The new Fractal Design Define R5 arrived in our lab while we were finishing our original Quiet ATX Gaming Build Guide article. Lacking time to do a full case review, we swapped the R5 in place of the R4 to get a practical assessment of the improvements the new case brings. In the process, we managed to drop the full load noise of our Quiet ATX Gamer to under 17 dBA@1m.
The Quiet ATX
Gaming Build Guide we posted last week was an obvious hit, judging by
the traffic and discussion it provoked. Nice that our work is appreciated, but
no resting on laurels this week.
In the middle of our work on that system, we took delivery of a sample of the
new Fractal Design Define R5, the next generation of the R4 which will
be in the stores in December. A 10-minute examination was enough to convince
us: It is a better case than the R4.
How much better? Usually we answer this kind of question in a series of exacting
tests with a reference system so acoustic and thermal performance can be compared
easily with other competitors. In this case, knowing there wasn’t enough time
to do a complete case review, we decided to do the next best thing. Just swap
it in place of the R4 used in our Quiet ATX Gaming Build and examine the differences.
Quicker, and about as good, killing two birds with one stone: A pretty thorough
silent-slanted review of the R5, and an upgrade to the ATX Gaming PC with a
single component swap.
So here is SPCR’s Quiet ATX Gaming Build, R5 Version. We’ll be referring to
our article on the first version, so hopefully you have read it already, or
have it open in another browser tab.
COMPONENT SELECTION
No other components have been changed, just the case. Here’s the component
list.
FRACTAL DESIGN DEFINE R5: A QUICK LOOK
Since this case is the only reason for our update, it deserves at least a quick
run through the key features. You can also check out the official Fractal
Design R5 page.
|
|
|
When the door is closed, the space between the dust filter and the side slots is over a centimeter, which means the R5 will draw in air a lot more easily, with lower RPM, than the R4. |
|
|
|
|
SYSTEM ASSEMBLY
Assembling the system was very straightforward. The Define R5 is spacious enough,
the cable routing is great, and we already had the CPU cooler mounted on the
motherboard from the first Quiet ATX Gaming rig. Here are photos of the finished
build. As in the original build, the 2TB Seagate HDD was suspension-mounted
in the optical drive bay using the same elastic cord.
|
|
|
TESTING
System Configuration:
- Intel Core i7-4770K processor – 3.5 GHz (3.9 GHz with Turbo
Boost), 22nm, 84W, integrated HD 4600 graphics - Scythe Kotetsu CPU cooler
- ASUS Z97-PRO motherboard – Intel Z97 chipset, ATX
- ASUS Strix GeForce GTX 980 graphics card – 2048 CUDA cores, 1178
MHz clock (1279 MHz with GPU Boost), 7010 MHz memory - Kingston HyperX Genesis memory – 2x4GB, DDR3-1600, C10
- Kingston HyperX 3K SSD – 240GB, SATA 6 Gbps, 2.5-inch
- Seagate Desktop SSHD hybrid drive – 2TB, 7200 RPM, 8GB NAND
Flash, SATA 6 Gbps, 3.5-inch - Fractal Design Define R5 case – ATX
- be quiet! Straight Power 10 power supply – ATX, modular, 600W
- Microsoft Windows 7 Ultimate operating system, 64-bit
Measurement and Analysis Tools
- Media Player
Classic – Home Cinema to play H.264/VC-1 video. - TMPGEnc
Xpress video encoder for stress testing. - Resident
Evil 6 standalone benchmark for stress testing. - Prime95 processor
stress software. - FurMark
openGL GPU stress software. - CPU-Z
to monitor CPU frequency and voltage. - GPU-Z to
monitor GPU temperature, load level, and fan speed. - AIDA64 to monitor
system temperatures and fan speeds. - ASUS Fan Xpert 3 to monitor system temperatures and monitor/change
fan speeds. - Extech AC Power Analyzer 380803 AC power meter, used to measure system
power consumption. - PC-based
spectrum analyzer: SpectraPlus with ACO Pacific mic
and M-Audio digital audio interfaces. - Anechoic
chamber with ambient level of 11 dBA or lower
Baseline Noise
Before any stress testing, we examined the noise of each part by leaving the
system on idle, and stopping/unplugging all the other fans (and the hard drive)
to isolate individual components.
