Radeon R300

维基百科，自由的百科全书

Radeon "R300" 架構(2002年八月推出)是ATI的第一個支援DirectX 9.0的顯示核心，亦是全球第一個完全支援DirectX 9的繒圖核心。它讓ATI在1999年的GeForce 256以來，在性能上首次勝過NVIDIA。R300和其衍生產品形成了ATI的消費級和專業級產品線足有三年那麼久。

它的整合式形號是Xpress 200，與桌面形號最大分別是它使用系統記憶體作為顯示記憶體。

目录 1 開發 1.1 架構 1.2 性能 2 進一步推出 2.1 更新 2.2 新接口 3 形號 4 參考

[编辑] 開發

有一段時間，ATI憑著Radeon 8500，成了效能領導者。但很快的，NVIDIA推出了GeForce 4 Ti產品線，重新成為效能領導者。原先在一個高端的更新，8500XT (R250)會被推出，用來對抗NVIDIA的高端產品，猶其是最強的Ti 4600。根據預先發佈的消息，8500XT的核心和顯示記憶體的頻率將會是300 MHz。ATI也許記得當年3dfx所發生的事，認為就算Radeon 8500以300 MHz的頻率來運行，都不足以對抗GeForce 4 Ti4600；而且還會拖慢下一代支援DirectX 9.0的產品的研發，令NVIDIA有機可乘。

R3xx晶片是由ATI的西岸團隊設計(ArtX Inc.的前身)，而第一個產品就是Radeon 9700 PRO(ATI的內部代號:R300 - ArtX的內部代號: Khan)，它於2002年8月推出。R300的架構與前代的產品，即是Radeon 8500 ("R200")，在很多地方存有差別。9700 PRO的核心以150納米製程製造，與Radeon 8500相似。但是，精鍊的架構和較先進的工藝都令晶片的電晶體數量和頻率都得到雙倍提升。其中一個重要的轉變，是利用了反轉晶片針腳柵格陣列技術製造核心。這是一種從未在製造顯示卡時使用過的技術。這個技術可以令核心能以較快的速度冷卻，因為核心直接接觸到散熱器，而核心就能以更高頻率運作。Radeon 9700 PRO的核心頻率是325 MHz，原先計劃設在300 MHz。核心擁有1.1億個電晶體，是當時最大型和最複雜的核心。一個較慢的核心，9700於於數月後發佈。它們的分別只是核心和顯示記憶體的頻率較慢。出人意外地，Radeon 9700 PRO的頻率顯然比Matrox Parhelia 512高。Matrox Parhelia 512比R300早幾個月推出，曾是顯示核心製造工藝的頂峰(八百萬個電晶體，核心頻率是220 MHz)。

[编辑] 架構

The chip adopted an architecture consisting of 8 pixel pipelines, each with 1 texture mapping unit (an 8x1 design). While this differed from the older chips using 2 texture units per pipeline, this did not mean "R300" could not perform multitexturing as efficiently as older chips. Its texture units could perform a new "loopback" operation which allowed them to sample up to 16 textures per geometry pass. The textures can be any combination of one, two, or three dimensions with bilinear, trilinear, or anisotropic filtering. This was part of the new DirectX 9 specification, along with more flexible floating-point-based Shader Model 2.0+ pixel shaders and vertex shaders. Equipped with 4 vertex shader units, "R300" possessed over twice the geometry processing capability of its predecessor and the GeForce4 Ti 4600, in addition to the greater feature-set offered compared to DirectX 8 shaders.

A noteworthy limitation is that all "R300"-generation chips were designed for a maximum floating point precision of 96-bit, or FP24, instead of DirectX 9's maximum of 128-bit FP32. DirectX 9.0 specified FP24 as a minimum level for conforming to the specification for full precision. This tradeoff in precision offered the best combination of transistor usage and image quality for the manufacturing process at the time. It did cause a usually visibly-imperceptible loss of quality when doing heavy blending. ATI's Radeon chips did not go above FP24 until R520. ATI demonstrated part of what was capable with pixel shader PS2.0 with their Rendering with Natural Light demo. The demo was a real-time implementation of noted 3D graphics researcher Paul Debevec's paper on the topic of high dynamic range rendering.^[1]

