AMD’s new stock cooling solution for their upcoming Zen processors features a taller two heatpipe design, a 92 mm fan, and an illuminated fan shroud.
February 22, 2016 by Lawrence Lee
| Product | AMD Wraith CPU Cooler |
| Manufacturer | AMD |
| Price | N/A |
Most retail processors ship with a simple down-blowing stock heatsink/fan that provides adequate CPU cooling and nothing more. They’re fairly compact which is good if you need to fit one in a small case, but the lack of mass and surface area equates to dismal performance compared to aftermarket coolers. Enthusiasts immediately toss them aside as they cannot handle the temperatures associated with heavy overclocking. Silent PC fanatics share their disdain as little consideration is placed on acoustics and high fan speeds are required under load which guarantees loud operation.
|
|
AMD recently unveiled new improved stock cooling units including a flagship cooler dubbed “Wraith” for their next generation Zen processors and included as sort of a preview with the 125W eight-core FX-8370. The current stock of FX-8370s with the older heatsink/fan are now discounted so the new improved cooler won’t cost you anything extra, though the SKU still carries a significant US$40 price premium over the FX-8350, which is the same chip with a 100 MHz slower Turbo frequency.
According to AMD, these new coolers “generate less than one-tenth the noise of their predecessors – running at a near-silent 39 decibels, about as quiet as a library” while the Wraith “delivers 34 percent more airflow and 24 percent more surface area for heat dissipation than its predecessor.” 39 dB is certainly not “near-silent” in our books but it’s clear just by looking at the Wraith that it’s a substantial improvement over the stock heatsinks we’re used to seeing.
|
|
Unfortunately we don’t have a FX-8370 stock cooler on hand for a direct comparison but it’s physically similar to the cooler pictured above. The basic anatomy of AMD boxed heatsink/fans has not changed in over a decade, featuring a boxy top-down design, a small fan clipped onto a plastic frame, and a spring clip mounting mechanism. The fins are tightly packed, dispersing air toward the sides toward the memory slots and VRM circuitry. The use of a copper base and heatpipes has been employed on hotter models since the days of the Athlon X2.
|
|
The Wraith has a similar two-heatpipe design but it’s noticeably taller than the older model (82 mm vs. 62 mm) and is equipped with a full-sized replaceable 92 mm fan and a fan shroud with LED lighting.
|
|
Two more stock coolers have also been announced alongside a crop of new AMD FM2+ CPUs/APUs, though their particulars have not been announced. The Athlon X4 870K/860K and A8-7670K/7650K (95W TDP) will be cooled by the heatsink picture above left, while the A10-7860K and Athlon X4 845 (65W TDP) will ship with the heatsink pictured on the right.
PHYSICAL DETAILS
According to my measurements, the heatsink weighs 340 grams and this increases to 460 grams (just over 1 lb) with the fan and shroud. The heatsink body stands 52 mm tall but with the frame and shroud installed, the total height is 82 mm which rules it out for some compact cases. It’s composed of 50 aluminum fins friction-fit to two 6 mm thick copper heatpipes, which in turn are soldered to a thin copper base plate.
|
|
|
|
|
|
|
|
|
|
|
|
TESTING
Before thermal testing, we took some basic physical measurements of the product(s) for comparison.
| Approximate Physical Measurements | |
| Weight | 340 g (heatsink only) 460 g (with stock fan and shroud) |
| Height | 82 mm |
| Fin count | 50 |
| Fin thickness | 0.29 mm |
| Fin spacing | 1.45 mm |
| Vertical Clearance* | N/A |
| * measured from motherboard PCB to bottom fin of heatsink | |
| Small Heatsink Comparison: Average Fin Thickness & Spacing | ||
| Heatsink | Fin Thickness | Fin Spacing |
| Scythe Big Shuriken 2 Rev.B | 0.29 mm | 1.13 mm |
| Noctua NH-L9i | 0.44 mm | 1.16 mm |
| Scythe Big Shuriken | 0.33 mm | 1.19 mm |
| Reeven Vanxie | 0.28 mm | 1.39 mm |
| Reeven Arcziel | 0.28 mm | 1.41 mm |
| AMD Wraith | 0.29 mm | 1.45 mm |
| Cooler Master GeminII M4 | 0.29 mm | 1.46 mm |
| Noctua NH-L12 | 0.49 mm | 1.51 mm |
| Scythe Kozuti | 0.12 mm | 1.69 mm |
| Noctua NH-D9L | 0.40 mm | 1.70 mm |
| Noctua NH-U12S | 0.45 mm | 1.72 mm |
| Scythe Samurai ZZ | 0.33 mm | 1.74 mm |
| Phanteks PH-TC14S | 0.41 mm | 1.76 mm |
| Thermalright TRUE Spirit 120M | 0.43 mm | 1.78 mm |
| Cryorig C1 | 0.42 mm | 1.78 mm |
| Noctua NH-U9S | 0.40 mm | 1.80 mm |
| Prolimatech Panther | 0.53 mm | 1.80 mm |
| SilverStone Argon AR02 | 0.30 mm | 1.85 mm |
| Phanteks PH-TC90LS | 0.47 mm | 1.90 mm |
Larger heatsinks are tested on our
LGA1366 heatsink testing platform, while smaller coolers tackle our
LGA1155
heatsink testing platform. Unfortunately AMD stock coolers are not cross-compatible so we’ve set up an AMD testing platform specifically for the Wraith. A summary of the test system and procedure follows.