GTX 980 + Define ATX System Component Noise Levels (PSU on idle) | |||
Components | Fan Speed | R4 SPL | R5 SPL |
PSU (Baseline) | N/A | 11 dBA | 11 dBA |
HDD hard-mounted | N/A | 14 dBA | 14 dBA |
GPU fan | 740 RPM (min) | 16 dBA | 16 dBA |
CPU fan | 340 RPM (min) | 11~12 dBA | 11~12 dBA |
650 RPM (~5V) | 12 dBA | 12 dBA | |
900 RPM (~7V) | 14~15 dBA | 14~15 dBA | |
1120 RPM (~9V) | 19~20 dBA | 19~20 dBA | |
1400 RPM (max) | 25 dBA | 25 dBA | |
Case fans | 500 RPM (min) | 14 dBA | <14 dBA |
620 RPM (~7V) | 17~18 dBA | 15~16 dBA | |
740 RPM (~9V) | 21 dBA | 20 dBA | |
920 RPM (max) | 25~26 dBA | 24~25 dBA | |
Mic positioned 1m at diagonal angle left/front of case. Ambient noise level 10~11 dBA@1m. |
A simple fact: The new fans in the R5 are a bit quieter. Holding them up close,
at full speed, the new ones feel like they might have slightly less airflow,
they spin a touch slower, and they have much less buzzing noise. As speed is
reduced, the 1~2 dBA noise advantage holds till the ambient floor of the anechoic
chamber is approached. There’s still an audible difference at 400 RPM, the new
fan sounds smoother, but it can’t be measured any more, the ambient noise begins
to intrude. This is promising.
Stress Testing: CPU-centric
We dispensed with lower power states and conducted only the higher power tests.
System Measurements | ||||
System State | Define R4 | Define R5 | ||
Video Encoding | Prime95x8 | Video Encoding | Prime95x8 | |
CPU Fan | 340 RPM | 650 RPM | 340 RPM | 650 RPM |
CPU | 53°C | 64°C | 53°C | 63°C |
MB | 33°C | 33°C | 33°C | 32°C |
SSD | 27°C | 24°C | 32°C | 35°C |
HDD | 33°C | 33°C | 33°C | 32°C |
GPU | 38°C | 38°C | 38°C | 38°C |
SPL@1m | 15 dBA | 14 dBA | ||
System Power (AC) | 106W | 131W | 106W | 131W |
System fans at 500 RPM (minimum), GPU fans off (auto). Ambient temperature: 21°C. |
This was an underwhelming result. The main points to note are:
1. The SSD mounted behind the motherboard in the R5 ran hotter: 5°C during
video encoding, and 11°C hotter during maximum CPU load. It’s easy to
see why. The SSD received airflow mounted in the front of the R4, and ran
cooler under the higher load because the CPU cooler fan provided a bit more
peripheral airflow. In contrast, when mounted behind the motherboard, the
SSD is isolated from airflow and affected somewhat only by the temperature
of the CPU. The highest temperature was just 35°C, way below the 70°C
maximum operational limit specified by Kingston.2. There was a measured 1 dBA difference in noise level but it may be too
small to be clearly audible to everyone.
Stress Testing: GPU-centric
Again, because we’ve already presented data on all the various power states
in the original article with the R4, we go straight to Prime95x2 + FurMark
results. As before, a custom fan profile was used in GPU Tweak to start the
GPU cooler fan at 35% when the GPU reaches 60°C, then 38% fan speed at 85°C,
and finally, 60% at 99°C
System Measurements Prime95x2 + FurMark | ||||
System State | Define R4 | Define R5 | ||
GPU Fan Speed | 990 RPM* | 1630 RPM (auto) | 950 RPM | 1600 RPM (auto) |
CPU Temp | 63°C | 61°C | 64°C | 61°C |
MB Temp | 46°C | 44°C | 47°C | 45°C |
SSD Temp | 26°C | 26°C | 36°C | 36°C |
HDD Temp | 36°C | 36°C | 36°C | 35°C |
GPU Temp | 85°C | 72°C | 87°C | 73°C |
SPL@1m | 18~19 dBA | 25 dBA | 17 dBA | 25 dBA |
Power (AC) | 291W | 290W | 291W | 290W |
*Speed adjusted to keep GPU temperature under 85°C. CPU fan at 650 RPM, system fans at 500 RPM (minimum). Ambient temperature: 21°C. |
There was a difference in the GPU fan speed according to the monitoring utilities,
but it is probably too small to make any acoustic difference. The GPU ran 2°C
hotter in the R5, and that system also measured upwards of 2 dBA quieter. This
is definitely audible.
When we reverted the GPU fans to default automatic control, the fan speed kicked
up by more than 600 RPM, with almost identical results in both cases: Extra
cooling for the CPU and MB, while the GPU core itself received 13°C of relief.
The noise level increased to 25 dBA, in both cases; any differences in the case
fan levels were swamped by the noise of the GPU fans.
Fan Swap?
We noted that the stock Fractal R4 case fans could be upgraded with quieter
fans. Is it worthwhile to try that with the quieter R5 fans? Let’s compare the
R4 and R5 fans to the Antec TrueQuiet 140 we suggested as an upgrade candidate
before. These are new measurements with each fan soft-mounted by itself outside
the case, with the mic 1m away from the intake side of the fan. (The Antec fan’s
top speed is just under 800 RPM, which is why its 870 RPM box is blank.)
