| Lian-Li Armorsuit PC-P50R Dragon Case |  |
| Reviews - Featured Reviews: Cases | |
| Written by Bruce Normann | |
| Tuesday, 25 May 2010 | |
|
Page 7 of 9
Lian Li Armorsuit PC-P50R Testing & ResultsTesting the cooling performance of a case is not that difficult, but it is time consuming. It takes a long time for component temperatures to reach steady state. They'll get to within a degree or two within 10 minutes, but they continue to climb, almost imperceptibly, for 3-4 times that long. It's not that any one component is getting that much hotter; it's the combined effect of all the components feeding off one another that causes the slow rise to the final maximum temperature. For the load tests, I used a mix of applications that I've used for benchmarking in the past and I'm familiar with how they stress individual components. No one application gives the maximum load for all components, so the load temps reported here are the maximum each piece of hardware reached during the testing. For comparison, I have results from the exact same hardware, installed in a CM Storm Sniper Black Edition case. Both cases were tested with the same hardware and software configuration. Cable management was optimized for each case, based on the internal layout and features available. Note for those looking back in the archives: I used a different set of hardware the first time I tested the SilverStone FT01BW, so the results from that test are not comparable. Test System
Benchmark Applications
Test Support Products
ResultsI've had several cases in house recently with side panel fans installed, or the capability to install them. In every case, the side fan improved the cooling of the video card, so I was a little concerned that the Lian Li PC-P50R would be at a disadvantage. The actual test results show that either I worry too much, or I was over thinking this one. All temperatures are reported as measured, in degrees Celsius, and the ambient temperature for all tests was 26C. (We were having a Spring heat wave on the day I tested.) I did both sets of tests on the same day, swapping the hardware from one case to the other after the first set of tests was done. Let's look at the results:
The idle results for the Lian Li PC-P50R are about 2 degrees centigrade hotter than the CM Storm Sniper BE for the Northbridge and the GPU. This is really good performance on the part of the Lian Li case, given the 100+ CFM of ambient air flowing into the Sniper case from the 200mm side fan blowing at its 100% setting. Notice the excellent performance of the CPU cooler; it's keeping the CPU just 3 degrees over ambient, at idle.
The load results are really excellent, and I had to run the test several times to convince myself that the smaller Lian Li case, with only one 120mm intake fan was really performing this well. The GPU results are much better than I ever imagined I would see without airflow directly in from the side, near the video card cooler. The twin 140mm fans on top apparently are doing a good job pulling in fresh air through the various vents in the case below. If there is a downside, it is noise. The smaller fans produce more of it as they spin up to higher RPMs, trying to push as much air as their larger counterparts. None of the included fans is really what I would call high speed; the two 120mm fans in the front and rear spin at 1200 and 1500 RPM, respectively. The two 140mm fans on top only go around at 1000 RPM, which is lower than many of the 140mm case fans on the market. I think it's probably the lack of obstructions more than anything else; there's very little to impede the airflow and the same goes for the noise from the fan blades. Some of the noise is also coming from the CPU cooler and video card, I can't pin all the blame on the case fans. Well, I hope I learned my lesson this time. I had imagined that the PC-P50R would be an average, run-of-the-mill performer for cooling, based on its unassuming complement of fans and solid side panels. It just goes to show the value of critical testing, over the guesses and conjecture that can cause us to make up our mind about a product before giving it a fair chance.
|
|
Comments
I admit the drive can wiggle a tiny bit when the OP Side screws are not used, but that's just cosmetic. The drive isn't going anywhere.
BTW, what drive wereyou installing that neede to use the rear set of pins?
Having built numerous computers over the years, including about 40 Lian Li's for various clients, I would offer the following comments.
1) I have never, ever cut myself (or a wire) on a Lian Li - all the likely competitors should hang their heads in shame on this one. Blood sacrifices to the machine god are wearisome after a few hundred occasions.
2) Everything fits well - I have recently built 3 PC50 machines and their snap in 5.25 drives work well. Of course, showing my age here I guess, I still use the screws as well.
