You are here

PowerColor Radeon HD 5850: Worth the Wait

PowerColor Radeon HD 5850: Worth the Wait

January 11, 2009 by Lawrence Lee

Product
PowerColor Radeon HD 5850 1GB

AX5850 1GBD5-MDHG

PCI-E Graphics Card
Manufacturer
Street Price
~US$300

For PC gamers one of the biggest stories of 2009 was the launch of the Radeon
HD 5000 series, most notably the higher-end HD 5870 and 5850. They received
rave reviews, extolling their high levels of performance and excellent energy
efficiency. As a result, they became massively popular and created a demand
with which ATI/AMD was unable to cope. Their initial production runs had very
poor yields, the same problem that plagued the release of their first 40nm GPU,
the HD 4770. Cards trickled
in slowly to distributors for most of the last quarter and the stock that did
become available in stores were quickly gobbled up. The predicament was exacerbated
by the allocation of a significant portion of 40nm chips for the release of
the dual GPU HD 5970.



The PowerColor HD 5850 box.

This issue was so widespread that we weren't able to obtain a sample until
recently through retail channels, a PowerColor Radeon HD 5850 1GB which current
ships with a free Steam copy of Dirt 2. Other than that, this model is basically
the same as most other HD 5850s, using the reference PCB and cooler. Our sample
had only a CrossFire bridge and DVI to VGA adapter inside the box which wasn't
too surprising as PowerColor has developed a reputation for being stingy with
accessories.



Package contents.




Technical specifications according to GPU-Z.

Compared to the Radeon HD 4890,
the fastest single GPU from ATI's last generation of graphics cards, the HD
5850 sports more than twice the transistor count (more than 2 billion), made
possible by an 18% larger die and upgrading the fabrication process from 55nm
to 40nm. The 5850 also has DirectX 11 support and an 80% increase in shader
units (1440 vs. 800). With all the extra hardware it will be interesting to
see whether how well they managed noise, cooling, and energy efficiency.

Gaming aside, the HD 5000 series has a couple of other benefits that can't
be assigned an objective value. ATI's new EyeFinity
multi-display technology allows users to output to three displays simultaneously
without requiring the use of a second GPU. Using three monitors one could immerse
themselves in a surreal gaming experience, or simply increase productivity with
the extra screen real estate. Another improvement is the ability to bitstream
Dolby TrueHD and DTS-HD audio directly over HDMI without uncompressing the signal,
a feature that until now has been available only on high-end sound cards.

Specifications: PowerColor Radeon HD 5850

(from the
product web page
)
Graphics Engine RADEON HD5850
Video Memory 1GB GDDR5
Engine Clock 725 MHz
Memory Clock 1000MHz (4.0Gbps)
Memory Interface 256bit
DirectX Support 11
Bus Standard PCIE 2.1

PHYSICAL DETAILS

The HD 5850 utilizes a closed-off box style cooler reminiscent to Nvidia's
GTX 200 series. (Editor's Tip: With this type of cooler, increasing
the airflow in the PC case beyond a certain point provides no improvement in
cooling becase the cooler's fan itself is the airflow gate; the air can't flow
through the heatsink fins any faster than that fan spins. Reducing the temperature
of the intake air can help. One way to optimize the noise/cooling/airflow is
to monitor the intake air temperature at the VGA cooler — ideally with
the GPU under full load — and adjust case fan speeds to the point where
increased speed provides very little temperature drop.)



As EyeFinity supports triple display configurations, the HD 5850 is equipped
with four digital outputs, one DisplayPort, one HDMI, and two DVI.




The PCB measures the same as the HD 4870/4890 — 24.1cm or 9.5",
slightly less the 9.6" standard width of an ATX motherboard. As a
result it will fit in the majority of ATX cases without difficulty, even
when you account for the extra inch required by the power connectors which
plug into the right side.




A small vent on the top side of the cooler may push some heat into the
case.




With four digital outputs on the back panel, there is only room for a
small exhaust port.




A pair of 6-pin PCI-E power connectors are required to drive the 5850.

HEATSINK

Warning — removing the heatsink from a card generally voids the product's
warranty. Do so at your own risk. Note that all testing on the card was performed
before the cooler was removed.