FANS COMPARED | |||
RPM | R4 | R5 | Antec TQ140 |
SPL: dBA@1m | |||
870 | 19.4 | 17 | – |
750 | 15.6 | 14 | 15 |
500 | 13.5 | 12.5 | 12* |
*We measured 11 dBA in our last 140mm fan roundup, but readings at such low levels are easily affected by outside noise, which could be higher this time around, and there could be a sample variance. |
This is a revelation. The stock R5 fan is at least as quiet as the Antec TrueQuiet
140, which is one of the quietest we’ve measured. So there’s no benefit at all
to swapping out the R5 fans. They are as quiet any 140mm fans we’ve heard.
An Extra Fan
Since the case fans are actually at a level lower than the dominant sound source
at load, would adding another fan help cooling and perhaps reduce GPU
fan speed/noise without increasing the overall noise of the system? It was worth
trying.
|
The first trial was with one of the fans from the R4 mounted on the floor vent
directly below the VGA card, to blow fresh air up to it. The fan speed was kept
at the same 500 RPM as the other two fans. The result was puzzling: There was
perhaps a half dBA rise in noise at idle and no change at maximum load, and
only a tiny reduction in GPU and PCH temperature, just a couple degrees, not
enough even to be out of the margin of measurement error. Is the bottom intake
too impeded by the close proximity of the floor beneath? We didn’t investigate
further.
Moving on to trial 2: This time, the bottom HDD cage was removed, and an Antec TQ140 mounted in the front under the stock intake fan.
|
Here, we saw a positive result. Temperatures of all the components dropped
by several degrees while overall noise was unchanged. Heartened by this, the
GPU fan profile was changed ever so slightly, down to 37% speed (a drop of 1%)
at 80°C. Only the GPU temperature rose, by just 3°C, but the slight
change in GPU fan speed was enough to drop the overall SPL down a bit as well,
to 16.3 dBA@1m. Now this is a gaming system we can definitely live with!
Define R5 System Measurements Prime95x2 + FurMark | |||
System State | 2 case fans | 3 case fans | |
GPU Fan Speed | 950 RPM | 950 RPM | 900 RPM |
CPU Temp | 64°C | 61°C | 61°C |
MB Temp | 47°C | 43°C | 43°C |
SSD Temp | 36°C | 36°C | 36°C |
HDD Temp | 36°C | 36°C | 36°C |
GPU Temp | 87°C | 81°C | 84°C |
SPL@1m | 17 dBA | 17 dBA | 16.3 dBA |
Power (AC) | 291W | 291W | 291W |
CPU fan at 650 RPM, case fans at 500 RPM (minimum). Ambient temperature: 21°C. |
CONCLUSIONS
The main change here from last week’s Quiet ATX Gaming Rig is the Fractal Design
Define R5. The new case is a success, better than the R4 in many ways, but from
a purely functional point of the view, the main improvements are improved intake
vents and quieter fans. The ~2 dBA improvement at full load of the R5 setup
over the R4 has to be attributed mostly to the new fans, though undoubtedly,
the improved front venting probably comes into play as well. The final system
SPL of <14 dBA@1m at low load and just 16.3 dBA at maximum load was achieved
by adding a single quiet front fan, and removing all the HDD drive cages so
the airflow from the front fans is completely unimpeded. The end result was
a slight reduction in GPU fan speed, which helped reduce the overall noise,
and slightly cooler temperatures at all the sensor points.
Again, we have to laud nVIDIA for its improvements in GPU power efficiency
which makes the semi-passive cooling in the ASUS Strix GTX 980 possible. This
build would be a lot more complex and expensive without the GTX 980 and the
ASUS Strix cooler.
The total cost of our build should be unchanged, because the R5 case is apparently
going to be selling for about the same price as the R4, though we have not received
official word from Fractal. The gaming system we present here is hard to beat
for performance or noise, however, unless you throw a lot more money
at the build. Choose some of the more value-conscious alternatives suggested,
like the GTX 970, and you can reduce the cost by about US$400 while still maintaining
a high quality gaming experience.
Many thanks to Intel,
ASUS, Kingston,
Seagate, Scythe,
Fractal Design,
and be quiet! for
sponsoring the components in this build guide.
* * *
Articles of Related Interest
Quiet ATX Gaming
Build Guide
ASUS Z97-PRO LGA1150 Motherboard
Crucial MX100 512GB
& Samsung 850 Pro 256GB SSDs
Silent Mid Gaming PC Build Guide
Recommended Power Supplies
Recommended
Hard Drives
Case
Basics & Recommendations
Recommended Heatsinks
* * *