AMD Heatsink Test Platform:
- AMD A10-6800K APU – 4.1 GHz, 32nm, 100W, socket FM2
- Asus F2A85-M Pro
motherboard – AMD A85X chipset, microATX - Kingston HyperX LoVo memory, 2 x 4GB, DDR3-1600 in dual channel
- Kingston HyperX 3K solid-state drive – 120GB, 2.5-inch, SATA 6 Gbps
- Seasonic X-460FL
power supply – 460W, ATX, 80Plus Platinum, passively cooled - Noctua NT-H1
thermal compound
The systems are silent under the test conditions, except for the CPU cooling
fan(s).
Normally, our reference fans are used whenever possible, the measured details
of which are shown below.
| Reference Noctua 140mm fan Anechoic chamber measurements | ||
| Voltage | Speed | |
| 12V | 1250 RPM | 28~29 dBA |
| 9V | 990 RPM | 21 dBA |
| 8V | 880 RPM | 18 dBA |
| 7V | 770 RPM | 15~16 dBA |
| 6V | 660 RPM | 13 dBA |
| Reference Nexus 120 mm fan Anechoic chamber measurements | ||
| Voltage | Speed | |
| 12V | 1080 RPM | 16 dBA |
| 9V | 880 RPM | 13 dBA |
| 7V | 720 RPM | 12 dBA |
| Reference Nexus 92 mm fan Anechoic chamber measurements | ||
| Voltage | Speed | |
| 12V | 1470 RPM | 17 dBA |
| 9V | 1280 RPM | 14 dBA |
| 7V | 1010 RPM | 12 dBA |
Measurement and Analysis Tools
- Extech 380803 AC power analyzer / data logger for measuring AC system
power. - 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 - Various other tools for testing fans, as documented in our
standard fan testing methodology. - SpeedFan,
used to monitor the on-chip thermal sensors. The sensors are not calibrated,
so results are not universally applicable. - Prime95,
used to stress the CPU heavily, generating more heat than most real applications.
All instances are used to ensure full stress. - CPU-Z,used to monitor the CPU speed to determine when overheating occurs.
Noise measurements are made with the fans powered from a separate, fanless system. Load testing was accomplished using Prime95 to stress the processor, and the
graph function in SpeedFan was used to ensure that the load temperature is stable
for at least ten minutes. The temperature recorded is the highest single core
reading. The stock fans were tested at various voltages to represent a good
cross-section of airflow and noise performance.
The ambient conditions during testing were 10~11 dBA and 21~23°C.
TEST RESULTS
Noise Measurements
|
|
| Specifications: AMD Wraith Stock Fan | |||
| Manufacturer | Delta Electronics | Power Rating | 3.84 W |
| Model Number | QFR0912H | Airflow Rating | 60.47 CFM |
| Bearing Type | ? | Speed Rating | 3200 RPM |
| Frame Size | 92 x 92 x 25 mm | Noise Rating | 37 dBA |
| Hub Size | 40 mm | Header Type | 4-pin PWM |
| Blade Diameter | 84 mm | Starting Voltage | < 4.0 V |
| Cable Length | 24 cm | Weight | 100 g |
| Corner Type | Open | Retail Availability | Yes |
Data in blue cells provided by the manufacturer or observed; data in the green cells were measured. | |||
Though the stock fan conforms to the same form factor as a standard 92 mm case fan, it’s design is clearly not intended for open air operation. It has a surprisingly large hub and short stubby blades arranged at overlapping angles. An identical fan is available at large electronic OEM suppliers though the specified power rating is a much lower 2.52W. This particular fan is probably rated higher as it has to power the LEDs on the shroud as well.