The "R300" was the first board to truly take advantage of a 256-bit memory bus. Matrox had released their Parhelia 512 several months earlier, but this board did not show great gains with its 256-bit bus. ATI, however, had not only doubled their bus to 256-bit, but also integrated an advanced crossbar memory controller, somewhat similar to NVIDIA's memory technology. Utilizing four individual load-balanced 64-bit memory controllers, ATI's memory implementation was quite capable of achieving high bandwidth efficiency by maintaining adequate granularity of memory transactions and thus working around memory latency limitations. "R300" was also given the latest refinement of ATI's innovative HyperZ memory bandwidth and fillrate saving technology, HyperZ III. The demands of the 8x1 architecture required more bandwidth than the 128-bit bus designs of the previous generation due to having double the texture and pixel fillrate.

Radeon 9700 introduced ATI's multi-sample gamma-corrected anti-aliasing scheme. The chip offered sparse-sampling in modes including 2X, 4X, and 6X. Multi-sampling offered vastly superior performance over the supersampling method on older Radeons, and superior image quality compared to NVIDIA's offerings at the time. Anti-aliasing was, for the first time, a fully usable option even in the newest and most demanding titles of the day. "R300" also offered advanced anisotropic filtering which incurred a much smaller performance hit than the anisotropic solution of the GeForce4 and other competitors' cards, while offering significantly improved quality over Radeon 8500's anisotropic filtering implementation which was highly angle dependent.

[编辑] 性能

Radeon 9700的先進架構是十分有效率，對比它在2002年時的舊對手是，效能當然是很強大了。在一個普通的情況下，它能比昔日的皇者，GeForce4 Ti 4600快上15-20%。但是，當執行抗鋸齒(AA)和各異向性過濾(AF)運算時，在所有情況下，它能比Ti 4600快上40-100%。在當時來說，這是頗為震驚的，使到人們廣泛接受AA和AF，使之成為必不可少的功能。

Besides advanced architecture, reviewers also took note of ATI's change in strategy. The 9700 would be the first of ATI's chips to be shipped to third-party manufacturers instead of ATI producing all of its graphics cards (ATI would still produce cards off of its highest-end chips). This freed up engineering resources that were channelled towards driver improvements, and the 9700 performed phenomenally well at launch because of this.

The performance and quality increases offered by the R300 GPU is considered to be one of the greatest in the history of 3D graphics, alongside the achievements GeForce 256 and Voodoo Graphics. Furthermore, NVIDIA’s response in the form of the GeForce FX 5800 was both late to market and somewhat unimpressive, especially when pixel shading was used. R300 would become one of the GPUs with the longest useful lifetime in history, allowing playable performance in new games at least 3 years after its launch.^[2]

[编辑] 進一步推出

數月之後，9500和9500 PRO推出。9500 PRO的記憶體頻寬只有9700 PRO的一半，而9500的像素處理單元亦被遮蔽到只有9500 PRO的一半，and the hierarchical Z-buffer optimization unit (part of HyperZ III). With its full 8 pipelines and efficient architecture, the 9500 PRO outperformed all of NVIDIA’s products (save the Ti 4600). Meanwhile, the 9500 also became popular because it could in some cases be modified into the much more powerful 9700 non-PRO (np). ATI only intended for the 9500 series to be a temporary solution to fill the gap for the 2002 Christmas season, prior to the release of the 9600. Since all of the "R300" chips were based on the same physical die, ATI's margins on 9500 products were low. Radeon 9500 was one of the shortest-lived product of ATI, later replaced by the Radeon 9600 series. The logo and box package of the 9500 was "resurrected" in 2004 to market the unrelated and slower Radeon 9550 (which is a derivative of the 9600).

