Hubbry Logo
Radeon HD 2000 seriesRadeon HD 2000 seriesMain
Open search
Radeon HD 2000 series
Community hub
Radeon HD 2000 series
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Radeon HD 2000 series
Radeon HD 2000 series
Radeon HD 2000 series
View on Wikipedia
from Wikipedia
ATI Radeon HD 2000 series
Release dateJune 28, 2007; 18 years ago (2007-06-28)
CodenameRadeon R600 series
ArchitectureTeraScale 1[citation needed]
Transistors180M 65nm (RV610)
  • 390M 65nm (RV630)
  • 700M 80nm (R600)
Cards
Entry-level2350, 2400
Mid-range2600
High-end2900
API support
Direct3DDirect3D 10.0[3]
Shader Model 4.0
OpenCLClose To Metal
OpenGLOpenGL 3.3[1][2]
History
PredecessorRadeon X1000 series
SuccessorRadeon HD 3000 series
Support status
Unsupported

The graphics processing unit (GPU) codenamed Radeon R600 is the foundation of the Radeon HD 2000 series and the FireGL 2007 series video cards developed by ATI Technologies. The HD 2000 cards competed with nVidia's GeForce 8 series.

Architecture

[edit]

This article is about all products under the brand "Radeon HD 2000 Series". They all contain a GPU which implements TeraScale 1, ATI's first Unified shader model microarchitecture for PCs.

Video acceleration

[edit]

The Unified Video Decoder (UVD) SIP core is on-die in the HD 2400 and the HD 2600. The HD 2900 GPU dies do not have a UVD core, as its stream processors were powerful enough to handle most of the steps of video acceleration in its stead except for entropy decoding and bitstream processing which are left for the CPU to perform.[4]

Other features

[edit]

HDTV encoding support is implemented via the integrated AMD Xilleon encoder; the companion Rage Theater chip used on the Radeon X1000 series was replaced with the digital Theater 200 chip, providing VIVO capabilities.

For display outputs, all variants include two dual-link TMDS transmitters, except for HD 2400 and HD 3400, which include one single and one dual-link TMDS transmitters. Each DVI output includes dual-link HDCP encoder with on-chip decipher key. HDMI was introduced, supporting display resolutions up to 1,920×1,080, with integrated HD audio controller with 5.1-channel LPCM and AC3 encoding support. Audio is transmitted via DVI port, with specially designed DVI-to-HDMI dongle for HDMI output that carries both audio and video.[5]

All variants support CrossFireX technology. CrossFire efficiency was improved and shows performance approaching the theoretical maximum of twice the performance of a single card.[6][7]

Desktop products

[edit]
Main article: Radeon R600 (HD 2xxx, HD 3xxx) Series

The R600 family is called the Radeon HD 2000 series, with the enthusiast segment being the Radeon HD 2900 series which originally comprised the Radeon HD 2900 XT with GDDR3 memory released on May 14, and the higher-clocked GDDR4 version in early July.

The mainstream and budget segment products were the Radeon HD 2600 and Radeon HD 2400 series respectively, both launched June 28, 2007.[8]

Previously there were no HD 2000 series products being offered in the performance segment while ATI used models from the previous generation to address that target market; this situation did not change until the release of variants of the Radeon HD 2900 series, the Radeon HD 2900 Pro and GT, which filled the gap of the performance market for a short period of time.

Radeon HD 2400

[edit]
ATI Radeon HD 2400 XT

The Radeon HD 2400 series was based on the codenamed RV610 GPU. It had 180 million transistors on a 65 nm fabrication process. The Radeon HD 2400 series used a 64-bit-wide memory bus.[9] The die size is 85 mm2.[10] The official PCB design implements only a passive-cooling heatsink instead of a fan, and official claims of power consumption are as little as 35 W.[citation needed] The core has 16 kiB unified vertex/texture cache away from dedicated vertex cache and L1/L2 texture cache used in higher end model.

Reports has that the first batch of the RV610 core (silicon revision A12), only being released to system builders, has a bug that hindered the UVD from working properly, but other parts of the die operated normally. Those products were officially supported with the release of Catalyst 7.10 driver, which the cards were named as Radeon HD 2350 series.[11]

Several reports from owners of HD 2400 Pro suggest the card do not fully support hardware decoding for all H.264/VC-1 videos. The device driver, even with the latest stable version, seem to only honor hardware decoding for formats specified in the Blu-ray and HD-DVD specification. As a result of such restriction, the card is not deemed very useful for hardware video decoding since the majority of the H.264/VC-1 videos on the net are not encoded in those formats (even though the hardware itself is fully capable of doing such decoding work). This device driver restriction has led to the development of a third party driver patch, "ExDeus ATI HD Registry Tweak", to unlock the potential of HD 2400 Pro for full support of H.264/VC-1 hardware video decoding.[12][13][14]

Radeon HD 2600

[edit]

The Radeon HD 2600 series was based on the codenamed RV630 GPU and packed 390 million transistors on a 65 nm fabrication process. The Radeon HD 2600-series video cards included GDDR3 support, a 128-bit memory ring bus and 4-phase digital PWM,[9] spanning a die size of 153 mm2.[15] Neither of the GDDR3 reference PCI-E designs required additional power connectors whereas the HD 2600 Pro and XT AGP variants required additional power through either 4-pin or 6-pin power connectors,[16] Official claims state that the Radeon HD 2600 series consumes as little as 45 W of power.[citation needed]

Radeon HD 2600 X2

[edit]

The Radeon HD 2600 X2 is a dual-GPU product which includes 2 RV630 dies on a single PCB with a PCI-E bridge splitting the PCI-E ×16 bandwidth into two groups of PCI-E ×8 lanes (each 2.0 Gbit/s). The card provides 4 DVI outputs or HDMI outputs via dongle and supports CrossFire configurations. AMD calls this product the Radeon HD 2600 X2 as seen by some vendors and as observed inside the INF file of Catalyst 7.9 version 8.411. Sapphire and other vendors including PowerColor and GeCube have either announced or demonstrated their respective dual GPU (connected by crossfire) products.[17] Catalyst 7.9 added support for this hardware in September 2007. However, AMD did not provide much publicity to promote it. A vendor may offer cards containing 256 MiB, 512 MiB, or 1 GiB of video memory. Although the memory technology utilized is at a vendor's discretion, most vendors have opted for GDDR3 and DDR2 due to lower manufacturing cost and positioning of this product for the mainstream rather than performance market segment and also a big success.

Radeon HD 2900

[edit]

The Radeon HD 2900 series was based on the codenamed R600 GPU and was launched on May 14, 2007. R600 packed 700 million transistors on an 80 nm fabrication process and had a 420 mm2 die size.[18] The Radeon HD 2900 XT was launched with 320 Stream Processors and a core clock of 743 MHz. The initial model was released with 512 MB of GDDR3 clocked at 828 MHz (1,656 MHz effective) with a 512-bit interface. A couple months after release ATI released the 1 GB GDDR4 model with a memory frequency of 1,000 MHz (2,000 MHz effective). Performance was on par compared to the 512 MB card. The HD 2900 XT introduced a lot of firsts. It was the first to implement a digital PWM on board (7-phase PWM), first to use an 8-pin PEG connector, and was the first graphics card from ATI to support DirectX 10.

The Radeon HD 2900 Pro was clocked lower at 600 MHz core and 800 MHz memory (1,600 MHz effective), configured with 512 MB of GDDR3 or 1 GB of GDDR4. It was rumored that some of the 1 GB GDDR4 models were manufactured using a 12" cooler borrowed from the prototype HD 2900 XTX.[19] The HD 2900 Pro had both 256-bit and 512-bit interface options for the 512 MB versions of the card. A few AIB partners offered a black and silver cooler exclusive to the 256-bit model of the Pro.[20][21]

The Radeon HD 2900 GT was a 240 Stream Processor variant clocked the same as the HD 2900 Pro, but with 256 MB of video memory on a 256-bit interface.

Mobile products

[edit]
Main article: List of R600 GPUs

All Mobility Radeon HD 2000 series share the same feature set support as their desktop counterparts, as well as the addition of the battery-conserving PowerPlay 7.0 features, which are augmented from the previous generation's PowerPlay 6.0.