| Stock Fan Measurements | |||
| PWM Setting | Speed | [email protected] (minus shroud) | |
| 100% | 2930 RPM | 42~43 dBA | 39 dBA |
| 70% | 2270 RPM | 33~34 dBA | 31 dBA |
| 50% | 1800 RPM | 27~28 dBA | 24~25 dBA |
| 40% | 1590 RPM | 23~24 dBA | 21 dBA |
| 30% | 1370 RPM | 19~20 dBA | 17 dBA |
| 20% | 1170 RPM | 17 dBA | 15 dBA |
| 10% | 980 RPM | 15 dBA | 13 dBA |
| 0% | 690 RPM | 12~13 dBA | 11~12 dBA |
| Measuring mic positioned 1m at diagonal angle from the center of the heatsink. Ambient noise level: 10~11 dBA. | |||
Originally it was reported that the Wraith would not have an adjustable fan speed but our sample can be controlled like any other fan, producing an almost inaudible 12~13 [email protected] at minimum PWM speed and a ear-shattering 42~43 [email protected] at maximum speed in our anechoic chamber. Noise ratings should generally be taken with a grain of salt as there are no strict industry standards for measuring noise. That being said, the Wraith precisely hits its 39 dBA claim in our anechoic chamber, though only when the fan shroud is removed. Taking the shroud off makes the cooler noticeably quieter at all speeds.
|
|
|
|
With the shroud on, not only does the cooler produce more noise, it sounds worse overall as well. Its acoustics are rather complex depending on the speed with moderate tonality at various frequencies. It’s incredibly whiny at 100% speed, drones at 70%, develops an uneven hum at 50%, buzzes at 40%, and below that, there is an audible rattle/clicking noise.
Many of these undesirable characteristics are lessened by removing the shroud or squeezing the shroud during operation. It still doesn’t sound very pleasant with the shroud off but it’s a clear improvement and the shroudless cooler’s frequency distribution is clearly smoother. The fan and shroud are not securely as tightly as they should be, and the heatsink/fan assembly has quite a few parts. It’s likely that all these components interact at some level, rattling and shaking one another, generating extra noise.
Performance (A10-6800K 100W)
For today’s testing, we’re using an A10-6800K 100W APU rather than the 125W FX-8370 processor with which it ships. This will better indicate how the Wraith will perform with AMD’s upcoming 95W Zen chips and allows us to make closer comparisons with heatsinks previously tested on our Intel Sandy Bridge based small CPU cooler test platform.
| AMD Wraith (Shroudless) | |||
| PWM/Volt. Setting | Fan Speed | Thermal Rise | |
| 100% | 2930 RPM | 39 dBA | 22°C |
| 70% | 2270 RPM | 31 dBA | 25°C |
| 50% | 1800 RPM | 24~25 dBA | 27°C |
| 40% | 1590 RPM | 21 dBA | 31°C |
| 30% | 1370 RPM | 17 dBA | 33°C |
| 20% | 1170 RPM | 15 dBA | 40°C |
| 10% | 980 RPM | 13 dBA | 52°C |
| Reference 92 mm Fan (Nexus) | |||
| 12V | 1460 RPM | 18 dBA | 44°C |
| 9V | 1280 RPM | 16~17 dBA | 49°C |
| Red box indicates insufficient cooler (thermal throttling). | |||
The Wraith performs surprisingly well on our test platform, keeping the thermal rise below 35°C throughout most of its range. Cooling proficiency drops off gradually as the fan speed is reduced with no substantial reduction until the jump from 30% to 20% speed. However, it’s already fairly quiet at this level, it the 15~17 [email protected] range. At 10% speed (about 1000 RPM), it finally fails as the CPU clock speed begins to fluctuate, frequently throttling down to 2.0 GHz.
The cooler sounds miles better with our reference 92 mm fan but it fails miserably at replicating the performance of the stock fan. The difference is startling with the Nexus fan producing temperatures more than 10°C worse at similar noise levels. It’s easy enough to swap the fan but it would seem that a typical case fan is not well suited for this heatsink.