3) Using a gigabyte UD-7 motherboard (inbuilt passive air cooling), two gigabyte silent cell 5770 graphic cards and a Noctua NH-D14 CPU cooler (quietest fan speed), 16Gb Kingston RAM, Corsair 950 watt non-modular PSU (less connections) in a non-airconditioned, bare brick environment in Australia, the client can sleep in the same room with the machine on. Further, I cannot hear it until I am literally at the keyboard. Highest temperature - AMD-1100BE CPU 48 Celsius, Intel i7-960 CPU (12 Gb RAM) 63 Celsius, case AMD - 58 Celsius, Intel - 61 Celsius, ambient - often over 30 Celsius. Yes, I checked it several times - the AMD with the Noctua runs at a lower temp than the case - even when she played Crysis at the top settings for a whole day... (the Intel was actually being built for a different client, however she tested it in the same room because she wanted to see if it was faster). Client's previous system - three quarter height steel tower (expensive, famous gamer brand) usin!
g the AMD CPU and AMD heatsink/fan. It sounded like a loosely mounted jet engine and ran over 10 Celsius hotter case temperature. Lian Li's don't resonate!
4) The plastic strap across the top of the 3.5 hard disk drives gave me a hot spot on a server system, so I removed it & went back to the screws - gamers beware... my only criticism of the PC-P50 may be disk threatening. The PC-P80 or PC-A77F are a much better, cooler cradle IMHO.
5) I am picking up another PC50 later today - this time a "dragon" - and will be placing more in less the same bits in it. Unless the fans are a lot noisier than the "black" case, it should be very close to silent.
6) This is being written on my workhorse - a (1999?) Lian Li PC62, my other is a (2009?) PC-P80 - which is quieter and cooler than either the PC-P50 or the PC62.
7) A very thorough review - well done.
Arron, Western Australia.
Thanks for sharing all this excellent info, based on real world experience. I'm curious how you detected the hot spot(s) on the drive. Thermal imaging, or low-level environmental reporting from the drive itself?
Thanks, Bruce.
I sent the first email in because your review was so thorough - some gamer type hardware reviews are so poor I would not waste my time. I commend you on your approach, and wish there were more like you in your profession.
I share your view on quality vs features. Pity some coders out there clearly don't...
My digital "instant read" thermometer plus thermal sensing strips are my essential friends in these days of massive "so far over the top I am stunned" power consumption. The original pentium 100 CPU pulled 4 (well, 3.9 actually) watts average - count 'em, one hand. Last year an nvidia graphics card pulled nearly 1000 watts - scary or obscene?
Anyway, back to the PC-P50 - I intended to hammer it the next day, however after only two hours of reasonably hard use just getting a feel for the machine, the drive's sound changed a little. I measured 85 celsius under the strap, 67 near it, both readings where the disk platter would be spinning. A temperature variance of zero to +18C to zero in less than three inches across a high precision device spinning faster than an angle grinder - shudder. Clearly HDD internal frame warping and / or platter shattering is a no cost option. I repeated it with temperature strips the next day - 86 and 67C after two hours.
Result - bin the strap, use those old obsolete things called spare aluminium screws from another Lian Li case. FWIW, that strap is probably the only thing Lian Li have ever come up with that makes no sense at all to me - fashion over function I guess. Even the individual disk cradle runs noticeably warmer than the PC-P80 or PC-A77F's older, lighter and more ventilated "3 drives at a time predecessors".
Oh, and yes, aluminium screws do make a small difference, especially in aluminium frames, conducting heat away a little quicker as well as avoiding hot spots compared to their steel cousins. Screws and closing all gaps also prevents clients losing expensive software - like a Novell server 50 licence disk that disappeared for 4 years until we scrapped the previous assembler's machine ? a technician originally installed it and it had vanished when they needed to reinstall it ? it was between the CD drive and the hard disk, scratched beyond use and already replaced by a not happy insurance company.
More than enough for now, please keep up the good work.
Thanks for your feedback.
Arron.