The Mobility Radeon HD 2300 is a budget product which includes UVD in silica but lacks unified shader architecture and DirectX 10.0/SM 4.0 support, limiting support to DirectX 9.0c/SM 3.0 using the more traditional architecture of the previous generation. A high-end variant, the Mobility Radeon HD 2700, with higher core and memory frequencies than the Mobility Radeon HD 2600, was released in mid-December 2007.

The Mobility Radeon HD 2400 is offered in two model variants; the standard HD 2400 and the HD 2400 XT.[22]

The Mobility Radeon HD 2600 is also available in the same two flavors; the plain HD 2600 and, at the top of the mobility lineup, the HD 2600 XT.[23]

The half-generation update treatment had also applied to mobile products. Announced prior to CES 2008 was the Mobility Radeon HD 3000 series. Released in the first quarter of 2008, the Mobility Radeon HD 3000 series consisted of two families, the Mobility Radeon HD 3400 series and the Mobility Radeon HD 3600 series. The Mobility Radeon HD 3600 series also featured the industry's first implementation of on-board 128-bit GDDR4 memory.

About the time of late March to early April, 2008, AMD renewed the device ID list on its website[24] with the inclusion of Mobility Radeon HD 3850 X2 and Mobility Radeon HD 3870 X2 and their respective device IDs. Later in Spring IDF 2008 held in Shanghai, a development board of the Mobility Radeon HD 3870 X2 was demonstrated alongside a Centrino 2 platform demonstration system.[25] The Mobility Radeon HD 3870 X2 was based on two M88 GPUs with the addition of a PCI Express switch chip on a single PCB. The development board used for demonstration was a PCI Express 2.0 ×16 card, while the final product is expected to be on AXIOM/MXM modules.

Chipset table

[edit]
Main article: TeraScale (microarchitecture) § TeraScale 1
Model Launch
Code name
Fab (nm)
Transistors (million)
Die size (mm2)
Bus interface
 Clock rate
Core config
 Fillrate Memory Processing power
(GFLOPS)
TDP (Watts)
Crossfire support
 API support (version)
Release price (USD)
Core (MHz)
Memory (MHz)
Pixel (GP/s)
Texture (GT/s)
Size (MB)
Bandwidth (GB/s)
Bus type
Bus width (Bit)
Max.
Direct3D
Radeon HD 2350 June 28, 2007 RV610 65 180 85 PCIe 1.0 ×16 525 400 40:4:4 2.10 2.10 64 onboard + up to 256 system 3.20 DDR2 32 42.0 — 20 No 10.0 3.3 APP Stream Only ?
Radeon HD 2400 PRO PCIe 1.0 ×16
AGP
PCI 		128
256
512 		6.40 64 $50–55
Radeon HD 2400 XT PCIe 1.0 ×16 650 500
700 		2.60 2.60 256 8.0
11.2 		DDR2
GDDR3 		52.0 25 $75–85
Radeon HD 2600 PRO RV630 390 153 PCIe 1.0 ×16
AGP 		600 120:8:4 2.40 4.80 256
512 		16.0
22.4 		128 144.0 35 $89–99
Radeon HD 2600 XT 800 800
1100 		3.20 6.40 25.6
35.2 		GDDR3
GDDR4 		192.0 45
50 		4-way CrossFire $119 (GDDR3) $149 (GDDR4)
Radeon HD 2900 GT November 6, 2007 R600 GT 80 720 420 PCIe 1.0 ×16 601 800 240:12:12 7.21 7.21 51.2 GDDR3 256 288.5 150 $200
Radeon HD 2900 PRO September 25, 2007 R600 PRO 600 800
925 		320:16:16 9.6 9.6 512
1024 		51.2
102.4
118.4 		GDDR3
GDDR4 		256
512 		384.0 200 $250 (GDDR3)
$300 (GDDR4)
Radeon HD 2900 XT May 14, 2007 R600 XT 743 828
1000 		11.9 11.9 105.6
128.0 		GDDR3
GDDR4 		512 475.5 215 $399

Mobility Radeon

[edit]

OpenGL 3.3 is possible with latest drivers for all RV6xx.

Model Launch
Model number
Code name
Fab (nm)
Core clock (MHz)
Memory clock (MHz)
Core config
 Memory
API compliance (version)
Processing power
(GFLOPS)
Notes
Pixel (GP/s)
Texture (GT/s)
Size (MB)
Bandwidth (GB/s)
Bus type
Bus width (bit)
Mobility Radeon X2300 March 1, 2007 M64 RV515 90 PCIe ×16 480 400 2:4:4:41 1.92 1.92 128 6.4
12.8 		DDR
DDR2
GDDR3 		64
128 		9.0c 2.0 Unknown renamed product, HyperMemory, no UVD, PowerPlay 6.0
Mobility Radeon X2500 June 1, 2007 M66 RV530 460 5:12:4:41 1.84 1.84 256 12.8 128 Unknown based on X1600/1700, HM up to 768 Mb, no UVD, PowerPlay 6.0
Mobility Radeon HD 2300 March 1, 2007 M71 RV515 480 2:4:4:41 1.92 1.92 128
256
512 		6.4
12.8 		64
128 		Unknown same as X2300, but with UVD, PowerPlay 6.0
Mobility Radeon HD 2400 May 14, 2007 M72S RV610 65 450 40:4:42 1.8 1.8 256+ Hyper Memory 6.4 DDR2 64 10.0 2.0 (3.3) 36 UVD, PowerPlay 7.0
Mobility Radeon HD 2400 XT M72M 600
600 		400
700 		2.4 2.4 6.4
11.2 		DDR2
GDDR3 		48
Mobility Radeon HD 2600 M76M RV630 500
500 		400
600 		120:8:42 2.0 4.0 12.8
19.2 		128 120
Mobility Radeon HD 2600 XT M76XT 680 750 2.72 5.44 24 GDDR3 168
Mobility Radeon HD 2700 December 12, 2007 M76 650 700 2.6 5.2 256+ Hyper Memory (total 768) 22.4

1 Vertex shaders : Pixel shaders : Texture mapping units : Render output units.
2 Unified Shaderss : Texture mapping units : Render output units

Radeon feature matrix

[edit]

The following table shows features of AMD/ATI's GPUs (see also: List of AMD graphics processing units).

