Customcooled pc watercooling line up

[SlaterSpeed]

Have you considered the possibility that kits the swiftech setup should beat (like the Zalman resorator) were attualy gaining an advantage from something? ie the problerm was not with the swiftech test but with the others?

You relly NEED to test each kit at least 3 times to draw a valid conclusion (thats middle school sceince )

On the resorator issue you say -

We then leave the CPU loaded until the temp peaks (maxes out) and record this temp.

How do you know its peaked? how long do you leave it before you diside its at its highest? The resorator would be very sensitve to this point. It could take a very long time to get to its equlibrium tempreture due to the large res capacity. If you diside its maxed out before it is then its going to scew results in its favor.

also i assume you controled airflow over the resorator somehow? its going to make a differance having it neer a fan or for that matter in direct sunlight?

[Tempus]

Lizard,

Thanks for your intelligent responses. Sounds like your testing is fairly comprehensive.

I'd like to make a special note that you might want to consider multiple mounting attempts, as I have seen significant variations due to the mounting of the waterblock.

All in all a great response. I'd actually be interested in reading more of your reviews. Too bad you are all the way in the UK.

[jaydee]

I see the mag cover in the pdf says Agust 2004. So why did you guys use an older version of the Swiftech kit instead of the new one? Pretty damn hard to compare kits if one of the kits contains older gear than what is sold today (August 2004). Should have either left it out or waited untill you got a new kit?

[lizard]

How do you know its peaked? how long do you leave it before you diside its at its highest? The resorator would be very sensitve to this point. It could take a very long time to get to its equlibrium tempreture due to the large res capacity. If you diside its maxed out before it is then its going to scew results in its favor.

also i assume you controled airflow over the resorator somehow? its going to make a differance having it neer a fan or for that matter in direct sunlight?

Yes indeed, the ambient temp and airflow within the room will obviously make for a difference to passively cooled kits than actively cooled radiators.

Our test lab does have air conditioning but it would be pretty silly for us to add extra fans to the Reserator when the whole point is that its a passive system designed for low noise. Each time we tested the Reserator it was left overnight (CPU fully loaded) to make sure the temp really had maxed out.

I'd like to make a special note that you might want to consider multiple mounting attempts, as I have seen significant variations due to the mounting of the waterblock.

All in all a great response. I'd actually be interested in reading more of your reviews. Too bad you are all the way in the UK.

Thanks for the comments. Well, you could get an international subscription, we do already have small numbers of subscribers in Canada and the US already.

Originally Posted by jaydee116

I see the mag cover in the pdf says Agust 2004. So why did you guys use an older version of the Swiftech kit instead of the new one? Pretty damn hard to compare kits if one of the kits contains older gear than what is sold today (August 2004). Should have either left it out or waited untill you got a new kit?

Because we wanted to review what was available in the market at the time, the kit was sourced from a retailer and was the most up to date Swiftech kit they had at the time. Remember this review was carried out over 4 weeks ago now.

[jaydee]

Originally Posted by lizard

Because we wanted to review what was available in the market at the time, the kit was sourced from a retailer and was the most up to date Swiftech kit they had at the time. Remember this review was carried out over 4 weeks ago now.

I can understand that. I would expect however an update with the new Swiftech kit as the MCW6000 is a MUCH better block than the one you tested. Just isn't fair to put an old kit into the mix and turn off people to Swiftech when thay already have a much better setup available now and may not even sell the kit you tested anymore!!! Just isn't right IMO.

[pauldenton]

sadly availability this side of the pond tends to lag behind the US... (esp with swiftech kit..... can't even get the 6000 here except in it's xeon setup....)

[jaydee]

Originally Posted by pauldenton

sadly availability this side of the pond tends to lag behind the US... (esp with swiftech kit..... can't even get the 6000 here except in it's xeon setup....)

I don't see how that justifies using old kits in a modern review. :shrug:

[pauldenton]

Originally Posted by jaydee116

I don't see how that justifies using old kits in a modern review. :shrug:

indeed - but when it was done it might well have been the best swiftech on offer in the UK (it's 6000 based replacement is about £24 cheaper...but iirc there were no 6000s in the UK when they would have gone to press.....)