Removing the stock cooler is a simple affair, involving the removal of
several screws accessible from the back of the circuit board. The four
mounting holes around the GPU core has the same dimensions as the HD 3000
and 4000 series.




The heatsink has an anatomy similar to the HD 4870/4890 cooler with four
removable parts, a two heatpipe heatsink for the GPU core, the fan, a
memory/VRM cooling plate, and the top cover. Unfortunately we weren't
able to separate the cooling plate as the screws securing the cover were
too small and tight for the tools we had on hand.




The stock fan manufactured by AVC has a massive 14.4 watt power rating.




One of our favorite heatsinks, the Accelero
S1
in incompatible with the HD 5850 as there is a box cover over
the DVI ports blocking it, though you could easily bend a few fins to
make it work. For reference, the distance between the edge of the DVI
cover and the center of the bottom left mounting hole is 57mm.




The Scythe Musashi
has much of its body shifted to the right side, so it cleared the DVI
cover by a few millimeters.

TEST METHODOLOGY

Our test procedure is an in-system test, designed to:

1. Determine whether the cooler is adequate for use in a low-noise system.
By adequately cooled, we mean cooled well enough that no misbehavior
related to thermal overload is exhibited. Thermal misbehavior in a graphics
card can show up in a variety of ways, including:

  • Sudden system shutdown, reboot without warning, or loss of display signal
  • Jaggies and other visual artifacts on the screen.
  • Motion slowing and/or screen freezing.

Any of these misbehaviors are annoying at best and dangerous at worst —
dangerous to the health and lifespan of the graphics card, and sometimes to
the system OS.

2. Estimate the card's power consumption. This is a good indicator of how efficient
the card is and will have an effect on how hot the stock cooler becomes due
to power lost in the form of heat. The lower the better.

Test Platform

Measurement and Analysis Tools

Testing Procedures

Our first test involves recording the system power consumption using a Seasonic
Power Angel as well as CPU and GPU temperatures using SpeedFan and GPU-Z during
different states: Idle, under load with CPUBurn running to stress the processor,
and ATI and FurMark running to stress both the CPU and GPU simultaneously. This
last state is an extremely stressful, worst case scenario test which generates
more heat and higher power consumption than can be produced by a modern video
game. If it can survive this torture in our low airflow system, it should be
able to function nominally in the majority of PCs.

The software is left running until the GPU temperature remains stable for at
least 10 minutes. If artifacts are detected by ATI's artifact scanner or by
eye or any other instability is noted, the heatsink is deemed inadequate to
cool the video card in our test system.

If the heatsink has a fan, the load state tests are repeated at various fan
speeds (if applicable) while the system case fan is left at its lowest setting
of 7V. If the card utilizes a passive cooler, the system fan is varied instead
to study the effect of system airflow on the heatsink's performance. System
noise measurements are made at each fan speed.

Our second test procedure is to run the system through a video test suite featuring
a variety of high definition clips. During playback, a CPU usage graph is created
by the Windows Task Manger for analysis to determine the average CPU usage.
High CPU usage is indicative of poor video decoding ability. If the video (and/or
audio) skips or freezes, we conclude the GPU (in conjunction with the processor)
is inadequate to decompress the clip properly. Power consumption during playback
of high definition video is also recorded.

H.264/VC-1 Test Clips

H.264 and VC-1 are codecs commonly used in high definition movie videos on
the web (like Quicktime movie trailers and the like) and also in Blu-ray discs.
To play these clips, we use Cyberlink PowerDVD.



1080p | 24fps | ~10mbps
1080p H.264:
Rush Hour 3 Trailer 2c
is a 1080p clip encoded in H.264
inside an Apple Quicktime container.





1080p | 24fps | ~8mbps
WMV-HD:
Coral Reef Adventure Trailer
is encoded in VC-1 using the
WMV3 codec commonly recognized by the "WMV-HD" moniker.


x264/MKV Video Test Clips

MKV (Matroska) is a very popular online multimedia container
used for high definition content, usually using x264 (a free, open source
H.264 encoder) for video. The clips were taken from two longer videos
— the most demanding one minute portions were used. To play them
we use Media Player Classic Home Cinema, configured in the most suitable
manner depending on the GPU. For Intel/ATI graphics the player is configured
to use DXVA (DirectX Video Acceleration) and for Nvidia graphics we use
CoreAVC to enable CUDA (Compute Unified Device Architecture) support.