[ VisualEditor ]
Name of GPU series Wonder Mach 3D Rage Rage Pro Rage 128 R100 R200 R300 R400 R500 R600 RV670 R700 Evergreen Northern
Islands 		Southern
Islands 		Sea
Islands 		Volcanic
Islands 		Arctic
Islands/Polaris 		Vega Navi 1x Navi 2x Navi 3x Navi 4x
Released 1986 1991 Apr
1996 		Mar
1997 		Aug
1998 		Apr
2000 		Aug
2001 		Sep
2002 		May
2004 		Oct
2005 		May
2007 		Nov
2007 		Jun
2008 		Sep
2009 		Oct
2010 		Dec
2010 		Jan
2012 		Sep
2013 		Jun
2015 		Jun 2016, Apr 2017, Aug 2019 Jun 2017, Feb 2019 Jul
2019 		Nov
2020 		Dec
2022 		Feb
2025
Marketing Name Wonder Mach 3D
Rage 		Rage
Pro 		Rage
128 		Radeon
7000 		Radeon
8000 		Radeon
9000 		Radeon
X700/X800 		Radeon
X1000 		Radeon
HD 2000 		Radeon
HD 3000 		Radeon
HD 4000 		Radeon
HD 5000 		Radeon
HD 6000 		Radeon
HD 7000 		Radeon
200 		Radeon
300 		Radeon
400/500/600 		Radeon
RX Vega, Radeon VII 		Radeon
RX 5000 		Radeon
RX 6000 		Radeon
RX 7000 		Radeon
RX 9000
AMD support Ended Current
Kind 2D 3D
Instruction set architecture Not publicly known TeraScale instruction set GCN instruction set RDNA instruction set
Microarchitecture Not publicly known GFX1 GFX2 TeraScale 1
(VLIW5)
(GFX3) 		TeraScale 2
(VLIW5)
(GFX4) 		TeraScale 2
(VLIW5)
up to 68xx
(GFX4) 		TeraScale 3
(VLIW4)
in 69xx [26][27]
(GFX5) 		GCN 1st
gen
(GFX6) 		GCN 2nd
gen
(GFX7) 		GCN 3rd
gen
(GFX8) 		GCN 4th
gen
(GFX8) 		GCN 5th
gen
(GFX9) 		RDNA
(GFX10.1) 		RDNA 2
(GFX10.3) 		RDNA 3
(GFX11) 		RDNA 4
(GFX12)
Type Fixed pipeline[a] Programmable pixel & vertex pipelines Unified shader model
Direct3D — 5.0 6.0 7.0 8.1 9.0
11 (9_2) 		9.0b
11 (9_2) 		9.0c
11 (9_3) 		10.0
11 (10_0) 		10.1
11 (10_1) 		11 (11_0) 11 (11_1)
12 (11_1) 		11 (12_0)
12 (12_0) 		11 (12_1)
12 (12_1) 		11 (12_1)
12 (12_2)
Shader model — 1.4 2.0+ 2.0b 3.0 4.0 4.1 5.0 5.1 5.1
6.5 		6.7 6.8
OpenGL — 1.1 1.2 1.3 1.5[b][28] 3.3 4.6[29][c]
Vulkan — 1.1[c][d] 1.3[30][e] 1.4[31]
OpenCL — Close to Metal 1.1 (not supported by Mesa) 1.2+ (on Linux: 1.1+ (no Image support on Clover, with by Rusticl) with Mesa, 1.2+ on GCN 1.Gen) 2.0+ (Adrenalin driver on Win7+)
(on Linux ROCm, Mesa 1.2+ (no Image support in Clover, but in Rusticl with Mesa, 2.0+ and 3.0 with AMD drivers or AMD ROCm), 5th gen: 2.2 win 10+ and Linux RocM 5.0+ 		2.2+ and 3.0 Windows 8.1+ and Linux ROCm 5.0+ (Mesa Rusticl 1.2+ and 3.0 (2.1+ and 2.2+ wip))[32][33][34]
HSA / ROCm — Yes ?
Video decoding ASIC — Avivo/UVD UVD+ UVD 2 UVD 2.2 UVD 3 UVD 4 UVD 4.2 UVD 5.0 or 6.0 UVD 6.3 UVD 7 [35][f] VCN 2.0 [35][f] VCN 3.0 [36] VCN 4.0 VCN 5.0
Video encoding ASIC — VCE 1.0 VCE 2.0 VCE 3.0 or 3.1 VCE 3.4 VCE 4.0 [35][f]
Fluid Motion [g] No Yes No ?
Power saving ? PowerPlay PowerTune PowerTune & ZeroCore Power ?
TrueAudio — Via dedicated DSP Via shaders
FreeSync — 1
2
HDCP[h] — ? 1.4 2.2 2.3 [37]
PlayReady[h] — 3.0 No 3.0
Supported displays[i] 1–2 2 2–6 ? 4
Max. resolution ? 2–6 ×
2560×1600 		2–6 ×
4096×2160 @ 30 Hz 		2–6 ×
5120×2880 @ 60 Hz 		3 ×
7680×4320 @ 60 Hz [38]
7680×4320 @ 60 Hz PowerColor 		7680x4320

@165 Hz

7680x4320
/drm/radeon[j] Yes —
/drm/amdgpu[j] — Optional [39] Yes
  1. ^ The Radeon 100 Series has programmable pixel shaders, but do not fully comply with DirectX 8 or Pixel Shader 1.0. See article on R100's pixel shaders.
  2. ^ R300, R400 and R500 based cards do not fully comply with OpenGL 2+ as the hardware does not support all types of non-power of two (NPOT) textures.
  3. ^ a b OpenGL 4+ compliance requires supporting FP64 shaders and these are emulated on some TeraScale chips using 32-bit hardware.
  4. ^ Vulkan support is theoretically possible but has not been implemented in a stable driver.
  5. ^ Vulkan support in Linux relies on the amdgpu kernel driver which is incomplete and not enabled by default for GFX6 and GFX7.
  6. ^ a b c The UVD and VCE were replaced by the Video Core Next (VCN) ASIC in the Raven Ridge APU implementation of Vega.
  7. ^ Video processing for video frame rate interpolation technique. In Windows it works as a DirectShow filter in your player. In Linux, there is no support on the part of drivers and / or community.
  8. ^ a b To play protected video content, it also requires card, operating system, driver, and application support. A compatible HDCP display is also needed for this. HDCP is mandatory for the output of certain audio formats, placing additional constraints on the multimedia setup.
  9. ^ More displays may be supported with native DisplayPort connections, or splitting the maximum resolution between multiple monitors with active converters.
  10. ^ a b DRM (Direct Rendering Manager) is a component of the Linux kernel. AMDgpu is the Linux kernel module. Support in this table refers to the most current version.

Graphics device drivers

[edit]

AMD's proprietary graphics device driver "Catalyst"

[edit]
Main article: AMD Catalyst

AMD Catalyst is being developed for Microsoft Windows and Linux. As of July 2014, other operating system are not officially supported. This may be different for the AMD FirePro brand, which is based on identical hardware but features OpenGL-certified graphics device drivers.

AMD Catalyst supports of course all features advertised for the Radeon brand.

Microsoft Windows

[edit]

The Purple Pill tool issue, which could allow unsigned drivers to be loaded into Windows Vista and tamper with the operating system kernel,[40] was resolved in the Catalyst 7.8 release (version 8.401).[41] The AVIVO video converter for Windows Vista, and color temperature control in Catalyst Control Center was added with the release of Catalyst 7.9, package version 8.411. Software CrossFire was enabled for HD 2600 and HD 2400 series video cards with the release of Catalyst 7.10 (package version 8.421)

The Catalyst 8.1, package version 8.451, supports for MultiView technology for accelerated OpenGL rendering on multiple video card setup (CrossFire). The driver also allows CrossFire configurations for Radeon HD 3850 and HD 3870 video cards.[42][43]

The Catalyst 8.3 is described by AMD as a milestone release,[44] supporting DirectX 10.1, ATI CrossFire X technology and allowing the mixing of different Radeon HD 3800 series video cards to form a CrossFire X setup with 2 to 4 GPUs. Catalyst 8.3 introduced to new video controls to further enhance the video playback quality, these controls includes edge enhancement and noise reduction settings. There is also the support for extended desktop in CrossFire X mode. The anti-aliasing support for Unreal Engine 3.0 in DirectX 9.0 games, support for CFAA filters (wide tent and box tent) to be enabled when Super AA is enabled, and other features as developer support for hardware surface tessellation, hardware accelerated wide aspect ratio LCD scaling, HydraVision support for Windows Vista allowing to add maximum 9 virtual desktops and new Folding@home client are also officially supported in this release.

The Catalyst 8.5, package version 8.493[45] brought new features include component video with 480i and 480p resolutions, SECAM TV output support, 1080p HDTV custom mode via HDMI, 1080p24 (1080p resolution at 24 Hz) support, HDMI Audio for non-standard TV modes (CEA 861b), support for adaptive anti-aliasing (and later, in Catalyst 8.6, also support for custom filters[46]) under OpenGL, Windows XP SP3 support and un-install utility enhancements. The driver also includes performance improvements and fixes some instability issues and rendering issues on some games.

The Radeon HD 2000 series has been transitioned to legacy support, where drivers will be updated only to fix bugs instead of being optimized for new applications.[47]

Current Catalyst drivers do not support the AGP versions of Radeon HD 2000/3000 series cards with RIALTO bridge. Installing Catalyst drivers on those cards will yield the following error message: "setup did not find a driver compatible with your current hardware or operating system." or simply fail outright. The AGP cards in question are supported unofficially by ATI/AMD with a hot-fixed Catalyst driver-set each month since May 2008 with the Catalyst 8.5 hotfix.[48] Their PCI vendor IDs are listed below:[49]

GPU core Product PCI device ID
RV610 Radeon HD 2400 Pro 94C4
RV630 Radeon HD 2600 Pro 9587
RV630 Radeon HD 2600 XT 9586

Free and open-source graphics device driver "Radeon"

[edit]
Main article: free and open-source "radeon" graphics device driver

The free and open-source drivers are primarily developed on Linux and for Linux, but have been ported to other operating systems as well. Each driver is composed out of five parts:

  1. Linux kernel component Direct Rendering Manager (DRM)
  2. Linux kernel component KMS driver: basically the device driver for the display controller
  3. user-space component libDRM
  4. user-space component in Mesa 3D;
  5. a special and distinct 2D graphics device driver for X.Org Server, which if finally about to be replaced by Glamor

The free and open-source "Radeon" graphics driver supports most of the features implemented into the Radeon line of GPUs.[50]