[编辑] 更新

In early 2003, the 9700 cards were replaced by the 9800 (a.k.a. R350). These were R300s with higher clock speeds, and improvements to the shader units and memory controller which enhanced anti-aliasing performance. They were designed to maintain a performance lead over the newly launched GeForce FX 5900 Ultra, which it managed to do without difficulty. The 9800 still held its own against the revised 5900, primarily (and significantly) in tasks involving heavy SM2.0 pixel shading. A later version with 256 MB of memory used GDDR2. The other two variants were the 9800 "non-pro" , which was simply a lower-clocked '9800 Pro, and the 9800 SE, which had half the pixel processing units disabled (could sometimes be enabled again). Official ATI specs dictate a 256-bit memory bus for the 9800 SE, but most of the manufacturers used a 128-bit bus. Usually, the 9800 SE with 256-bit memory bus was called "9800 SE Ultra" or "9800 SE Golden Version".

Alongside the 9800, the 9600 (a.k.a. RV350) series was rolled out in early 2003, and while the 9600 PRO didn't outperform the 9500 PRO that it was supposed to replace, it was much more economical for ATI to produce by way of a 130nm process (all ATI's cards since the 7500/8500 had been 150nm) and a simplified design. Radeon 9600's "RV350" core was basically a 9800 PRO cut in half, with exactly half of the same functional units, making it a 4x1 architecture with 2 vertex shaders. It also lost part of HyperZ III with the removal of the hierarchical z-buffer optimization unit, the same as Radeon 9500. Using a 130nm process was also good for pushing up the core clock speed. The 9600 series, all with high default clocking, was shown to have quite a bit of headroom by overclockers (achieving over 500 MHz, from 400 MHz on the Pro model). While the 9600 series was less powerful than the 9500 and 9500 PRO it replaced, it did largely manage to maintain the 9500's lead over NVIDIA’s GeForce FX 5600 Ultra, and it was ATI's cost-effective answer to the long-time mainstream performance board, GeForce4 Ti 4200.

During the summer of 2003, the Mobility Radeon 9600 was launched, based upon the RV350 core. Being the first laptop chip to offer DirectX 9.0 shaders, it enjoyed the same success of the previous Mobility Radeons. The Mobility Radeon 9600 was originally planned to use a RAM technology called GDDR2-M. The company developing that memory went bankrupt and the RAM never arrived, so ATI was forced to use regular DDR SDRAM. Undoubtedly there would have been power usage savings, and perhaps performance gains with GDDR2-M. In fall 2004, a slightly faster variant, the Mobility Radeon 9700 was launched (which was still based upon the RV350, and not the older R300 of the desktop Radeon 9700 despite the naming similarity).

Later in 2003, three new cards were launched - the 9800 XT (R360), the 9600 XT (RV360), and the 9600 SE (RV350). The 9800 XT was slightly faster than the 9800 PRO had been, while the 9600 XT competed well with the newly launched GeForce FX 5700 Ultra.^[3] The "RV360" chip on 9600 XT was the first graphics chip by ATI that utilized Low-K chip fabrication and allowed even higher clocking of the 9600 core (500 MHz default). The 9600 SE was ATI's answer to NVIDIA’s GeForce FX 5200 Ultra, managing to outperform the 5200 while also being cheaper. Another "RV350" board followed in early 2004, on the Radeon 9550, which was a Radeon 9600 with a lower core clock (though an identical memory clock and bus width).

Worthy of note regarding the R300-based generation is that the entire lineup utilized single-slot cooling solutions. It was not until the R420 generation's Radeon X850 XT Platinum Edition, in December 2004, that ATI would adopt an official dual-slot cooling design.^[4]

[编辑] 新接口

Also in 2004, ATI released the Radeon X300 and X600 boards. These were based on the "RV370" and "RV380" GPU respectively. They were nearly identical to the chips used in Radeon 9550 and 9600, only differing in that they were native PCI Express offerings. These were very popular for Dell and other OEM companies to sell in various configurations; connectors: DVI vs. DMS-59, card height: full-height vs. half-height.

