolance, Swiftech, Danger Den, D-TEK, or any other water cooling manufacturer. If you think it sounds written as if I already knew what the results were, then you're right. I already did. I tested the block before I wrote the review. I tried to write as if I didn't already know, but you caught me. Honestly I had a new block in my hands and like any of you, I wanted to put it on a CPU as soon as I could to see how it would perform.

All of that said, your reason for distrusting my review that has "physical substance" still does not. There is no reason that a radiator that is already adequate for the CPU we're cooling would have that kind of effect on the way block stack up against each other.

Quote:
Lol. I have no argument with you SG, in general you've shown yourself to be very positive towards just about everyone on the forums. So I'll take this with a grain of salt

Other than my reply to Ranker I took a few minutes to read the review itself and I have to say, despite MY comments that I'VE been unhappy with the blocks performance (based on the claims of other forum members with other blocks) I'm less than impressed with this reviewer's, well, review.

For one thing unless the guy is a total idiot he should have NO problems installing that block even WITHOUT instructions. That he can actually say he could not install it in the recommended manner is absurd. I have one of these blocks and trust me, it's one of the easiest blocks I've ever seen to install.

He also shows using the koolance thermal paste for the koolance block (in a picture) then says below he's using Arctic Ceramique for "all" the blocks, but never shows it's application anywhere.

He then mentions how his numbers vary widely from "other" reviews of the block on the web, then just out of hand dismisses those reviews as flawed. I mean, they have to be right? Obviously his review can't be wrong.

In any case I've already mentioned that I've also seen unsatisfactory results but have yet purchased another block to compare to. I think I'll order a Fuzion soon and see what comparative results I get. I'm pretty confident that the block is NOT performing as well as would be expected from a block of it's pricepoint.

As for Koolance's other products, well, same rules apply. Until they are tested and compared I for one will make no blanket hype statements that they suck out of hand. The new GPU blocks are excellent.
I've already addressed my issues with mounting the block. Please see the pictures in my review or (as posted earlier in this thread) the Virtual Hideout review for any verification that the thru-bolts are not long enough for the block to be easily mounted. Of course I can mount it without instructions, but let's be realistic... Are most of the people buying Koolance products water cooling veterans? NO. My review is written to a target audience. You are not it.

Arctic Cooling Ceramique was used for all blocks. I mentioned that in my review. I don't think I need to show a picture of Ceramique for everyone here to know what it is and how it works.

The other review is flawed. How do I know? Well for one, the other review is clearly biased (read it). For two, I know for a FACT that the Koolance CPU-330 does NOT perform that well. How do I know this? I have one in front of me. I also have an APOGEE GT in front of me. They are in close proximity to the FuZion I have in front of me. Those results are BS.

All of that said, I agree with your last statement. I did not make any blanket condemnation of Koolance products and neither should you. This is ONE block. That's all.

Quote:
But until more DIY biased review sites are willing to conduct ethical, UNBIASED reviews of Koolance products I think it's tough to make blanket statements about their quality or performance.
HWC is as unbiased and as DIY as any review site out there. As far as whether or not my review is ethical, I fail to see what could be unethical... I guess I could add "no small furry bunnies were harmed in the making of this Koolance review" at the end.

When it's all said and done, you don't like my results, you're not going to believe them whether you have a real reason for not liking them, or if you simply don't trust me. I can tell you right now though, that there is a LOT of politics in reviewing water blocks. That's why I try to obtain my samples from retail stores instead of directly from manufacturers. That way they have no say in your review.

I'm going to stop monitoring that thread about me, but if anyone else has any feedback for me, then let's work together and turn my reviews into something that you CAN believe. I'm not here to waste my time turning out reviews that no one believes are credible. Let's find the truth.
Reply With Quote
  #16 (permalink)  
Old August 3, 2007, 01:07 PM
3.0charlie's Avatar
3.0 "I kill SR2's" Charlie
F@H
 
Join Date: May 2007
Location: Laval, QC
Posts: 9,627

My System Specs

Default

Misoprostol and I had a discussion via PM regarding all this. Here it is, as he proposed it. I see no reason why everyone should not see this information

John

Quote:
I have read both your review and discussion thread regarding subject matter.

About the mounting studs. I don't have the actual block at hand, but I do know about that mounting system. If you look at the pic, you will see that the thumb screw has a stud on its lower section. This would allow to reach the threads while adding pressure on the spring and start torquing 2 thumb screws at the same time.





It's not easy, but can be done. The manual on sheet 4 shows as such. You may already know this, and if so I apologise accordingly.