Documentation release

[edit]

The free and open-source "Radeon" graphics device drivers are not reverse engineered, but based on documentation released by AMD.[51]

Initial register documentation and parser code to execute the AtomBIOS ROM routines were released in September 2007. The R600 family Instruction Set Architecture guide was released on June 11, 2008.[52] Sample code and register headers for the R600 and R700 3D engines were released in December 2008. AMD released the specifications for both the r6xx and r7xx families on January 26, 2009.[53]

See also

[edit]

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Radeon HD 2000 series

View on Grokipedia
from Grokipedia
The Radeon HD 2000 series is a family of graphics processing units (GPUs) developed by and launched by on May 14, 2007, as the company's first 10-compliant product line featuring a unified . This series marked a significant shift from previous ATI generations by introducing a modular design with stream processors that could handle both vertex and tasks efficiently, enabling enhanced performance in high-definition gaming and multimedia applications. The lineup targeted both desktop and mobile platforms, spanning from entry-level to high-end segments, and was positioned to compete with NVIDIA's in the emerging 10 ecosystem. Key models in the desktop category included the flagship Radeon HD 2900 XT, built on the R600 chip with 320 stream processors, 512 MB of GDDR3 memory on a 512-bit bus, and an 80 nm manufacturing process; the mid-range Radeon HD 2600 XT and HD 2600 Pro; and the entry-level Radeon HD 2400 XT and HD 2400 Pro, the latter two utilizing 65 nm and 55 nm processes for improved power efficiency. Mobile variants, such as the Mobility Radeon HD 2600, HD 2400, and HD 2300, extended the series to laptops with features like ATI PowerPlay 7 for dynamic and options. Initial pricing ranged from $99 for the HD 2400 series to $199 for the HD 2600 series and $399 for the HD 2900 XT, making HD gaming more accessible while supporting multi-GPU configurations via technology on select models. The architecture introduced several notable technologies, including the second-generation Unified Shader Architecture for Shader Model 4.0 compliance, the (UVD) for hardware-accelerated playback of formats like H.264 and , and Avivo HD for enhanced image quality with support for 128-bit HDR rendering and 1.3 output including digital audio. Additional innovations encompassed a for faster data access, hardware derived from technology for procedural geometry generation, and Custom Filter (CFAA) offering up to 24x sampling rates for superior visual fidelity without excessive performance penalties. Despite its ambitious feature set, the series faced challenges with driver maturity and power efficiency at launch, particularly in the high-end segment, but it laid foundational advancements for subsequent GPU generations.

Overview

Development history

The development of the Radeon HD 2000 series, centered on the R600 (GPU), began in the mid-2000s as sought to counter Nvidia's advancements in graphics architecture, particularly the upcoming G80 GPU with its for 10 support. ATI's prior R500-based relied on separate vertex processing units (VPUs) and pixel shaders, which limited efficiency in handling diverse workloads under emerging APIs like 10. To address this, ATI shifted to a unified shader architecture in the TeraScale design for R600, allowing a single pool of stream processors to handle both vertex and operations, thereby improving utilization and performance for next-generation gaming and compute tasks. In July 2006, AMD announced its acquisition of ATI for $5.4 billion, with the deal finalized in October 2006, integrating ATI's graphics division into AMD's operations and providing additional engineering resources to refine the R600 project amid intensifying competition from Nvidia's G80 launch in November 2006. This merger, however, introduced organizational disruptions that temporarily slowed R600's progress, as teams adjusted to new structures while aiming to deliver a DirectX 10-capable product. A key design pillar was the integration of ATI Avivo HD technology, which enhanced hardware-accelerated video decoding and processing for high-definition content, supporting formats up to 1080p and enabling smoother playback with reduced CPU load. Development faced significant challenges, including repeated delays from initial targets in late 2006 to a launch in mid-2007, primarily due to excessive power consumption and thermal issues in the high-end R600 chip, which featured 710 million transistors on a 420 mm² die fabricated at 80 nm. Engineers addressed these by optimizing the architecture, drawing inspiration from the unified shaders in ATI's earlier Xenos GPU for the Xbox 360, but the complexities of scaling for desktop performance while maintaining efficiency under DirectX 10 requirements extended the timeline. Despite these hurdles, the R600's TeraScale foundation laid the groundwork for ATI's (now AMD's) future graphics innovations, emphasizing balanced compute capabilities over raw pixel throughput.

Release timeline and market reception

The Radeon HD 2000 series was officially introduced by on May 14, 2007, marking the company's first major graphics product line following its acquisition of in October 2006. The lineup featured a staggered release schedule, with the high-end Radeon HD 2900 XT launching immediately on that date, while the mid-range Radeon HD 2600 series and entry-level Radeon HD 2400 series followed on June 28, 2007. Initial suggested retail pricing positioned the series competitively in the 2007 market, with the Radeon HD 2900 XT at $399 USD, the Radeon HD 2600 XT at $169 USD, and the Radeon HD 2400 XT at $99 USD. The series targeted Nvidia's 8 lineup, emphasizing 10 compatibility as a key differentiator in the emerging Windows Vista ecosystem. Market reception was mixed, with praise for the series' pioneering 10 support and integrated (UVD) for hardware-accelerated HD video playback, which enhanced multimedia capabilities across all models. The mid-range HD 2600 series received positive feedback for its value in gaming and efficient power use (around 65W TDP), often outperforming Nvidia's 8600 GTS at similar price points. In contrast, the flagship HD 2900 XT faced significant criticism for its 215W TDP, resulting in high heat output, loud cooling, and up to 50% higher power draw than the competing 8800 GTX, alongside early driver immaturity that caused instability in 10 titles. These issues contributed to lackluster high-end sales, as the card failed to match Nvidia's performance leadership, though the overall lineup strengthened AMD's position in mid-range segments.

Architecture

Core design and TeraScale

The R600 (GPU), serving as the foundation for the high-end models in the Radeon HD 2000 series, was manufactured using TSMC's 80 nm high-speed (HS) process technology, resulting in a die size of 420 mm² and a of 700 million. This large-scale integration marked a significant advancement in GPU density at the time, enabling complex computational capabilities within a single chip. The core incorporates 320 unified shader processors, organized into 16 quads, where each quad handles scalar operations across multiple shader types through a vectorized execution model. The TeraScale microarchitecture underlying the R600 represents ATI's (later AMD's) shift to a fully unified shader design, departing from the discrete fixed-function pipelines of prior generations like the Radeon X1000 series. In this architecture, vertex, pixel, and geometry shaders are processed interchangeably via single instruction, multiple data (SIMD) arithmetic logic units (ALUs), supporting DirectX 10 and Shader Model 4.0 for enhanced programmability and efficiency in rendering diverse graphical workloads. This unification allowed for better resource utilization by dynamically allocating processing power based on application demands, a key innovation that improved scalability over the asymmetric processing units in earlier architectures. The memory subsystem features support for GDDR3 memory with a maximum bus width of 512 bits in flagship configurations, facilitating high-bandwidth access for graphics operations. A ring bus interconnect design links the memory controllers to and render output units (ROPs), providing a balanced data distribution that reduces latency compared to traditional crossbar implementations in previous ATI GPUs. Power and thermal management in the R600 design introduced native multi-GPU support, allowing scalable performance through communication between cards. However, the architecture's aggressive clock speeds and high transistor density led to elevated (TDP) ratings, often exceeding 215 W, which posed challenges for cooling and efficiency in early implementations.