Comparison Tables
For a direct comparison, two of our favorite small aftermarket CPU coolers were re-tested on the same A10-6800K system. The Noctua NH-L12 is one of the best coolers on the market for smaller systems, shipping with both a 120 and 92 mm fan; This is somewhat unfair, so it was tested in single fan configuration, such that the cooler stands 66 mm rather than 93 mm tall. The Scythe Shuriken 2 is a sleeker 58 mm high cooling solution with only a slim 120 mm fan; We’ve used it to great success in low profile cases like the SilverStone FTZ01 and RVZ02.
| CPU Thermal Rise Comparison (°C): A10-6800K Test System | ||||||||||||||
| SPL ([email protected]) | 28 | 27 | 26 | 25 | 24 | 23 | 22 | 21 | 20 | 19 | 18 | 17 | 16 | 15 |
| Noctua NH-L12 (120 mm fan only) | – | 25 | – | – | 26 | – | – | 28 | – | – | 30 | – | ||
| AMD Wraith (no shroud) | – | – | – | 27 | – | – | 31 | – | – | – | 33 | – | 40 | |
| Scythe Big Shuriken 2 Rev.B | 27 | – | – | – | 30 | – | 37 | – | – | – | 51 | |||
| Noctua NH-L12 (92 mm fan only) | 29 | – | 31 | – | 33 | – | – | – | 38 | – | 47 | – | ||
Surprisingly, the Wraith stacks up quite nicely, edged out by a few degrees by the NH-L12 with its larger 120 mm fan installed. The Wraith beats the Big Shuriken 2 by a similar margin and its performance holds up much better at very low noise levels.
Using previous results from our Intel-based small heatsink test system and our even older Athlon II test platform, the relative performance of the Wraith and a pair of older Phenom II stock coolers has been estimated below:
| CPU Thermal Rise Comparison (°C): SPCR Small HSF Test System | ||||||||||||||
| SPL ([email protected]) | 25 | 24 | 23 | 22 | 21 | 20 | 19 | 18 | 17 | 16 | 15 | 14 | 13 | 12 |
| Noctua NH-L12 (120 & 92 mm fans) | – | – | – | – | – | 33 | – | – | 34 | – | 35 | – | 36 | |
| Thermalright TRUE Spirit 120M | 31 | – | – | – | – | – | 32 | – | – | – | 34 | 36 | 38 | |
| Noctua NH-U12S | – | – | – | – | – | – | 33 | – | 34 | – | – | 35 | 38 | |
| Phanteks PH-TC14S | – | – | – | 31 | – | – | – | 33 | – | – | 34 | – | 38 | – |
| Noctua NH-D9L | 30 | – | – | – | 33 | – | – | – | 34 | – | 38 | – | ||
| Cryorig C1 | – | – | 31 | – | – | – | 32 | – | – | 36 | – | – | – | |
| Noctua NH-L12 (120 mm fan only) | – | – | – | – | 36 | – | 37 | – | – | 38 | – | 39 | 42 | |
| SilverStone Argon AR02 | – | – | – | – | 36 | – | – | – | 38 | – | – | – | 43 | |
| AMD Wraith (no shroud) | ? | |||||||||||||
| Noctua NH-U9S | – | – | – | 34 | – | – | – | 36 | – | – | 40 | – | 45 | |
| Noctua NH-D9L (single fan) | – | – | – | 35 | – | – | – | 37 | – | – | 41 | – | 46 | |
| Scythe Big Shuriken 2 Rev.B | – | 37 | – | – | – | 39 | – | – | – | 43 | – | 48 | ||
| Scythe Samurai ZZ | 42 | – | – | – | – | – | – | 45 | – | – | 46 | – | 52 | |
| Noctua NH-L12 (92 mm fan only) | – | – | 39 | – | – | – | – | – | 44 | – | 47 | – | 51 | 57 |
| Scythe Big Shuriken | – | – | – | – | – | 43 | – | – | – | – | 46 | – | 61 | |
| Cooler Master GeminII M4 | – | 47 | – | – | – | – | – | – | 53 | 56 | – | 64 | ||
| Noctua NH-L9i | – | – | – | 54 | – | – | – | 56 | – | – | 61 | – | – | – |
| Scythe Kozuti | – | 53 | – | – | – | – | 57 | – | – | 62 | – | 65 | ||
| AMD Phenom II AM2+ Stock Cooler | ? | |||||||||||||
| Phanteks PH-TC90LS | – | – | – | – | 63 | – | – | 67 | – | – | – | 69 | – | – |
| AMD Phenom II AM3 Stock Cooler | ? | |||||||||||||
| Reeven Vanxie | 59 | – | – | – | – | – | 66 | – | – | – | – | 77 | – | F |
| SPL ([email protected]) | 25 | 24 | 23 | 22 | 21 | 20 | 19 | 18 | 17 | 16 | 15 | 14 | 13 | 12 |
The Wraith’s place on the chart is similar to that of tower heatsinks equipped with a 92 mm fan like the SilverStone Argon AR02. The older stock cooling solutions are near the bottom, in the same league as anemic low profile models like the Scythe Kozuti and Phanteks PH-TC90LS.
