My latest package came through on Thursday, this time Corsair had something for me to look over, I would be reviewing th-
Okay, so I’m not the kind of person who gets sent free hardware from manufacturers. However, as followers of PC Gamer’s Twelve Days of Christmas competition will know, I did win a lovely cooler, and technically it was shipped and paid for by Corsair. Close as I’m ever going to get.
The Corsair H50 High-Performance CPU Cooler is a closed-system water cooler. I did once look into water cooling, it offers impressive levels of cooling over traditional heatsink\fan systems and can also be a lot quieter. However it’s still a mostly DIY area of computer modding, people design their own systems from the myriad of water cooling components available, there are dangers of leaks and regular maintenance is required, not to the mention the huge financial cost. I decided it wasn’t the kind of project I should be messing about with. At the time I had only seen one example of a pre-built kit that was anywhere near within my price range, a CPU water cooler like this one but manufactured by a company I hadn’t heard of. After a quick search I found very few reviews, none of which were very convincing. So I let the idea pass.
I had entered a couple of the previous days competitions and didn’t expect to get any further this time, this time though I won. Not only was I happy that I’d won such a great prize, I also thought I could easily get a few informative blog posts out of it, and here we are. By performing several tests both before and after I would not only be able to provide people curious about this product, or water cooling in general, with some useful information, I would also find out how much of a difference it was actually making to my system.
Aside from watching the large yellow numbers of FRAPS in the bottom right hand corner of my screen I have never performed tests on my computer before. I’m sure my testing won’t stand-up against some of the professional hardware review sites, but I just wanted to explain how I have put some thought into the tests.
Firstly I won’t be performing any synthetic tests, they always seem very professional in reviews but I’m only interested in what direct impact it will be having on my gaming experiences. So I selected three games based on what little knowledge I have of CPU\GPU balances in games.
What performance testing would be complete without one of the best looking (from a technical stand-point, discounting art-style) games that brings even the latest computer hardware to it’s knees. From what I have gathered from various performance reports, my graphics card is holding back any performance increases I might otherwise get from a faster CPU, so Crysis will be a sort of control sample for the next two. It will also be good to see if I can improve the performance of this debilitating game, later on in the tests.
From reading various posts on the official Elder Scrolls forums my 8800GT should be able to run at a good FPS on the highest settings, yet I’m still only getting low-ish FPS plus some occasional stutter. The system requirements list a 2Ghz CPU as the minimum required for playing, and the recommend says 3Ghz. Now I’ve known since AMDs rise to fame in the Athlon 64 days that clock rate isn’t as important to performance as the other inner workings of a CPU, but I don’t have any other form of measurement. Oblivion was developed in the pre-dual core days, and whilst you do get extra performance if background tasks have another core to run on, the game will always be limited to the limits of the individual cores. In my case that’s 2.4Ghz, so my reasoning is that this could be the main source of the poor performance in Oblivion. Not the soundest of reasoning, but it will make for an interesting test.
Not only is this the game that I am playing the most at the moment (…), Wurm Online also manages to kick my CPU fan into overdrive every time I play. Although it may not look like it Wurm Online seems to tax my computer more than most games, whether this is down to poor optimisation or just because of how complex the items system is I don’t know. But I’m almost certain my graphics card is not breaking much of a sweat.
Using Core Temps I took readings from each game before I installed the new cooler. In Crysis and Oblivion I just played around for twenty minutes or so and in Wurm Online I left it running for about half-an-hour whilst doing some of my usual jobs.
With the pre-installation tests out of the way it was time to fit the cooler. I was originally thinking of taking some step by step photographs like I did with the Akasa Vortexx Neo, but installation proved to be a little more long winded than I thought. So I’ve just taken a before and after picture.
After I removed my Arctic Cooling Freezer 7 Pro, I set about reading the instructions that came with my new cooler. They were nice and simple, if lacking a bit in certain details. The cooler comes with two sets of brackets and screws one for Intel, the other for AMD CPUs. Unfortunately this was when I found out that the brackets came in two parts, and one of them needed to go on the back of the motherboard. This of course meant removing my graphics card and TV tuner, then unscrewing my motherboard from the case. Not a particularly hard task, I was the one that fitted it after all, but I would certainly have preferred not to have needed to do it. At first I had a bit of trouble working out how to line up the rear frame, this was the first (of only two) oversights in the instructions, admittedly they are listed as the quick start instructions but it is such a small thing. There are four metal plugs which go into the frame to give the screws from the other side something to grip on. The correct holes are marked clearly enough on the frame so it wasn’t too much of a problem once I realised.
With the rear frame now in place it was time to start thinking about radiator placement. This was actually the second step in the instructions but I preferred to sort out the motherboard first before finding a home for it. As you can see from the picture above, the little round section with the Corsair logo on it sits on top of the processor, the recommended radiator position is where most rear case fans are. Strangely it suggests the optimum flow should be in from the rear fan, with my case (and most cases I think) the idea is for air to flow from the front then out the back. The suggested position could possibly cause problems with my graphics card, since air would be coming from the front fan and the back.