Graphics and compute features

The Radeon HD 2000 series introduced ATI's TeraScale 1 architecture, featuring a that allowed flexible allocation of processing resources across vertex, geometry, and pixel shading tasks, enabling efficient handling of 10 workloads. This unified approach contrasted with prior fixed-function pipelines by treating shaders as general-purpose stream processors capable of scalar operations, with the high-end HD 2900 XT exemplifying the design through its 320 unified shaders organized into 16 quads, each supporting up to five scalar instructions per clock cycle in VLIW packet execution for multiply-add operations. Lower-tier models scaled down accordingly, such as the HD 2600 XT with 120 shaders and the HD 2400 XT with 40 shaders, paired with 8 and 4 units (TMUs), respectively, while maintaining the same per-shader execution efficiency to balance performance and power. Across the series, 16 render output units (ROPs) in the HD 2900 models and 4 ROPs in the others handled final pixel blending and output, contributing to a peak pixel fill rate of up to 11.9 GPixel/s on the flagship at its 742 MHz core clock. The geometry engine in the Radeon HD 2000 series provided robust support for 10's geometry shader stage, including hardware-accelerated to subdivide primitive surfaces into finer meshes for enhanced detail without excessive vertex counts. This unit, programmable via instructions, operated on input post-vertex processing, generating up to a 15x subdivision factor in adaptive modes to optimize complex scenes like or character models. Complementing this, the engine featured programmable vertex fetch capabilities, allowing to dynamically load vertex data from index buffers and attribute streams, reducing CPU overhead and enabling more efficient geometry pipelines compared to 9 limitations. These elements collectively supported up to 500 million triangles per second in geometry processing on the HD 2900 XT, facilitating real-time rendering of intricate 3D environments. Anti-aliasing and enhancements in the series improved image quality by mitigating jagged edges and distortion in rendered scenes. The GPUs supported up to 8x (MSAA) modes, including adaptive and temporal variants that selectively applied samples based on to balance quality and performance, with coverage sampling anti-aliasing (CSAA)-like options for sub-pixel precision in 10 titles. reached 16x levels, applying perspective-correct texture sampling to distant surfaces for sharper details without significant fill-rate penalties, as demonstrated in benchmarks where enabling 16x AF yielded minimal drops under 4x AA. These features were driver-managed through ATI's Catalyst Control Center, allowing users to override application settings for consistent visual fidelity across games. Compute capabilities in the Radeon HD 2000 series marked an early foray into general-purpose GPU (GPGPU) computing via ATI's SDK, which leveraged the unified shaders for parallel processing tasks beyond , such as scientific simulations or data-parallel algorithms using the Brook+ language or Close to Metal (CTM) interface. At launch in 2007, this provided foundational GPGPU potential on R600-based cards like the HD 2900 XT, achieving up to 0.95 TFLOPS in single-precision floating-point operations, though it lagged behind NVIDIA's in ease of development and ecosystem maturity due to proprietary APIs. Full 1.0 support arrived later in 2009 with Stream SDK v2.0, enabling cross-platform compute shaders on HD 2000 hardware, but initial implementations were limited by the architecture's scalar focus and lack of double-precision native hardware.

Video acceleration and display capabilities

The Radeon HD 2000 series introduced ATI Avivo HD, a dedicated hardware subsystem designed to enhance and playback efficiency. At its core is the first-generation (UVD1), an integrated engine that provides full for decoding H.264 (AVC), , and video formats, supporting playback up to resolution without significant CPU overhead. This capability marked the Radeon HD 2000 as the first ATI GPUs to enable smooth, native hardware decoding for Blu-ray and content, allowing for high-quality reproduction of protected high-definition media while minimizing system resource demands. Avivo HD further incorporates specialized engines for video post-processing, including HD upscaling to higher resolutions, noise reduction to minimize artifacts in compressed sources, and color correction for improved vibrancy and accuracy across displays. These features operate independently of the GPU's 3D rendering pipeline, ensuring efficient handling of multimedia tasks even during concurrent graphics workloads. The unified architecture of the series complements this by allowing seamless integration of video decode with compute shaders for enhanced effects, though Avivo HD primarily relies on its dedicated hardware for core acceleration. For display capabilities, the series supports dual-link DVI outputs capable of resolutions up to 2560x1600 at 60 Hz, alongside 1.3 connectivity via included adapters that transmit both video and audio. All outputs include integrated HDCP () support to enable secure playback of encrypted HD content, such as from Blu-ray discs. While lower-end models like the HD 2400 and HD 2600 series stick to DVI and configurations, the flagship HD 2900 series maintains these standards without native , focusing instead on robust analog options like for legacy compatibility.

Desktop products

Radeon HD 2400 series

The Radeon HD 2400 series represents the entry-level desktop graphics cards in AMD's Radeon HD 2000 lineup, powered by the RV610 GPU fabricated on a node and containing 180 million transistors. It incorporates 40 unified shading units, 4 units (TMUs), and 4 render output units (ROPs) based on the TeraScale architecture. Designed for cost-effective systems, these cards emphasize low power consumption and compatibility with 10, making them suitable for budget gaming and (HTPC) builds. Key variants include the Radeon HD 2400 XT, a PCIe 1.0 x16 card with a 650 MHz core clock and 256 MB of GDDR3 memory on a 64-bit bus running at 500 MHz (effective 1000 MT/s). The Radeon HD 2400 PRO targets low-profile chassis with slower specifications, featuring a 525 MHz core clock and 256 MB of DDR2 memory at 400 MHz on the same 64-bit bus. OEM models offered further customization, such as up to 512 MB VRAM configurations using DDR2. Performance positioned the series for light-duty tasks, supporting 10 features but constrained by modest VRAM and bandwidth, typically delivering playable frame rates in 2007-era games at 1024x768 resolution. These GPUs excelled in non-gaming applications like video playback thanks to integrated . Distinguishing features include a 25 W TDP for the XT and 20 W for the Pro, facilitating single-slot, passively cooled designs without external power requirements. Additionally, native support enabled dual-card configurations for scaled performance in multi-GPU setups, similar to SLI but optimized for hardware.

Radeon HD 2600 series

The Radeon HD 2600 series, based on the RV630 graphics processor, represented AMD's mid-range desktop graphics solution in the TeraScale lineup, targeting gamers and multimedia users seeking 10 support without excessive power demands. Fabricated on a node with 390 million transistors, the RV630 featured 120 unified stream processors, 8 units (TMUs), and 4 render output units (ROPs), enabling efficient handling of shader-intensive workloads. The series emphasized power efficiency, with single-GPU models drawing as little as 45 W TDP, a significant improvement over the power-hungry high-end offerings in the lineup. Key variants included the HD 2600 XT, the flagship single-GPU model clocked at 800 MHz with 256 MB of GDDR3 memory on a 128-bit bus running at 700 MHz effective (1400 MT/s), providing a memory bandwidth of 22.4 GB/s. The HD 2600 PRO, primarily targeted at OEM systems, operated at lower clocks around 600 MHz with similar memory configurations but reduced performance for budget builds. A notable innovation was the HD 2600 X2, a dual-GPU card integrating two RV630 cores at 800 MHz each, paired with 512 MB total GDDR3 memory, and a 110 W TDP; it supported on-die for multi-GPU rendering without requiring an external bridge, simplifying setup and boosting frame rates in compatible titles. In terms of performance, the series was optimized for resolutions up to 1280x1024, delivering playable frame rates of 40-60 FPS in contemporary 9 and 10 games such as or at medium settings, though it lagged behind competitors like the 8600 GTS in more demanding scenarios. The X2 variant approximately doubled the output in CrossFire-enabled applications, offering a cost-effective path to higher performance without the complexity of separate cards. Unique to the series was its integration of the Avivo HD engine with (UVD) for hardware-accelerated decoding of formats like H.264 and , enhancing multimedia playback efficiency on a 128-bit interface that balanced cost and bandwidth. The 65 nm shrink contributed to superior thermal and power efficiency compared to the 80 nm flagship models, making it suitable for compact systems.

Radeon HD 2900 series

The Radeon HD 2900 series represented AMD's high-end flagship offering in the HD 2000 lineup, powered by the fabricated on a 80 nm process node with approximately 700 million transistors. The R600 featured 320 unified stream processors, 64 texture mapping units (TMUs), and 16 render output units (ROPs), enabling 10 support and unified for both graphics and compute workloads. This design marked a significant step in AMD's TeraScale , emphasizing high parallelism for rasterization tasks while introducing early support for programmable . The primary variant, the Radeon HD 2900 XT, shipped with 512 MB or 1 GB of GDDR3 across a 512-bit interface—the first such wide bus in a consumer GPU—operating at a core clock of 742 MHz and clock of 1000 MHz (effective 2000 MHz), delivering up to 128 GB/s of bandwidth. It had a (TDP) of 250 W, requiring a single 6-pin and 8-pin power connector, and recommended a minimum W system power supply unit (PSU) to accommodate its demands. A lower-tier Radeon HD 2900 GT variant used a cut-down R600 with 240 stream processors, 48 TMUs, 12 ROPs, 256 MB GDDR3 on a 256-bit bus, reduced clocks of 600 MHz core and 800 MHz (effective 1600 MHz), and a TDP of 150 W for better efficiency in mid-high-end builds. Rumors circulated of a higher-clocked Radeon HD 2900 XTX with an 800 MHz core and enhanced cooling, but ultimately canceled its release amid performance tuning challenges. In performance, the HD 2900 XT delivered competitive rasterization against NVIDIA's 8800 GTX, achieving approximately 50-60 frames per second (FPS) in titles like and at 1600x1200 resolution with 4x , though it trailed in shader-heavy 10 workloads due to architectural inefficiencies. multi-GPU scaling proved strong, often exceeding 90% efficiency in supported games, allowing dual-card setups to surpass single 8800 GTX SLI configurations in bandwidth-limited scenarios. However, the series faced notable drawbacks at launch, including high power draw necessitating 450 W PSUs for stability, excessive heat output leading to throttling under sustained loads, and a noisy reference cooler that reached around 50 dBA. These issues, combined with initial driver instability, contributed to mixed market reception despite the innovative 512-bit memory subsystem.

