• Home
  • blog
  • ATI Radeon HD 4770: ATI’s First 40nm GPU

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

Not content with the performance/price edge the Radeon HD 4830 currently enjoys over the GeForce 9800GT, ATI has decided to hammer the $100 graphics market further with the new Radeon HD 4770 released today. Built using a 40nm manufacturing process, ATI positions it as a better performing, more energy efficient replacement for the HD 4830.

Apr. 28, 2009 by Lawrence Lee

Product
ATI Radeon HD 4770 512MB
PCI-E Graphics Card
Manufacturer
MSRP
~US$99

As the underdog in the chip wars, AMD has traditionally targeted
users with products that offer good relative value, rather than maximum performance.
A prime current example is their fastest desktop processor, the Phenom II. Their current
flagship CPU, the Phenom II 955 Black
Edition
falls a bit short of Intel’s fastest and and more costly Core 2 or
Core i7 chips, but it was well received by the most of the tech web press
due to AMD’s aggressive pricing. Their graphics division, ATI, has been following
a similar path, offering great value cards like the energy efficient Radeon
HD 4670
and HD 4830 for gamers with tighter pockets. In these troubled financial times, it’s could well turn out to be a very effective strategy.

Currently, consumers looking for a budget gaming card for about $100 have ATI’s
Radeon HD 4830 and nVidia’s
GeForce 9800 GT to
choose between. They are neck-in-neck in terms of performance, but as of date,
9800 GT’s are retailing for about $10 more. To hammer this position further,
ATI has released its first GPU built with a 40 nanometer manufacturing process,
the Radeon HD 4770. ATI says it is a better performing, more energy efficient
replacement for the HD 4830.


The Radeon HD 4770.

ATI rated the HD 4830 as a 110W card and employed a single-slot cooler. Strangely,
the new HD 4770, despite having a 80W thermal envelope, has a dual-slot reference
heatsink. Both cards require a 6-pin PCI-E power cable.

Tale Of The Tape: HD 4770 vs. HD 4830
Model HD 4770 HD 4830
GPU Core RV740 RV770LE
Manufacturing Process 40nm 55nm
Transistor Count 826 million 956 million
Stream Processors 640 640
Core Clock 750 MHz 575 MHz
Memory Clock 800 MHz 900 MHz
Memory 512MB GDDR5 512MB GDDR3
Memory Bandwidth 128-bit
51.2 GB/s
256-bit
57.6 GB/s
Average board power 80W 110W

Compared to the HD 4830, the 4770 has an equal number of shaders units, a higher
clock speed, but smaller transistor count, slower memory and less memory bandwidth. The die-shrink from
55nm to 40nm is no doubt the reason for lower power requirement. With the smaller die, it’s surely cheaper to build.


Technical specifications according to GPU-Z.

PHYSICAL DETAILS

The HD 4770 has a familiar cooler, but the board layout has been altered slightly.


The reference HD 4770 has a PCB length of 20.3 cm and is equipped with
a dual-slot heatpipe cooler. Unlike most of the HD 4000 series, a separate
ramsink module is used to cool the memory chips.

 


From the outside the heatsink looks more or less like the one used on
the HD 3870/4870/4890, though the sticker art is more stylized. The card’s
backpane includes a TV-out and dual DVI ports.

 


A strange black metal cover runs along the length of the top side of the
card secured by screws through the PCB and the back panel

 


Installed in our testbed.

THE COOLER

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.


Like most ATI cards the primary cooler is secured to the card with a backplate
and four spring loaded screws. The mounting holes form a 4.8 cm square,
half a cm less than the HD 4850/4870/4890.

 


The memory heatsink and black metal side-cover are a fused together in
a separate unit. It is likely this would need to be removed if a third
party cooling solution were to be used.

 


The VRM circuitry do not have heatsinks of their own but the MOSFETs are
positioned in such a way take advantage of any tertiary cooling provided
by the fan directly above.

 


Unlike the HD 4870 and 4890, the 4770’s heatsink is immobile, being fixed
to the plastic housing. The main fin mass is connected to two 6mm thick
heatpipes soldered to the base.

