| Article Index |
|
HIS Radeon HD 4870 IceQ 4+ Turbo 1GB video card review
|
|
RV770 architecture&heading=HIS Radeon HD 4870 IceQ 4+ Turbo review
|
|
HIS Radeon HD 4870 IceQ 4+ Turbo&heading=HIS Radeon HD 4870 IceQ 4+ Turbo
|
|
Test setup, synthetic benchmarks&heading=Test setup, synthetic benchmarks
|
|
Fallout 3, ET:QW&heading=Fallout 3, ET:QW
|
|
Left 4 Dead, Crysis&heading=Left 4 Dead, Crysis
|
|
World in Conflict,GRID&heading=World in Conflict,GRID
|
|
Far Cry 2, Unreal Tournament 3&heading=Far Cry 2, Unreal Tournament 3
|
|
High IQ, Overclocking, video playback
|
|
Power, Temperature, Noise
|
|
Conclusions
|
Page 2 of 11 - RV770 architecture
RV770 architecture
We covered AMD's RV770 architecture in far more thorough detail in a previous article, so be sure to check it out if you want a more comprehensive description of the technology which powers this particular graphics board.
Starting out with the basics, AMD's RV770 core uses the now-established 55 nanometre manufacturing process, packing in 956 million transistors to bring us this DirectX 10.1 and PCI Express 2.0 supporting part, and again featuring tessellation hardware which won't see support in an official API until the release of DirectX 11.
RV770 certainly packs plenty into that transistor count, sporting a massive 800 Stream Processors to handle and process vertex, pixel and geometry shader data. These Stream Processors are arranged as arrays of sixteen units (or SIMD cores), of which there are ten. Each of these SIMD cores has a dedicated texturing unit (again making for ten in total), housing four texturing addressing processors, sixteen 32-bit floating point texture samplers and four 32-bit floating point texturing filtering units.
Much like its predecessor, RV770 employs four ROP partitions, making for a total of sixteen units (with four ROPs in each partition). The core as a whole can process sixteen pixels per clock in 2x and 4x multi-sample anti-aliasing modes, and eight pixels per clock with 8x multi-sampling enabled. All of the same anti-aliasing modes seen with AMD's Radeon HD 2000 and 3000 series are supported, meaning that Custom Filter anti-aliasing is also available alongside standard multi-sampling, with transparency anti-aliasing also possible using either multi-sampling or super-sampling.
As with RV670, the Radeon HD 4800 series employs a 256-bit memory controller, but replaces its predecessors ring bus design with a distributed controller, which supports GDDR3, GDDR4 and GDDR5 memory types.
RV770 also features AMD's second generation UVD video decoding engine, which now supports dual stream decoding (i.e. handling two High Definition video streams so that playback of both is GPU accelerated) and dynamic contrast enhancement. Full HDMI output is still supported, with the addition of full eight-channel audio output with support for all of the major formats used by DVD and Blu-Ray movies.
The Radeon HD 4870 features a fully functional RV770 core, with a reference board sporting a core clock speed of 750MHz. A standard part also features either 512MB or (as in this case) 1GB of GDDR5 memory, clocked at 900MHz - However, due to the unique properties of this new memory standard over its predecessors which allows for double the bandwidth to be transferred per pin per clock cycle, the effective speed of these memory modules are 3.6GHz. The HIS board under the spotlight today sports higher clock speeds than this, which we'll examine in just a second.
To round off, here are the highlights of the Radeon HD 4800 series architecture in full.
