Originally Posted by ez12a
I'll see how it performs with its stock fans when i receive mine. I have a good idea on how the h100i performed in my system and if the h220 is comparable or better then we know Linus' review is flawed.
From what I've seen in the comments on his video, they're retesting.
Linus review is flawed, there is no denying that. Its simply not possible to get 20ºC difference, PERIOD. Either way, its his fault for publishing non-consistent and ilogical data.
Originally Posted by sikkly
I would bother if I was them. If a "normal" thing can go wrong for a guy like Linus, than it may go wrong for a lot of end users as well. Solving his problem may help prevent costumers from making the same mistake.
In the end most people aren't Martin. We don't have the tools to test and see the results that he has. I'm sure a lot of people get a high temp due to improper installations and think the cooler is junk and leave it at that.
Swiftech can help those people by figuring out where Linus went wrong. Maybe his review won't help you, but a lot of people will watch his and no one else's review.
The unit tested by Linus was defective or he has no idea how to test things consistently or anything similar. The damage is already done, there is no fixing that, and its that simple. Do you truly thing that If a normal user had the H100i and the H220 and got a 15ºC difference he would accept those results as "normal"? Let alone publish them as some data that accurately reflect the performance of a product? I'm sure the lesson learned by Swiftech here is... do not hand products to people that do not know how to handle them and, thus, they won't screw up your brand image. Right now, Swiftech not only didn't get any good PR, they got bad one. Guess how hard the bad image is hard to erase.... because its VERY, specially when people were waiting results after the CES... and got disappointed when their trusted review wouldn't show the desired results...
And no, the mounting system Swiftech uses is "fool-proof" as is the easiest in the industry to set up with consistent results. I don't think its that hard to screw a few things here and there and not skew results but, apparently, it is hard since some reviewers have screwed their results.
Or... wait a minute... were they REAL reviewers? The problem we have today is that everybody considers himself a reviewer. If somebody offered me a PSU to test I'd decline, since I lack the equipment or the knowledge to do it... and the same for cooling related devices. On the other hand, nobody (most, at least) press-related in this industry cares to do proper reviews using normal testing methods that are standard in any field in which you plan to obtain data to be compared with something... specially when other people's brands are at stake. The guy at OC3D stated that he wouldn't want to spend a ton of time because he had other things to do... would you trust the data of somebody that doesn't even dedicate enough time to do something? I don't, and neither should you.
Originally Posted by Neo Zuko
Should be ok, one 120mm rad area a block plus one more 120mm rad area for overhead. Depends on the overclock really.
I've never understood this sort of logic, really. It depends on what you want to cool and not on the amount of heat you plan to dissipate (to an extent). A double rad could be used to dissipate 2 x high-end gpu cards (500-600W) withouth a hassle, and you would be getting very good temperatures. On the other hand, the same radiator might struggle with 200w cpu's...
Originally Posted by WiSK
But how many of those who watched it have any idea that there was incorrect information in it? How can Swiftech help those people after the fact? They will have ordered their H220 and 2xNoctua's now, just as Linus suggested and likely they will never know that they didn't need to buy them. If I was cynical I would say that for this "accidental" situation is exactly the reason that Noctua sends Linus free fans.
Noctua's.... never understood why they are so frigging expensive, specially when they look TERRIBLE. Give me some nidec GT's and call it a day, please...Edited by prava - 3/5/13 at 9:58am