TEST METHODOLOGY

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

1. Determine whether the card’s 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, bluescreen or reboot without warning.
  • 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 life span 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.

3. Determine the card’s ability to play back high definition video, to see
if whether it is a suitable choice for a home theater PC.

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 with CPUBurn and ATITool’s artifact scanner (or FurMark — whichever
produces higher power consumption) running to stress both the CPU and GPU simultaneously.
This last state mimics the stress on the CPU and GPU produced by a modern video
game. The software is left running until the GPU temperature remains stable
for at least 10 minutes. If artifacts are detected by ATITool 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 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.

Power consumption is also measured during playback of a variety of video clips
with PowerDVD to test the efficiency of the card’s H.264/VC-1 hardware acceleration.

Video Test Suite


1080p | 24fps | ~10mbps
H.264:
Rush Hour 3 Trailer 1
is a H.264 encoded clip 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.

 


1080p | 24fps | ~19mbps
VC-1: Drag Race is a recording of a scene from
network television re-encoded with TMPGEnc using the WVC1 codec, a more
demanding VC-1 codec.

 

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
73W
Unknown
CPUBurn
144W
115W
Ambient temperature: 21°C

 

ATI Radeon HD 4770 512MB:

VGA Test Bed: ATI Radeon HD 4770 512MB
System State
Fan Speed
SPL
@1m
GPU
Temp
Mem. Temp
Shader Temp
System Power
AC
DC (Est.)
Idle
1240
RPM
15 dBA
53°C
53°C
51°C
103W
79W
CPUBurn
55°C
55°C
53°C
177W
143W
CPUBurn + ATITool
1560
RPM
17 dBA
69°C
72°C
69°C
202W
165W
CPUBurn + FurMark
1720
RPM
17 dBA
73°C
74°C
72°C
214W
175W
Temperatures recorded via GPU-Z.
Ambient temperature: 21°C
Ambient noise level: 11 dBA
System noise level: 12 dBA

During testing, temperatures stayed well within reasonable levels — at
full load, the GPU core temperature measured only 73°C. The relatively low
temperatures are due to decreased power with the 40nm die and the card’s dual slot cooler, which
is a bit of overkill for a card rated for 80W TDP. It could probably cool
the GPU adequately with 1000 RPM or less.

System Noise Level Comparison
Card
Idle
Load
ATI HD 4670 512MB
13 dBA
16 dBA
ATI HD 4770 512MB
15 dBA
17 dBA
ATI HD 4830 512MB
14 dBA
21 dBA
ATI HD 4870 1GB
13 dBA
20 dBA
Ambient noise level: 11 dBA
System noise level: 12 dBA

With the 4770’s fan idling at 1240 RPM, the system noise level measured 15 dBA,
slightly higher than most of ATI’s recent offerings. The noise
was fairly unobtrusive with the motor’s grind being muffled by our case’s
side panel to the point where only a low-pitched hum was audible. The fan speed
began to increase once the GPU heated up to about 65°C, finally settling
at 1720 RPM at full load. The noise level at this point was only 17 dBA —
a significant improvement over the HD
4830
. It can be easily masked by the music and sound effects from your
average PC game. We are also happy to report our HD 4770 did not have even a hint
of coil whine or cause such whine in our test system’s power supply or motherboard.

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.

Power Consumption Comparison (DC)
Card
Est. Power (Idle)
Est. Power (ATITool)
Est. Power (FurMark)
Asus EN9400GT 512MB
11W
N/A
23W
PowerColor HD4650 512MB
15W
N/A
28W
ATI HD 4670 512MB
3W
50W
60W
ATI HD 4770 512MB
28W
50W
60W
ATI HD 4830 512MB*
18W
80W
87W
Asus EN9800GT 512MB
32W
72W
80W
Diamond HD 4850 512MB
50W
101W
N/A
* sample with unknown number of stream processors

Our power consumption measurements yielded mixed results. Our HD 4770 sample used
considerably more power in idle than the HD 4670 and 4830 we tested, but considerably less power than the 4850.
(See notes in the conclusions for more thoughts about the idle power.) With full 3D load, however. it drew an impressively modest 60W, 27W lower than our 4830
sample. If the card’s performance is anywhere near that of the 4830, this is
an impressive feat.