[jaydee]

Originally Posted by pauldenton

indeed - but when it was done it might well have been the best swiftech on offer in the UK (it's 6000 based replacement is about £24 cheaper...but iirc there were no 6000s in the UK when they would have gone to press.....)

The MCW6000 is more than a month old by a LONG shot (over 3 months old in fact). There for this is pretty irrelevant. Using old stock because no one has new stock dosn't justify using the old period. In fact it just shows how weak the WC market really is... Should have been excluded and I see no reason why it wasn't.

Also if your writting a freaking magazine it is their responsibility to make sure their content is UP TO DATE and accurate. What would the people at Car and Driver have to deal with if they used a 1999 Ford Mustang against a 2004 Corvett because they couldn't get a hold of a 2004 Mustang? They would get their ass sued off by Ford for starters but more importantly they would loose credibility and a large reader base. No one that is interested in a 2004 Mustang or Corvett wants to hear the comparison from 1999 Mustang to 2004 Corvett!

It would be amazingly biased not to get a modern Swiftech kit and update that article. That is if the people there are as legit and unbiased as they claim.

[Etacovda]

I still fail to see how the swiftech kit was performing that badly... something was wrong. I wonder what billA's take on this is

[pauldenton]

Originally Posted by jaydee116

The MCW6000 is more than a month old by a LONG shot (over 3 months old in fact). There for this is pretty irrelevant. Using old stock because no one has new stock dosn't justify using the old period. In fact it just shows how weak the WC market really is... Should have been excluded and I see no reason why it wasn't.

jaydee116 i'm not defending it, i'm really complaining about the availability (or not) of stuff here but i do think you're not taking account of the fact that as a printed mag it's totally UK centered...

personally i've always found printed media to be hopelessly out of date in the PC field - hence why i haven't bought any ever since i got adsl.....

Originally Posted by jaydee116

Also if your writting a freaking magazine it is their responsibility to make sure their content is UP TO DATE and accurate. What would the people at Car and Driver have to deal with if they used a 1999 Ford Mustang against a 2004 Corvett because they couldn't get a hold of a 2004 Mustang? They would get their ass sued off by Ford for starters but more importantly they would loose credibility and a large reader base. No one that is interested in a 2004 Mustang or Corvett wants to hear the comparison from 1999 Mustang to 2004 Corvett!

Cars>watercooling again??

i'm not a driver, but it strikes me that.....
1) there is 1 generation between the kit they should have had and the one they used.... so that would imply a 2003 mustang to 2004 corvette comparison. i don't know about printed car mags, but TV car shows i've seen seem to be full of "comparing a new car to existing similar cars" items....

Originally Posted by jaydee116

It would be amazingly biased not to get a modern Swiftech kit and update that article. That is if the people there are as legit and unbiased as they claim.

well for that to be of use the reason for their weird results would need to be determined.....
pHaestus has 2-2.5C between a 5000 and a 6000 - i doubt their mildly overclocked p4 would have enough extra heat to up that much - so if we say 3C better that would still leave the swiftech way underperforming all the others... unless the reason for it's freak result is determined and corrected...

hmm - don't suppose they could have had the fans turned right down??

[jaydee]

Originally Posted by pauldenton

jaydee116 i'm not defending it, i'm really complaining about the availability (or not) of stuff here but i do think you're not taking account of the fact that as a printed mag it's totally UK centered...

I don;t care were it is printed from to be honest.

personally i've always found printed media to be hopelessly out of date in the PC field - hence why i haven't bought any ever since i got adsl.....

I don't know about that. They all seem to have the newest products in them. Get to read a lot being I spend way to much time in layovers in airports.

Cars>watercooling again??

i'm not a driver, but it strikes me that.....
1) there is 1 generation between the kit they should have had and the one they used.... so that would imply a 2003 mustang to 2004 corvette comparison. i don't know about printed car mags, but TV car shows i've seen seem to be full of "comparing a new car to existing similar cars" items....