Mobile and integrated products

Mobility Radeon HD series

The Mobility Radeon HD series adapted the Radeon HD 2000 desktop architecture for laptops, emphasizing power efficiency through process nodes of 80 nm for entry-level models like the HD 2300 and 65 nm for mid-range models like the HD 2400 and HD 2600, along with features like dynamic power management to suit mobile form factors. These discrete GPUs targeted performance-oriented notebooks, delivering DirectX 10 support where applicable and hardware-accelerated video decoding for multimedia tasks. Unlike desktop variants, mobile models prioritized lower thermal output and battery conservation, with no direct equivalent to the power-intensive HD 2900 series due to laptop thermal constraints. Key cores included the entry-level M71 for the HD 2300 (based on RV5xx with 4 pixel and 2 vertex shaders, though some variants used reduced configurations), the mid-range M82 for the HD 2600 (RV630-based with 120 unified shaders), and no high-end mobile counterpart to the R600 core. The HD 2300, often positioned as a budget option, supported 9 with limited shader capabilities, while higher models like the HD 2400 and 2600 embraced the full TeraScale for improved efficiency in graphics workloads. Variants such as the Mobility Radeon HD 2400 XT featured thermal design power (TDP) ratings typically ranging from 5 W to 25 W, paired with 256 MB of DDR2 or GDDR3 memory across a 64-bit interface, enabling integration into slim notebooks from 2007 onward by OEMs like Dell and HP. The Mobility Radeon HD 2600 XT, aimed at higher performance, supported up to 50 W TDP with 256 MB GDDR3 memory, offering better handling of demanding applications within mobile constraints. These GPUs excelled in mobile scenarios like gaming at 1024x768 resolutions and HD video playback, where the (UVD) in models like the HD 2400 and 2600 facilitated efficient Blu-ray decoding on battery-powered systems. Dynamic clocking adjusted core and speeds based on load, extending battery life during light tasks while maintaining playable frame rates in titles like those from the 9 era. ATI PowerPlay 7.0 provided advanced thermal management by optimizing voltage and clock states to reduce heat and power draw, supporting external displays up to 1920x1200 via LVDS or emerging eDP interfaces for dual-monitor setups in productivity workflows. This technology, combined with Avivo HD video decode, ensured smooth multimedia experiences without excessive drain on batteries.

Integrated graphics in chipsets

The AMD RS690 , introduced in 2007 as part of the AMD 600-series, featured the integrated Xpress 1250 graphics processor (IGP), derived from the RV515 core of the earlier R500 architecture but designed for compatibility with HD 2000-series drivers and features. This IGP supported up to 1 GB of shared system memory via HyperMemory technology, allowing dynamic allocation from the host system's RAM for graphics tasks. The RS690 targeted budget desktop systems with AMD processors like and Sempron, providing entry-level graphics suitable for office productivity, web browsing, and basic playback. A variant, the RS600 chipset, integrated the Radeon Xpress 1250 IGP, optimized for Intel processor platforms while maintaining similar architecture and capabilities to the RS690's X1250, including support for shared system memory up to 1 GB. Both IGPs offered partial 9 support with Model 2.0b, relying on software emulation for enhanced features, and included Avivo for improved 2D and HD video decoding without hardware UVD . Display outputs encompassed analog TV-Out, DVI, and with HDCP support for protected content, enabling dual-monitor configurations in budget setups. Performance was limited to light workloads, achieving approximately 10-20 FPS in older games like F.E.A.R. at low resolutions and settings (e.g., 76 FPS at 640x480 low, dropping to 7 FPS at 1024x768 high), with no dedicated shaders and heavy dependence on the host CPU for rendering. These IGPs excelled in non-gaming scenarios, such as Aero effects and standard-definition video playback, but struggled with demanding 3D applications. A key feature was support for configurations combining the integrated IGP with a compatible discrete Radeon card (e.g., HD 2400 series) to balance power efficiency and performance in AMD platforms, with driver-enabled switching for intensive tasks. This marked an early step toward switchable graphics in consumer systems.

Software support

Proprietary drivers

The proprietary drivers for the Radeon HD 2000 series were delivered through AMD's Catalyst software suite, a closed-source package that managed graphics acceleration, display configuration, and additional utilities for compatible operating systems. The suite debuted with version 7.6 in June 2007, providing initial support for the HD 2900 XT and subsequent models like the HD 2600 and HD 2400 series upon their releases later that year. Key components included HydraVision for multi-monitor management, enabling seamless extension and spanning across up to six displays, and OverDrive for user-controlled overclocking of engine and memory clocks on supported cards. Support continued through legacy branches until version 13.9, released in April 2013, after which no further updates were issued for the series. On Windows platforms, Catalyst drivers enabled full DirectX 10 compatibility, allowing the HD 2000 series to run feature-complete shaders and tessellation effects from that era's applications. OpenGL support reached version 3.3 in later releases such as Catalyst 10.2., facilitating professional workloads and games requiring advanced geometry processing. Hardware-accelerated HD video decoding via the Unified Video Decoder (UVD) was optimized through version 13.9, supporting formats like H.264 and VC-1 up to 1080p resolutions. Full driver maintenance ended with Windows 8 compatibility in the legacy branch, while Windows Vista and 7 received their final certified updates by 2013 with version 13.9; Windows 8.1 and 10 offered basic functionality only through Microsoft Windows Update thereafter, without AMD-specific enhancements. Linux and Unix support was provided via ports of the Catalyst suite using the fglrx kernel module, offering OpenGL acceleration up to version 4.3 for 3D rendering and basic 2D operations, though hardware limits capped effective support at OpenGL 3.3 for R600 GPUs. These drivers were limited compared to Windows counterparts, lacking advanced features like HydraVision, but included essential controls. The legacy fglrx branch culminated in version 13.1 in January 2013, with deprecating further development by mid-2016 in favor of open-source alternatives. Post-launch updates focused on stability, with early releases like 7.7 addressing synchronization bugs that caused artifacts in multi-GPU setups and improving power efficiency to reduce idle consumption on HD 2900 series cards. Later iterations refined UVD utilization for smoother video playback and fixed intermittent crashes in 10 titles. The drivers never added support for modern APIs like , as the underlying R600 architecture predated those requirements.