System Power Consumption (AC) During Video Playback
Card
Rush Hour
(H.264)
Coral Reef
(WMV-HD)
Drag Race
(VC-1)
Asus EN9400GT 512MB
112W
110W
137W
PowerColor HD4650 512MB
102W
115W
137W
ATI HD 4670 512MB
94W
105W
129W
ATI HD 4770 512MB
132W
129W
148W
ATI HD 4830 512MB*
108W
134W
160W
Asus EN9800GT 512MB
119W
127W
141W
Diamond HD 4850 512MB
136W
151W
176W
* sample with unknown number of stream processors

Due to the HD 4770’s high idle power consumption, video playback was not very
power efficient with our system using between 129W and 148W during playback. Only the HD 4850 drew more power on average.

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.


ATI HD 4870 1GB: BIOS Clock/voltage settings.

According to the BIOS, the card is set to 750/850 MHz and 1.263V during boot
up and downclocks to 500/800 MHz when UVD is active. Displayed in green, yellow
and red are the card’s three clock/power modes during regular use. When idle
and during video playback, our card downclocked to 500/800 MHz according to
GPU-Z.

The BIOS’ fan settings were not accessible.

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.

FINAL THOUGHTS

Gaming: Please check out gaming-oriented reviews of the HD 4770 at xbit labs, techPowerUp,
HardwareCanucks,
and The Tech Report.
Preliminary reports indicate that the performance level is between
that of the HD 4830 and 4850
and easily surpasses the GeForce 9800 GT. If this holds true, the card is easily
the highest performance $100 GPU on the market.

Cooling/Noise: The stock cooler is audible but unobtrusive when idle and
its fan ramps up smoothly. When placed under stress in our testbed, the fan
speed leveled out only a few hundred RPM higher than the idle speed, resulting
in only a small increase in noise. The temperatures we recorded during testing
were excellent. The card’s power draw does not warrant a dual-slot cooler, and
a reduction in fan speed would probably have little impact on maximum load temperature.

Power Consumption: By our estimates, the ATI Radeon HD 4770 requires
about 28W idle and up to 60W when stressed to the limit. The idle number is
10W more than the HD 4830, while the load figure is 27W less. All things equal,
we prefer lower idle power consumption as most systems, especially ones equipped
with low/mid-tier graphics cards sit idle most of the time. However, xbit labs, whose video card power measurements are usually quite consistent and trustworthy, found their HD 4770 idling >10W lower than their 4830, so we are a bit puzzled at this point. We’ll try more testing, perhaps with another sample, and report back on this again.

As a 60W card, the GPU runs fairly cool and thus the stock fan barely needs
to exert any real effort. The cooler is quiet enough
not to bother most users, especially during load. Some with sensitive
hearing will probably still want to opt for a third party solution. Assuming the HD 4770 performs
as expected, we thank ATI for bringing unparalleled value to budget PC gamers.
The HD 4830 was already an excellent
bang-for-your-buck card, and to replace it with a card that is faster and has
a lower thermal envelope without increasing the price is an admirable feat.

The only negative point is our sample card’s poor idle power consumption which can’t
be overlooked. It makes little sense for a card to draw almost half the amount
of power it does when placed under load, which suggests that ATI’s power management
feature, PowerPlay, just isn’t working properly on this sample. Despite this, it still drew
less power than the GeForce
9800 GT
, the HD 4770’s chief competitor.

ATI Radeon HD 4770 512MB
PROS

* Fairly quiet
* Well cooled
* Low power consumption during load

CONS

* High idle power consumption

Our thanks to ATI
for the video card sample.

* * *

Articles of Related Interest
HIS
Radeon HD 4890 Turbo Edition

Sparkle
GeForce GTS 250 1GB Graphics Card

Asus
EN9400GT Silent Edition

Asus
Radeon HD 4870 Matrix

Asus
EN9800GT Matrix Edition

Filling
the Gap: ATI Radeon HD 4830

* * *

Discuss this article in the
SPCR forums.

Leave a Comment

Your email address will not be published. Required fields are marked *