720p | 24fps | ~11mbps

x264 720p: Undead Battle is a 720p x264 clip encoded
from the Blu-ray version of a major motion picture. It features
a battle with undead warriors.




1080p | 24fps | ~14mbps

x264 1080p: Spaceship is a 1080p x264 clip encoded from
the Blu-ray version of an animated short film. It features a hapless
robot trying to repair a lamp on a spaceship.


Estimating DC Power

The following power efficiency figures were obtained for the
Seasonic S12-600
used in our test system:

Seasonic S12-500 / 600 TEST RESULTS
DC Output (W)
65.3
89.7
148.7
198.5
249.5
300.2
AC Input (W)
87.0
115.0
183.1
242.1
305.0
370.2
Efficiency
75.1%
78.0%
81.2%
82.0%
81.8%
81.1%

This data is enough to give us a very good estimate of DC demand in our
test system. We extrapolate the DC power output from the measured AC power
input based on this data. We won't go through the math; it's easy enough
to figure out for yourself if you really want to.

TEST RESULTS

BASELINE, with Integrated Graphics: First, here are the results of
our baseline results of the system with just its integrated graphics, without
a discrete video card. We'll also need the power consumption reading during
CPUBurn to estimate the actual power draw of discrete card later.

VGA Test Bed: Baseline Results

(no discrete graphics card installed)
System

State
System Power
AC
DC (Est.)
Idle
75W
Unknown
CPUBurn
151W
121W
Ambient temperature: 22°C

PowerColor Radeon HD 5850 1GB:

VGA Test Bed: PowerColor Radeon HD 5850 1GB
System

State
Fan Speed
SPL

@1m
Temps
System Power
GPU
VRM
AC
DC
Idle
1160 rpm
18 dBA
40°C
36°C
99W
77W
CPUBurn
44°C
38°C
175W
142W
CPUBurn + ATITool
1570 rpm
19 dBA
67°C
53°C
243W
199W
CPUBurn + FurMark
2120 rpm
24 dBA
83°C
82°C
309W
253W
Ambient temperature: 22°C

Ambient noise level: 11 dBA

System noise level: 12 dBA

The reference cooler kept the graphics card very well cooled by modern standards,
with the GPU and VRM temperatures well below 50°C when idle and only slightly
above 80°C at full load.

Noise

The GPU exhibited some of the screeching associated with coil whine particularly
during ATITool's 3DView, but it was not as sharp as the HD 4000 series.

The noise generated by the fan when idle was measured at 18 dBA, low enough
for most users. Still, it's a lot more tolerable placed under a desk rather
than atop it — we'd recommend an aftermarket cooler for the latter scenario.
The fan had some tonal properties, audible in the form of a low-pitched hum,
but it was well muffled by our case's side panel. When placed on load, the fan's
higher speed generated enough turbulence to mask the hum, blending the overall
noise into smoother broadband profile.



Our VGA test system idle with the PowerColor Radeon HD 5850 installed.
Note the sharp tonal peak between at ~240Hz.



At load.


Noise & Cooling Comparison
Card
Idle
Load
SPL @1m
GPU Temp
SPL @1m
GPU Temp
ATI HD 4870
13 dBA
82°C
20 dBA
91°C
HIS HD 4890 Turbo
15 dBA
66°C
26 dBA
80°C
PowerColor HD 5850
18 dBA
40°C
24 dBA
83°C
Ambient noise level: 11 dBA

System noise level: 12 dBA

When we compare the SPL and GPU temperature measurements with that of the HD
4870 and HD 4890, it seems that the HD 5850's fan is much more aggressive when
idle. In most systems the GPU is dormant during the majority of operation so
a reduction in idle noise would be welcome. The HD 5850 certainly has plenty
of headroom as 40°C is extremely low for a high-end graphics card. We would
gladly sacrifice 20°C or more in exchange for a reduction in fan speed.
On load, the fan behavior is more sensible, producing 2 dBA less than the HD
4890 while keeping the GPU temperature within a few degrees, which is surprising,
given the smaller exhaust vent compared to the 4800 series cards. Perhaps those
small vents on the top help out after all.