Open-source drivers

The open-source driver stack for the Radeon HD 2000 series, based on the R600 GPU architecture, primarily consists of the Radeon kernel module and the Mesa graphics library, providing support for Linux and other open operating systems. This stack enables both 2D and 3D acceleration without relying on proprietary components, with ongoing maintenance ensuring compatibility nearly two decades after the hardware's release. The R600 architecture is hardware-limited to OpenGL 3.3 due to lacking features like double-precision floating point. The Radeon kernel module, part of the Linux Direct Rendering Manager (DRM) subsystem, introduced initial support for R600 GPUs in late 2007, coinciding with the hardware launch. This early implementation provided basic DRM functionality for command processing but lacked full 2D acceleration on R600 cards at the time, with EXA-based 2D support developing in subsequent kernel releases around 2008 through community efforts. By kernel version 3.11 in 2013, dynamic power management (DPM) was added to the Radeon driver, allowing clock and voltage adjustments for improved efficiency on R600 hardware. Experimental integration with the newer amdgpu kernel module has been available since around 2016 for select pre-GCN GPUs including R600, though the legacy Radeon module remains the primary and stable option. For 3D rendering, the Mesa project's Gallium3D r600g driver, introduced experimentally in Mesa 7.9 in October 2010, marked the shift from the classic Mesa driver to a more modular architecture supporting OpenGL on R600 GPUs. Initial features included basic texture support and simple rendering demos like glxgears, with progressive enhancements enabling broader application compatibility. Feature progression continued steadily; by 2017, OpenGL 4.3 conformance was achieved for higher-end R600 variants, though hardware limits cap the series at OpenGL 3.3. Recent updates in Mesa 25.2, released in August 2025, include fixes for OpenGL conformance test failures on R600 and R700 GPUs, improving stability for legacy workloads such as older games and CAD software. Vulkan support remains unavailable in the mainline RADV driver, which targets GCN and newer architectures; an experimental "Terakan" Vulkan driver for pre-GCN AMD GPUs, including R600, entered development in 2023 but is not yet production-ready. Key limitations of the open-source stack for R600 include the absence of native support, as it focuses on and APIs; Windows games require translation layers like Wine or Proton, which use DXVK for DirectX-over- emulation but may suffer performance overhead on older hardware. Power management via the experimental amdgpu path can introduce instability, such as screen artifacts at high refresh rates, prompting most users to stick with the Radeon driver's mature DPM implementation. AMD facilitated open-source development by releasing partial R600 specifications, including the AMD Intermediate Language (IL) reference in October 2011, which detailed instructions and enabled deeper reverse-engineering and implementation of features like unified . Earlier documents, such as the R600 from 2009, provided foundational ASIC overviews and acceleration details. Ongoing improvements stem from community contributions, with developers like those at X.Org and maintaining the stack through bug fixes and optimizations as recently as 2025.

Technical specifications

Chip variants and performance

The Radeon HD 2000 series utilized three primary GPU dies: the high-end R600, mid-range RV630, and entry-level RV610, with mobile variants derived from these cores under the M8xx designations. The R600 powered the flagship HD 2900 lineup, featuring 320 stream processors and a 512-bit memory interface, while the RV630 in the HD 2600 series offered 120 stream processors on a 128-bit bus, and the RV610 for the HD 2400 series provided 40 stream processors with a narrower 64-bit interface. These chips were fabricated on TSMC's 80 nm for the R600 and 65 nm for the others, enabling a shift to unified shader architecture that improved versatility over the prior scalar-focused designs. Memory configurations varied by model and tier, ranging from 128 MB of DDR2 on low-end discrete cards like the HD 2400 Pro to 512 MB of GDDR3 on the standard HD 2900 XT (with a later 1 GB GDDR4 variant available). For instance, the standard HD 2900 XT paired 512 MB GDDR3 at 800 MHz (1,600 MHz effective) on its 512-bit bus, yielding 102.4 GB/s bandwidth, while a 1 GB GDDR4 variant ran at 1,000 MHz (2,000 MHz effective) for 128 GB/s; the HD 2600 XT used 256 MB GDDR4 at 1,000 MHz (2,000 MHz effective) for 32 GB/s, and the HD 2400 XT employed 256 MB GDDR3 at 500 MHz (1,000 MHz effective) for 8 GB/s. Mobile implementations, such as the M82 (Mobility HD 2600 XT), scaled down to 256 MB GDDR3 at 750 MHz on a 128-bit bus for 24 GB/s, prioritizing power efficiency in laptops.
ChipSeriesCore Clock (MHz)Stream ProcessorsMemory Type/SizeMemory Clock (MHz, effective)Bus Width (bit)Bandwidth (GB/s)TDP (W)Mobile Variant
R600HD 2900 XT742320GDDR3 / 512 MB800 (1600)512102.4215M86x (limited)
RV630HD 2600 XT800120GDDR4 / 256 MB1000 (2000)1283245M82 (700 MHz core)
RV610HD 2400 XT65040GDDR3 / 256 MB500 (1000)64845M81 (525 MHz core)
In performance, the HD 2000 series delivered approximately 2x the theoretical shader throughput compared to the , thanks to the unified architecture and increased processor counts—e.g., the HD 2900 XT's 320 shaders versus the X1900 XTX's effective 48 pixel shaders—enabling better handling of 10 workloads. Against Nvidia rivals, the HD 2900 XT matched or slightly trailed the 8800 GTX in rasterization-heavy games like : Episode Two at 1680x1050 (around 60-70 FPS versus 65-75 FPS), but lagged in efficiency due to higher power draw. Mid-range models like the HD 2600 XT offered 20-50% uplift over the X1950 Pro in 9 titles such as , while providing competitive parity with the 8600 GT at lower resolutions. Power efficiency ratios highlighted the series' mixed results: the 65 nm RV630 and RV610 achieved favorable perf/W in mainstream scenarios (e.g., HD 2600 XT at ~1 FPS/W in 3DMark06 versus ~0.8 FPS/W for the X1950 GT), but the 80 nm R600 consumed up to 215 W with lower efficiency (~0.3 FPS/W) compared to the 8800 GTX's ~0.5 FPS/W. OEM variants often featured custom clocks for stability or , such as HIS's HD 2900 XT at 828 MHz core, while professional FireGL rebrands like the V5600 (RV630-based, 800 MHz core, 512 MB GDDR3) added support for error correction in CAD workflows, though without significant clock deviations from consumer models. CrossFire scaling reached up to 1.8x in supported titles for dual-GPU setups.

Feature comparison matrix

The Radeon HD 2000 series introduced several key graphical features standardized across its main sub-series, including full 10 compliance for advanced effects and . All models supported 3.3, enabling compatibility with early programmable pipelines in professional and gaming applications; open-source drivers received enhancements for 4.6 conformance as of 2025 on compatible hardware. capabilities reached up to 24x via Custom Filter Anti-Aliasing (CFAA), offering improved edge smoothing over traditional multi-sample methods, particularly beneficial for high-resolution rendering. Video decoding utilized the first-generation (UVD1), providing for H.264/AVC and formats to enable efficient HD playback without taxing the CPU. The series lacked dedicated hardware for ray tracing or AI-upscaling technologies like DLSS, reflecting its 2007-era focus on rasterization and unified s rather than modern real-time or enhancements.
FeatureHD 2400 SeriesHD 2600 SeriesHD 2900 Series
DirectX SupportFull DirectX 10.0Full DirectX 10.0Full DirectX 10.0
OpenGL Version3.33.33.3
Anti-Aliasing ModesUp to 24x (CFAA/MSAA)Up to 24x (CFAA/MSAA)Up to 24x (CFAA/MSAA)
Video DecodeUVD1 (H.264/VC-1)UVD1 (H.264/VC-1)UVD1 (H.264/VC-1)
Multi-GPU (CrossFire)External bridge connectorNative (e.g., X2 dual-GPU variant); external for singlesNative CrossFire X
Display OutputsDual DVI; HDMI via adapter; no native DisplayPortDual DVI; HDMI via adapter; no native DisplayPortDual DVI; HDMI via adapter; no native DisplayPort
Memory Interface Width64-bit128-bit512-bit
TDP Range20-25 W35-45 W215 W (XT variant)
Process Node65 nm65 nm80 nm
Professional VariantsFireGL V3350 (CAD/3D modeling)FireGL V3600 (CAD/3D modeling)FireGL V7600 (CAD/3D modeling)