Actually no!!! The block used is a couple generations old. It isn't even the final version of the 5000 series blocks for christ sake. If that kind of stuff is still found in the market place as NEW then they need to rethink their product line. BTW the final version, THE ONE PH TESTED 5000A (the one in the review is the older original 5000), has a different top and was a couple C better than the one in the CustomPC review! So now we are talking 4ish C. Now do you understand what I am getting at? This kit is well over 8 months old and couple revisions old plus a whole new block to replace it old. It is just crazy this made it into a current review.

well for that to be of use the reason for their weird results would need to be determined.....
pHaestus has 2-2.5C between a 5000 and a 6000 - i doubt their mildly overclocked p4 would have enough extra heat to up that much - so if we say 3C better that would still leave the swiftech way underperforming all the others... unless the reason for it's freak result is determined and corrected...

hmm - don't suppose they could have had the fans turned right down??

Who knows. My personal opinion will be with held though being I got no factual basis for it.

[msv]

http://www.gtek.se/index.php?mode=it...42b527080c0bc6

Cut this a minute ago. Yep, Sweden sometimes is faaar up in the northern wastelands.
regards
Mikael S.

[pHaestus]

Swiftech had some distribution issues getting the 6000s and 6002s out. I think I got one of the first retail 6002s in fact, and that review is what a week old now?

[BillA]

Originally Posted by Etacovda

I still fail to see how the swiftech kit was performing that badly... something was wrong. I wonder what billA's take on this is

Swiftech hat ON

Swiftech has no objection to reviewers purchasing products where they will, buying from a distributor also ensures that nothing 'special' occurred.
- but would not a 'major' magazine be interested in knowing how 'current' was the product being tested ?
I do understand the lag between a product's introduction, and its availability; and between the gathering of parts, their testing, and the report appearing in a printed publication. Is this what occurred ?

Swiftech welcomes the independence of reviewers, and accepts that negative comments/opinions and comparisons with other products are a part of the review process itself, and help us develop better products.
- but should not such comments relate to the product ?
Is this what happened ?
Yes with respect to the fill and bleed kit; some can follow instructions w/o problems; others cannot. (The instructions, when followed literally, work.) Some hate the kit, some like it; hey, the review process at work. No problem here.

A quick-connect fitting that leaked ? fixed with Teflon tape ? In 99.99% of the cases, a leak in a quick-connect fitting is caused by improper tube insertion. Using Teflon tape does not fix the problem, and may actually worsen it. Either the leak was part of the user installation, and an e-mail or phone call to customer support would have solved the problem, or the leak was part of the pre-installed components, in which case it would be a factory default, justifying calling the manufacturer, which by your own admission is something you do when such events do occur. Why didn’t we receive such call ?

The kit as shown on these pictures is not a 22501 Extreme. It shows the MCW5002 water-block, which was never shipped in a kit. So essentially, this article could be viewed as a misrepresentation of our product.

Notwithstanding the 'kit' components, the performance ranking is different than we would have expected. The shortfall is so great that one suspects that there was a problem. Would it have been so unreasonable to contact Swiftech to see if we had any insights into the kit's performance ? (We sell hundreds a month so we do have probably more experience than the reviewer as to potential problems.)

Or was there a desire to trash a Swiftech product ?

Bill Adams
VP Engn, Swiftech Inc.

Gabe’s comments:

It is fairly obvious that the reviewer had serious problems with this kit. Whatever these problems were, I find it objectionable that a magazine who claims to be writing for “more average end users” (or newbies, as quoted in another one of your responses) , would not include customer support in the write-up. The premise -“we buy a kit as an average consumer, and try to install it as a consumer would”, is good in itself, but when consumers have problems, they do call tech support. Why is it that your magazine did not do so ?

If there was a problem with the water-block retention mechanism for example, it could explain the low performance. And this is something that could have been instantaneously recognized –and fixed, if your reviewer had picked-up the phone or sent us an e-mail.

The reviewed kit, does not appear to be a genuine Swiftech kit, and should not have been presented as such. The reviewed “collection” of parts seems to be including obsolete parts (by close to a year or so), and is also compared to the latest offerings from other manufacturers. The fact that the H20-22600 kit (with the MCW6000 water-block) was not available in the UK, is no excuse. I would submit that it does not make much sense to include a kit in a review claiming to test the latest and greatest, if such kit is no longer made by the manufacturer.

Altogether, this article is certainly causing us undue tort, and I suppose that our next step will be to see what appropriate course of action needs to be taken to have these damages repaired.

Gabriel Rouchon
Chairman, Swiftech Inc.

[jaydee]

Geez, not even a true swiftech kit? Worst than I thought....