Power

The power consumption of an add-on video card can be estimated by comparing
the total system power draw with and without the card installed. Our results
were derived thus:

1. Power consumption of the graphics card at idle - When CPUBurn is
run on a system, the video card is not stressed at all, and stays in idle mode.
This is true whether the video card is integrated or an add-on PCIe 16X device.
Hence, when the power consumption of the base system under CPUBurn is subtracted
from the power consumption of the same test with the graphics card installed,
we obtain the increase in idle power of the add-on card over the
integrated graphics chip (Intel GMA950). (The actual idle power
of the add-on card cannot be derived, because the integrated graphics does draw
some power — we'd guess no more than a watt or two.)

2. Power consumption of the graphics card under load - The power draw
of the system is measured with the add-on video card, with CPUBurn and FurMark
running simultaneously. Then the power of the baseline system (with integrated
graphics) running just CPUBurn is subtracted. The difference is the load power
of the add-on card. (If you want to nitpick, the 1~2W power of the integrated
graphics at idle should be added to this number.) Any load on the CPU from FurMark
should not skew the results, since the CPU was running at full load in both
systems.

Both results are scaled by the efficiency of the power supply (tested
here
) to obtain a final estimate of the DC power consumption.

Estimated Power Consumption Comparison (DC)
Card
Idle
Load
Asus GeForce 9400GT 512MB
11W
23W
PowerColor HD 4650 512MB
15W
28W
ATI HD 4830 512MB
18W
87W
PowerColor HD 5850 1GB
21W
132W
ATI HD 4770 512MB
28W
60W
Asus GTX 260 896MB
35W
145W
ATI HD 4870 1GB
67W
134W
HIS HD 4890 Turbo 1GB
71W
149W

We've criticized ATI in the past for the relatively high idle power consumption
of their graphics cards when compared to Nvidia's offerings, but the HD 5000
series seems to have finally knocked out that particular problem. The HD 5850
uses about 21W when idle, a massive reduction compared to the HD 4890, impressively
low for a high-end video card. ATI also managed a small reduction on load which
is also admirable when you consider than its supposed to be 20~30% faster than
the HD 4870/4890.

Video Playback

Test Results: Video Playback
Test State
Average

CPU Usage
Additional

DC Power*
Rush Hour

(1080p H.264)
5%
+25W
Coral Reef

(WMV-HD)
27%
+44W
Undead Battle

(720p x264)
4%
+22W
Spaceship

(1080p x264)
3%
+22W
*compared to idle

The HD 5850 passed our video playback test with flying colors, with H.264 and
x264 playback adding only 25W and 22W respectively to DC system power, much
of which can be attributed to the GPU as CPU usage was very low. Our WMV-HD
clip required more CPU cycles and we noticed a bug: the core and memory clocks
of the graphics card increased to the maximum 700/1025Mhz setting rather than
the UVD speeds of 400/900MHz. As a result, the total system power draw increased
by 44W, which is unusually high.

BIOS

Some of the card's inner workings can be revealed by taking a look at the BIOS.
We used GPU-Z to extract the board's BIOS and Radeon
BIOS Editor
to examine its contents.



Clock settings. Boot/3D setting in red, idle in green, UVD in yellow.

Clock speeds of 725/1000MHz in 3D, 157/300MHz idle, and 400/900MHz when UVD
is at work were confirmed using GPU-Z.



Fan settings.

The stock fan's speed rises steeply, running at 22% until the core temperature
reaches 55°C and increasing linearly until 102°C when it reaches its
maximum speed.

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 5~10 seconds of room ambiance, followed by 5~10 seconds
of the VGA test system without a video card installed, and then the actual product's
noise 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.

Comparable system sound files:

FINAL THOUGHTS

Gaming: Please check out gaming-oriented reviews of the HD 5850 at techPowerUp,
HardwareCanucks,
and The
Tech Report
. The general consensus is that the HD 5850 is the second fastest
single GPU ATI video card currently on the market and is aptly suited for gaming
at 1920x1200 resolution and above. It holds a sizable performance advantage
over the HD 4890 and is only
moderately slower than the HD 5870 which holds a $100 price premium. In most
tests, the 5850 also beats Nvidia's closest competitor, the GTX 285, which retails
for $70 to $100 more.