My first thought was to put it on one of the fan spaces on the top of the case, this would then blow the air down towards the graphics card, which would then push it out of the back. Unfortunately the radiator and fan took up too much space. Limited by the length of the hoses the only position that would work was on the rear exhaust. Upon being told I had won this cooler I had a closer look into what it was I had won, on the main Corsair product page it links to an article discussing dual fans, one on either side of the radiator. It was an interesting read but was not something I would have considered doing, however if I had to fit it to the rear exhaust I would get an extra fan. I thought, performance might be optimum coming into the case, but with two fans it would surely work well enough blowing out. So that’s what I did. Unfortunately this meant sharing the screws provided, they are much longer than traditional fan screws since they pass through the case, then through the fan before securing themselves on the radiator. Two seemed stable enough though, and I used two traditional screws to further secure the first fan to the case.
Next I needed to fit the front bracket to the motherboard, this is where the second problem cropped up. It was partly my fault, I should have paid closer attention to the parts I had available. After fitting the correct plugs to the bracket it was time to insert the screws, the first two went in very smoothly but the third was a much tighter fit. What I didn’t realise at the time was that there were two sets of almost identical screws, the only difference been the small thread at the end, one was larger than the other and I was trying to put one of the larger ones in a hole made for the smaller ones. After taking things to far and having to undo most of my work I did eventually discover my error and fitted the bracket.
The final step was to attach the pump\heatsink to the processor, thermal paste came pre-applied which was very helpful so all I had to do was press it against the processor, give it a small twist, then tighten the bracket around it. This was quite fiddly but just took a bit of working, I had tightened the screws on the motherboard a bit too much, so after loosening them I managed to get the bracket to grip.
Not the perfect installation, but good enough. On a second glance at the picture it does look like the hoses are a little twisted but they aren’t, it’s just the funny angle I took the picture at. As you can see on this picture from the official article I mentioned earlier, they do stick out and away from the motherboard a little.
With the cooler fitted I was eager to see how much of a difference it was going to make. Regrettably one of the other fans in my computer has developed an annoying whirr during boot up so the sound difference wasn’t apparent at first. Sticking my ear close to the case I could hear little bubbles travelling around the cooler for it’s first activation since been packaged (assuming they test them when they are manufactured). As the annoying whirr cut out I listened carefully and heard…nothing really. One of the fans is attached to the standard motherboard fan connections and the other is connected to the CPU fan connection. I had a quite peek at the BIOS screen and saw that with smart fan control active, the main fan was travelling at a stately something-hundred RPM.
I had almost gotten used to the build up of the CPU fan during heavy use, but now I heard nothing but the standard hum that constantly comes from my case. I don’t have a sound meter, and I’ve never quite understood how the decibels translate into a real life situation (beyond the simple, more is louder), so all I have are my impressions. The difference is definitely noticeable.
But enough rambling, it’s time for cold hard facts. I didn’t practice the most accurate testing methodology (I think the slow build up on the Crysis graph is from when I was messing around on the settings screen), but the difference speaks for itself.
I think it’s safe to say I was blown away. I forgot to take idle readings but I’m fairly sure they used to hover around the 40’s-50’s, now I’m getting low 30’s occasionally dipping lower. When it comes to handling the increased heat that comes with gaming the cooler also takes it in it’s stride, never even reaching 50°c.
The difference is so startling that I can only think I didn’t place my original fan correctly. I had originally placed it the wrong way (not that you’d know since the computer seemed to run fine), maybe when I turned it around I didn’t quite put it on level, who knows? Yet surely if it wasn’t placed correctly I would have had overheating issues? Whatever the answer, I don’t think it could have come much closer to the Corsair H50, which seems to be running brilliantly.
However I’m guessing you, the reader, have a question, and that’s how much does it cost? I got it for free, knowing what I know now would I pay money for it? To know that we first need to know the price, and a quick search using Google’s Shopping search shows prices around £60. The cheapest price from a retailer I know of (and so would be comfortable ordering from) is £55.98 delivered. Which is a pretty amazing price I feel.
It is out of my comfort zone though, I paid around £15 for my original cooler and it’s served me well over almost 2 years. That leaves a difference of about £40, and I’m not one for spending lots of money on hardware (aside from my original payment), I probably won’t be upgrading my graphics card for at least another year. But I think people who spend more often on hardware would be quite happy with this price, especially if their processor is one of the fastest available for their motherboard and so would have to spend a lot more to buy faster hardware.
So I probably wouldn’t buy this product…as things stand at the moment. Cool components are all well and good, they have decreased wear-and-tear, but they don’t have any other benefits on their own. However more cooling opens up another avenue that could prove more exciting…
Part Two coming soon…