References

  1. https://www.beyond3d.com/content/pr/26/
  2. https://hardwaresecrets.com/amd-ati-radeon-hd-2000-series-architecture/
  3. https://gizmodo.com/ati-radeon-hd-2000-series-launched-from-99-to-399-260135
  4. https://www.theregister.com/2007/05/14/amd_graphics_card_launch/
  5. https://www.computer.org/publications/tech-news/chasing-pixels/amd-unified-shader-gpu-history
  6. https://www.copetti.org/writings/consoles/xbox-360/
  7. https://www.nytimes.com/2006/07/24/technology/24cnd-semi.html
  8. https://www.cbc.ca/news/business/amd-completes-takeover-of-ati-1.571612
  9. https://www.techspot.com/article/2689-ati-technologies-history/
  10. http://ixbtlabs.com/articles2/video/r600-part1.html
  11. https://pcper.com/2007/05/amd-ati-radeon-hd-2900-xt-review-r600-arrives/4/
  12. https://www.ign.com/articles/2007/02/22/ati-r600-gpu-delayed-again
  13. https://www.copetti.org/writings/consoles/wiiu/
  14. https://bjorn3d.com/2007/05/hd2000-the-first-gpus-under-the-amd-name-2/
  15. https://arstechnica.com/gadgets/2007/05/amd-launches-the-hd-2000-series/
  16. https://www.techpowerup.com/gpu-specs/radeon-hd-2900-xt.c192
  17. https://www.techpowerup.com/gpu-specs/radeon-hd-2600-xt.c193
  18. https://www.techpowerup.com/31126/amd-introduces-the-ati-radeon-hd-2000-series-for-desktop-and-mobile-platforms
  19. https://www.cnet.com/reviews/ati-radeon-hd-2600-xt-review/
  20. https://www.techpowerup.com/review/ati-hd-2900-xt/13.html
  21. https://www.tomshardware.com/picturestory/575-graphics-radeon-geforce-2.html
  22. https://courses.cs.washington.edu/courses/cse548/11au/AMD-r600.pdf
  23. https://www.techpowerup.com/gpu-specs/ati-r600.g431
  24. https://www.tomshardware.com/picturestory/735-history-of-amd-graphics-3.html
  25. https://www.beyond3d.com/content/reviews/16/2
  26. https://www.techpowerup.com/gpu-specs/radeon-hd-2400-xt.c194
  27. https://www.guru3d.com/review/ati-radeon-hd-2900-xt-512mb-review/page-9/
  28. https://hothardware.com/reviews/ati-radeon-hd-2600-and-2400-performance
  29. https://www.phoronix.com/review/amd_stream_qa
  30. https://www.guru3d.com/review/ati-radeon-hd-2900-xt-512mb-review/page-11/
  31. https://www.gamespot.com/articles/ati-radeon-hd-2000-series-hands-on/1100-6170621/
  32. https://www.techpowerup.com/gpu-specs/ati-rv610.g55
  33. https://www.beyond3d.com/content/pr/47/
  34. https://www.techpowerup.com/gpu-specs/radeon-hd-2400-pro.c997
  35. https://www.tweaktown.com/articles/1165/sapphire_radeon_hd_2400_xt_hd_2600_xt/index.html
  36. https://pcper.com/2007/06/amd-ati-radeon-hd-2600-xt-2600-pro-and-2400-xt-review/
  37. https://www.techpowerup.com/gpu-specs/radeon-hd-2600-pro.c212
  38. https://www.techpowerup.com/gpu-specs/radeon-hd-2600-xt-x2.c1005
  39. https://bit-tech.net/reviews/tech/graphics/his_radeon_hd_2600_xt_iceq_turbo_gddr3_/1/
  40. https://hexus.net/tech/reviews/graphics/9187-amd-radeon-hd-2600-xt-radeon-hd-2400-xt-saviours-sinners/?page=10
  41. https://www.techpowerup.com/review/ati-hd-2600-xt/
  42. https://pcper.com/2007/05/amd-ati-radeon-hd-2900-xt-review-r600-arrives/5/
  43. https://www.techpowerup.com/gpu-specs/radeon-hd-2900-gt.c220
  44. https://www.techpowerup.com/gpu-specs/radeon-hd-2900-xtx.c2140
  45. https://bit-tech.net/reviews/tech/graphics/r600_ati_radeon_hd_2900_xt/20/
  46. https://hothardware.com/reviews/ati-radeon-hd-2900-xt--r600-has-arrived?page=21
  47. https://www.notebookcheck.net/ATI-Mobility-Radeon-HD-2400.3773.0.html
  48. https://www.techpowerup.com/gpu-specs/mobility-radeon-hd-2600-xt.c1620
  49. https://www.notebookcheck.net/ATI-Mobility-Radeon-HD-2300.6940.0.html
  50. https://www.techpowerup.com/gpu-specs/mobility-radeon-hd-2300.c1614
  51. https://www.notebookcheck.net/ATI-Mobility-Radeon-HD-2400-XT.3772.0.html
  52. https://www.techpowerup.com/gpu-specs/mobility-radeon-hd-2400-xt.c1615
  53. https://notebooks.com/2007/05/14/ati-mobility-radeon-hd-2000-series-announced/
  54. https://www.techpowerup.com/gpu-specs/radeon-xpress-1250-igp.c847
  55. https://www.legitreviews.com/amds-690gv-series-chipset-preview-and-benchmarks_467/2
  56. https://bit-tech.net/reviews/tech/amd_690g_chipset/1/
  57. https://www.notebookcheck.net/ATI-Radeon-Xpress-X1250.6946.0.html
  58. https://www.theregister.com/2006/06/29/ati_rs600_is_radeon_1250/
  59. https://www.theregister.com/2007/02/28/amd_launches_690g_chipset/
  60. https://www.pcstats.com/articles/2117/2.html
  61. https://drivers.amd.com/relnotes/catalyst_76_release_notes.html
  62. https://www.amd.com/en/resources/support-articles/release-notes/RN-RAD-WIN-15-7-1.html
  63. https://drivers.softpedia.com/get/GRAPHICS-BOARD/AMD/AMD-Catalyst-Legacy-Graphics-Driver-139.shtml
  64. https://www.techpowerup.com/31181/ati-drivers-for-radeon-hd-2900-series
  65. https://www.anandtech.com/show/3674/amds-catalyst-102-whql-released-opengl-33-support-for-radeon-hd-2000
  66. https://www.guru3d.com/download/ati-catalyst-7-6-windows-xp-32-bit/
  67. https://www.amd.com/en/resources/support-articles/release-notes/rn-rad-win-legacy.html
  68. https://drivers.amd.com/relnotes/catalyst_77_release_notes.html
  69. https://www.phoronix.com/review/amd_catalyst_legacy2
  70. https://www.phoronix.com/news/R600-RV770-Fixes-Mesa-25.2
  71. https://airlied.livejournal.com/53129.html
  72. https://www.phoronix.com/review/960
  73. https://askubuntu.com/questions/324733/how-to-enable-the-radeon-dynamic-power-management-feature
  74. https://askubuntu.com/questions/756258/which-graphics-cards-are-supported-by-the-new-amdgpu-driver-in-ubuntu-16-04
  75. https://docs.mesa3d.org/relnotes/7.9.html
  76. https://www.phoronix.com/news/ODQ0Mw
  77. https://www.phoronix.com/news/OpenGL-43-Lands-In-R600g
  78. https://www.phoronix.com/news/Mesa-Terakan-R600-Vulkan-Driver
  79. https://forums.linuxmint.com/viewtopic.php?t=389489
  80. https://wiki.archlinux.org/title/AMDGPU
  81. https://www.amd.com/content/dam/amd/en/documents/radeon-tech-docs/programmer-references/AMD_Intermediate_Language_IL_Specification_v2.pdf
  82. https://www.phoronix.com/news/NzAxNg
  83. https://www.techpowerup.com/gpu-specs/ati-radeon-hd-2900-xt-1-gb.b6362
  84. https://www.anandtech.com/show/2411/7
  85. https://www.notebookcheck.net/ATI-Mobility-Radeon-HD-2600-XT.3770.0.html
  86. https://pcper.com/2007/05/amd-ati-radeon-hd-2900-xt-review-r600-arrives/
  87. https://www.techpowerup.com/30150/nvidia-8800-gtx-beats-amd-ati-radeon-hd-2900-xtx
  88. https://hothardware.com/reviews/ati-radeon-hd-2600-and-2400-performance?page=10
  89. https://www.techpowerup.com/gpu-specs/firegl-v5600.c849
  90. https://videocardz.net/browse/his/radeon-hd-2000
  91. https://www.techpowerup.com/gpu-specs/radeon-hd-2400.c2087
  92. https://www.videocardbenchmark.net/gpu.php?gpu=Radeon+HD+2400+XT&id=571
  93. https://www.videocardbenchmark.net/gpu.php?gpu=Radeon+HD+2600+PRO&id=572
  94. https://www.videocardbenchmark.net/gpu.php?gpu=Radeon+HD+2900+XT&id=578
  95. https://www.phoronix.com/news/Mesa-25.2-R600g-OpenGL-4.6
  96. https://www.techpowerup.com/review/ati-hd-2900-xt/
  97. https://www.tomshardware.com/reviews/FireGL-Quadro-Workstation%2C1995-9.html
  98. https://hothardware.com/news/diamonds-2gb-vfx-2000-series-pro-graphics-card
Add your contribution
Related Hubs
User Avatar
No comments yet.