Cooling: The box style stock cooler adopted by ATI is too aggressive
when idle. It is clearly audible and noticeably louder than previous generation
ATI HD coolers. That being said, most users will find the noise level tolerable,
though those used to a quiet PC will want to put a third party cooler on, place
the system under a desk, or attempt to modify the fan behavior in the BIOS.
On load, it is fairly quiet compared to most high-end graphics cards and the
noise level is low enough that in-game volume will undoubtedly drown it out.

Power Consumption: By our estimates, the Radeon HD 5850 requires only
21W when idle which is extremely impressive for one of the fastest video cards
available. By comparison the HD 4870 and 4890 use more than 3 times as much.
The high idle power consumption penalty typically associated with a topnotch
gaming cards does not apply to the HD 5850. On load, it uses slightly less than
the HD 4870/4890 even
though it offers stronger performance.

The biggest problem with the Radeon HD 5850 has been its high popularity. Plagued
by low yields at first, it was near impossible to find a 5850 after the initial
launch. The limited availability resulted in a rare price increase, but this
did not reduce demand at all. As Nvidia attention has been focused elsewhere,
ATI is the only game in town when it comes to high-end graphics cards, at least
if you want good value. With the stock situation finally beginning to reach
resolution we can say with confidence that the wait was worth it.

We have a few quibbles with the card, but none particularly significant. The
DVI cover interferes with some third party coolers, though heatsinks like the
Accelero series with malleable fins can be tweaked to avoid it. The stock cooler,
while quiet enough for most, can certainly be improved, but this point of contention
is far from notable as finding a reference heatsink we genuinely like seems impossible.
The package provided by PowerColor is very basic with a CrossFire bridge and
DVI to VGA adapter, but no 6-pin PCI-E power adapter cables. The card and cooler
are basically the same as the reference design as are the majority of HD 5850's
on the market, excluding the overclocked versions.

Its strengths are numerous and substantial. While energy efficiency is only
slightly better than ATI's previous generation on load, its low idle draw is
an amazing accomplishment. Using just a shade over 20W, gamers no longer have
to feel guilt over their GPU sucking up copious amounts of energy just surfing
the web or performing other mundane tasks. The EyeFinity feature is a landmark
on its own; the HD 5000 is the first series of mainstream desktop graphics card
that can drive triple display configurations without help from an IGP or a second
discrete card. Audiophiles have a lot to cheer about, as the ability to bistream
Dolby TrueHD and DTS-HD audio has been added. And even if you judge the HD 5850
on gaming performance alone, it still offers a far better value than anything
in its price bracket. Even the physical dimensions of the card are commendable
— an inch shorter than Nvidia's GTX 200 series, making it compatible with
more cases. It doesn't have the ultra-quiet acoustics we demand for an Editor's
Choice product, but the PowerColor Radeon HD 5850 1GB clearly deserve a strong
recommendation.

PowerColor Radeon HD 5850 1GB
PROS



* Excellent 3D Performance

* Well cooled

* Superb power consumption in idle

* Lower power consumption than HD 4870/4890 on load

* EyeFinity

* Bitsreams Dolby TrueHD/DTS-HD audio

* Free copy of Dirt 2 (temp. promo)

* Fits in most ATX cases
CONS



* Stock cooler spins too fast when idle

* DVI cover interferes with some third party coolers

* No power adapters included

* High price

PowerColor Radeon HD 5850 1GB



Recommended by SPCR

* * *

Articles of Related Interest

Scythe
Musashi Dual Fan GPU Cooler


ATI
Radeon HD 4770: ATI's First 40nm GPU


HIS
Radeon HD 4890 Turbo Edition


Sparkle
GeForce GTS 250 1GB Graphics Card


Asus
EN9400GT Silent Edition


Asus
Radeon HD 4870 Matrix

* * *

Discuss this article in the
SPCR forums.

Sections: 

Google

www SPCR