MP3 SOUND RECORDINGS
These recordings were made with a high
resolution, lab quality, digital recording system inside SPCR’s
own 11 dBA ambient anechoic chamber, then converted to LAME 128kbps
encoded MP3s. We’ve listened long and hard to ensure there is no audible degradation
from the original WAV files to these MP3s. They represent a quick snapshot of
what we heard during the review.
These recordings are intended to give you an idea of how the product sounds
in actual use — one meter is a reasonable typical distance between a computer
or computer component and your ear. The recording contains stretches of ambient
noise that you can use to judge the relative loudness of the subject. Be aware
that very quiet subjects may not be audible — if we couldn’t hear it from
one meter, chances are we couldn’t record it either!
The recording starts with 5 second segments of room ambiance, then the fan
at various levels. For the most realistic results, set the volume so that
the starting ambient level is just barely audible, then don’t change the volume
setting again.
- AMD Wraith stock fan at 1m
— at 20% PWM / 1170 PWM (17 dBA)
— at 30% PWM / 1370 RPM (19~20 dBA)
— at 40% PWM / 1590 RPM (23~24 dBA)
— at 50% PWM / 1800 RPM (27~28 dBA)
— at 70% PWM / 2270 RPM (33~34 dBA)
— at 100% PWM / 2930 RPM (42~43 dBA)
- AMD Wraith stock fan without shroud at 1m
— at 20% PWM / 1170 PWM (15 dBA)
— at 30% PWM / 1370 RPM (17 dBA)
— at 40% PWM / 1590 RPM (21 dBA)
— at 50% PWM / 1800 RPM (24~25 dBA)
— at 70% PWM / 2270 RPM (31 dBA)
— at 100% PWM / 2930 RPM (39 dBA)
FINAL THOUGHTS
The Wraith is considerably more effective than past AMD stock coolers, performing on the same level as small towers and modest-sized downblowing models. Standing 82 mm high, it’s considerably taller than its predecessors which makes it incompatible with many small cases. That’s one area it can’t compete with shorter solutions like the Noctua NH-L12 and Scythe Big Shuriken 2, which were designed with more compact enclosures in mind.
With the shroud removed, it manages to hit its 39 dB claim right on the money at full speed, though it’s not exactly an accomplishment. It’s very loud by any standards and the two AMD heatpipe coolers we reviewed six years ago were actually quieter. The shroud is poorly designed as the illuminated AMD logo is difficult to see from most angles and it makes the cooler louder. The stock fan doesn’t sound great; It’s tolerable but the shroud amplifies all its negative characteristics. The fan is easily replaced but the stock fan absolutely destroyed our Nexus 92 mm reference fan performance-wise, so swapping it out may result in a substantially higher temperatures.
The Wraith makes the price difference between the FX-8370 and FX-8350 less of an issue but its significance is more pertinent to AMD’s future than their current aging line of FX processors. Offering it only with the 125W FX-8370 is an amuse bouche, to whet our appetite for the release of their highly anticipated new CPU microarchitecture. Zen is scheduled to drop in Q4, reportedly, with a maximum TDP of just 95W. Conceivably this means both the noise-conscious and overclocking enthusiasts, or at least a portion of them, will think twice before considering aftermarket cooling.
Our thanks to AMD
for the Wraith CPU cooler sample.
* * *
SPCR Articles of Related Interest:
Phanteks PH-TC14S & Cryorig C1 CPU Coolers
New 92mm-fan Tower Coolers from Noctua
Cryorig R1 Ultimate & Universal CPU Coolers
Sub-$20 CPU Coolers: A Reader’s Roundup
SilverStone Argon AR02 CPU Cooler
AMD Phenom II Stock Coolers
* * *