John
His answer:

Quote:
No apologies necessary. I know it can be done. It's just hard for no reason. If they'd just made longer thru-bolts, then this whole problem could have been avoided. I don't think that a novice user would like this mounting system because you need to do two at the same time.

I think they were trying to avoid having the bolt stick out of the top of the thumb nut for aesthetics, but that still could have been done without making it so difficult to tighten them down.

I'm accessible and open, so feel free to post this in our discussion forum, as well as my response. I'm all about open communication, and not about behind-others'-backs communication.

I will edit the review to reflect that it can be done, but I will maintain that I *REALLY* don't like the mounting system.
By torquing 2 screws at a time, it avoids putting un-equal pressure on the core, which could crush it. Swiftech has a similar design, using Dome nuts instead of thumb nuts. You must put pressure on the springs in order to engage the threads, and that on opposite studs simultaneously.

John
Reply With Quote
  #17 (permalink)  
Old August 3, 2007, 01:44 PM
Misoprostol
Guest
 
Posts: n/a
Default

While it's true that Swiftech's design is similar, I guess what bothered me most about the Koolance one vs. the Swiftech one is that the Koolance springs are A) harder to compress than the swiftech ones and B) stick out further over the top of the bolt.

To avoid this problem entirely, the system D-TEK uses (nice long bolts) makes a lot of sense. You just stack everything you need, thread all the nuts on, then tighten criss-cross.
Reply With Quote
  #18 (permalink)  
Old August 3, 2007, 01:55 PM
sswilson's Avatar
Moderator
F@H
 
Join Date: Dec 2006
Location: Moncton NB
Posts: 14,481

My System Specs

Default

Heh.... sounds like you're making friends and influencing people Miso... :)

Talk about trying to nitpick..... who cares how well "XXX" hardware or TIM performs as long as the tests were all performed on the same platform using the same methodology.

The only serious question that might be asked is if there is possibly a performance benefit to using the stock Kooldance mounting method no matter how hard it might be........

I guess some of this was to be expected considering you "called out" several reviews for their testing methodology.
__________________
MSI Z87I Gaming AC / i5 4670K / 2X 4G Gskill 1866 DDR3 / XFX XTR 750 / EVGA GTX 680 SC+ 2GB / Intel DC S3700 200G / random 160G Sata HDD
Inwin 904 / Swiftech MCP655-b / Alphacool NexXxos XT45 120 Rad / 2X Scythe GT AP-15 / EK Supreme HF / Dell UltraSharp U2412M

Asrock AM1H-ITX / AM1 Athlon 5350 / 2X4G Gskill PC3-14900 / Intel 6235 Wi-Fi / 90W Targus Power Brick / 320G Seagate Momentus / Mini-Box M350 / 1X 22" Dell IPS / 1X 22" HP
Reply With Quote
  #19 (permalink)  
Old August 3, 2007, 01:56 PM
3.0charlie's Avatar
3.0 "I kill SR2's" Charlie
F@H
 
Join Date: May 2007
Location: Laval, QC
Posts: 9,627

My System Specs

Default

A major difference between Swiftech's Dome nuts and Koolance's thumb screws: you cannot over-torque the dome nuts. Once you have reach the bottom of it, you cannot torque anymore. Only the spring is left to put pressure on the block.
Reply With Quote
  #20 (permalink)  
Old August 3, 2007, 02:22 PM
Misoprostol
Guest
 
Posts: n/a
Default

personally I really like Swiftech's dome nut mounting system. It's idiot-proof and relatively easy to use.

For the more extreme user I like the D-tek mounting system because it gives you more flexibility.

Honestly I would be absolutely FLOORED if the Koolance block performed better using stock mounting hardware. I'd be happy to try it, but I'm pretty sure it's going to be a fruitless exercise. If it's close to my own mounting method, then I'll say so. If it's completely off, then I'll be happy to update my performance charts and make a public apology to Koolance.

And yeah, I'm not making any friends with this one...
Reply With Quote
Reply


Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
Thermalright XWB-1 Review Discussion Thread Misoprostol Water Cooling 11 October 25, 2007 07:53 PM
Discussion thread for QNAP TS-109 Pro NAS review... 3oh6 Storage 15 September 7, 2007 11:50 PM
Discussion thread for Hitachi 1TB drive review Misoprostol Storage 6 July 18, 2007 09:55 PM
Discussion thread for QNAP TS-201 review Misoprostol Storage 4 July 18, 2007 04:23 AM
Discussion Thread for Ballistix vs. Firestix review Babrbarossa Reviews & Articles from the Web 10 June 12, 2007 04:43 AM