- • 956 million transistors on 55nm fabrication process
• PCI Express 2.0 x16 bus interface
• 256-bit GDDR3/GDDR5 memory interface
• Microsoft® DirectX® 10.1 support
o Shader Model 4.1
o 32-bit floating point texture filtering
o Indexed cube map arrays
o Independent blend modes per render target
o Pixel coverage sample masking
o Read/write multi-sample surfaces with shaders
o Gather4 texture fetching
• Unified Superscalar Shader Architecture
o 800 stream processing units
- Dynamic load balancing and resource allocation for vertex, geometry, and pixel shaders
- Common instruction set and texture unit access supported for all types of shaders
- Dedicated branch execution units and texture address processors
- 128-bit floating point precision for all operations
o Command processor for reduced CPU overhead
o Shader instruction and constant caches
o Up to 160 texture fetches per clock cycle
o Up to 128 textures per pixel
o Fully associative multi-level texture cache design
o DXTC and 3Dc+ texture compression
o High resolution texture support (up to 8192 x 8192)
o Fully associative texture Z/stencil cache designs
o Double-sided hierarchical Z/stencil buffer
o Early Z test, Re-Z, Z Range optimization, and Fast Z Clear
o Lossless Z & stencil compression (up to 128:1)
o Lossless color compression (up to 8:1)
o 8 render targets (MRTs) with anti-aliasing support
o Physics processing support
• Dynamic Geometry Acceleration
o High performance vertex cache
o Programmable tessellation unit
o Accelerated geometry shader path for geometry amplification
o Memory read/write cache for improved stream output performance
• Anti-aliasing features
o Multi-sample anti-aliasing (2, 4 or 8 samples per pixel)
o Up to 24x Custom Filter Anti-Aliasing (CFAA) for improved quality
o Adaptive super-sampling and multi-sampling
o Gamma correct
o Super AA (ATI CrossFireX™ configurations only)
o All anti-aliasing features compatible with HDR rendering
• Texture filtering features
o 2x/4x/8x/16x high quality adaptive anisotropic filtering modes (up to 128 taps per pixel)
o 128-bit floating point HDR texture filtering
o sRGB filtering (gamma/degamma)
o Percentage Closer Filtering (PCF)
o Depth & stencil texture (DST) format support
o Shared exponent HDR (RGBE 9:9:9:5) texture format support
• OpenGL 2.0 support
• ATI Avivo™ HD Video and Display Platform
o 2nd generation Unified Video Decoder (UVD 2)
o Enabling hardware decode acceleration of H.264, VC-1 and MPEG-2
o Dual stream playback (or Picture-in-picture)
o Hardware MPEG-1, and DivX video decode acceleration
- Motion compensation and IDCT
o ATI Avivo Video Post Processor
- New enhanced DVD upconversion to HD new!
- New automatic and dynamic contrast adjustment new!
- Color space conversion
- Chroma subsampling format conversion
- Horizontal and vertical scaling
- Gamma correction
- Advanced vector adaptive per-pixel de-interlacing
- De-blocking and noise reduction filtering
- Detail enhancement
- Inverse telecine (2:2 and 3:2 pull-down correction)
- Bad edit correction
- Full score in HQV (SD) and HQV (HD) video quality benchmarks
o Two independent display controllers
- Drive two displays simultaneously with independent resolutions, refresh rates, color controls and video overlays for each display
- Full 30-bit display processing
- Programmable piecewise linear gamma correction, color correction, and color space conversion
- Spatial/temporal dithering provides 30-bit color quality on 24-bit and 18-bit displays
- High quality pre- and post-scaling engines, with underscan support for all display outputs
- Content-adaptive de-flicker filtering for interlaced displays
- Fast, glitch-free mode switching
- Hardware cursor
o Two integrated DVI display outputs
- Primary supports 18-, 24-, and 30-bit digital displays at all resolutions up to 1920x1200 (single-link DVI) or 2560x1600 (dual-link DVI)3
- Secondary supports 18-, 24-, and 30-bit digital displays at all resolutions up to 1920x1200 (single-link DVI only)3
- Each includes a dual-link HDCP encoder with on-chip key storage for high resolution playback of protected content4
o Two integrated 400MHz 30-bit RAMDACs
- Each supports analog displays connected by VGA at all resolutions up to 2048x15363
o DisplayPort™ output support
- Supports 24- and 30-bit displays at all resolutions up to 2560x16003
o HDMI output support
- Supports all display resolutions up to 1920x10803
- Integrated HD audio controller with up to 2 channel 48 kHz stereo or multi-channel (7.1) AC3 enabling a plug-and-play cable-less audio solution
o Integrated AMD Xilleon™ HDTV encoder
- Provides high quality analog TV output (component/S-video/composite)
- Supports SDTV and HDTV resolutions
- Underscan and overscan compensation
o MPEG-2, MPEG-4, DivX, WMV9, VC-1, and H.264/AVC encoding and transcoding
o Seamless integration of pixel shaders with video in real time
o VGA mode support on all display outputs
• ATI PowerPlay™
o Advanced power management technology for optimal performance and power savings
o Performance-on-Demand
- Constantly monitors GPU activity, dynamically adjusting clocks and voltage based on user scenario
- Clock and memory speed throttling
- Voltage switching
- Dynamic clock gating
o Central thermal management – on-chip sensor monitors GPU temperature and triggers thermal actions as required
• ATI CrossFireX™ Multi-GPU Technology
o Scale up rendering performance and image quality with two GPUs
o Integrated compositing engine
o High performance dual channel bridge interconnect
|