[编辑] 形號

桌面平臺顯示卡
形號	核心類型	晶粒製程	頻率 (MHz) 核心/顯存	核心配置1	填充率2 (MTex/s)	Geometry3 (MTri/s)	記憶體 介面	記憶體 頻寬	註釋
9500	R300	150 nm	275/270	4:4	1100	275	128-bit	8.6 GB/s	Hierarchical-Z disabled.
9500 PRO	R300	150 nm	275/270	8:4	2200	275	128-bit	8.6 GB/s	Core identical to 9700.
9700	R300	150 nm	275/270	8:4	2200	275	256-bit	17.3 GB/s
9700 PRO	R300	150 nm	325/310	8:4	2600	325	256-bit	19.8 GB/s
9800 SE	R350	150 nm	325/270	4:4	1300	325	128-bit	8.6 GB/s
9800 SE	R360	150 nm	325/290	4:4	1300	380	256-bit	18.6 GB/s
9800	R350	150 nm	325/290	8:4	2600	325	256-bit	18.6 GB/s
9800 PRO	R350, R360	150 nm	380/340	8:4	3040	380	256-bit	21.8 GB/s	340 MHz 128 MB DDR or 350 MHz 256 MB GDDR2.
9800 XL	R350	150 nm	350/310	8:4	2800	350	256-bit	19.8 GB/s	Cost reduced model produced for OEMs like Medion.
9800 XXL	R360	150 nm	390/337.5	8:4	3120	390	256-bit	21.6 GB/s	Higher clocked version of 9800 XL.
9800 XT	R360	150 nm	412/365	8:4	3296	412	256-bit	23.4 GB/s	256 MB GDDR2
9550 SE	RV350	130 nm	250/200	4:2	1000	125	64-bit	3.2 GB/s
9550	RV350	130 nm	250/200	4:2	1000	125	128-bit	6.4 GB/s
9600 SE	RV350	130 nm	325/200	4:2	1300	163	64-bit	3.2 GB/s
9600	RV350	130 nm	325/200	4:2	1300	163	128-bit	6.4 GB/s
9600 PRO	RV350	130 nm	400/300	4:2	1600	200	128-bit	9.6 GB/s
9600 XT	RV360	130 nm	500/300	4:2	2000	250	128-bit	9.6 GB/s	first Low-K 130 nm.
X300 SE	RV370	110 nm	325/300	4:2	1300	163	64-bit	4.8 GB/s	RV370 is PCIe native. HyperMemory model avail.
X300 LE	RV370	110 nm	325/200	4:2	1300	163	128-bit	6.4 GB/s
X300	RV370	110 nm	325/200	4:2	1300	163	128-bit	6.4 GB/s
X550	RV370	110 nm	400/250	4:2	1600	200	128-bit	8.0 GB/s
X600 PRO	RV380	130 nm	400/300	4:2	1600	200	128-bit	9.6 GB/s	RV380 is PCIe native
X600 XT	RV380	130 nm	500/370	4:2	2000	250	128-bit	11.8 GB/s
X1050	RV370	110 nm	325/200	4:4:4:2	1300	165	128-bit	6.4 GB/s	renamed Radeon X300, SM2.0 support
Mobility Radeons和整合式圖像處理器
MR9550 32 MB	M10	130 nm	300/200	4:2	1200	150	64-bit	3.2 GB/s	Mobile RV350. Powerplay power management.
MR9550	M10	130 nm	300/200	4:2	1200	150	128-bit	6.4 GB/s
MR9600 32 MB	M10	130 nm	300/200	4:2	1200	150	64-bit	3.2 GB/s
MR9600	M10	130 nm	300/200	4:2	1200	150	128-bit	6.4 GB/s
MR9600 PRO	M10	130 nm	333/200	4:2	1332	167	128-bit	6.4 GB/s
MR9600 PRO Turbo	M10	130 nm	333/240	4:2	1332	167	128-bit	7.7 GB/s
MR9700	M11	130 nm	450/260	4:2	1800	225	128-bit	8.3 GB/s	Low-K. RV360-based
Xpress 200	RS480	130 nm	300/266	2:0	666	?	64/128-bit	NA	Based on X300. Mostly host based vertex processing w/ support for VS 2.0. Hypermemory-like memory set up. 0 MB - 128 MB local RAM. Uses system RAM as well. SurroundView 3-display (with separate ATI card).

• Bold rows designate initial showings of the major core types.
• ¹ (# of Pixel pipelines) : (# of vertex shaders). All chips of this generation have 1 texture mapping unit (TMU) per pixel pipeline.
• ² MTex/s = Million Texels per second, a measure of texturing fillrate. All chips of this generation have equal texture and pixel fillrates because of having only a single TMU per pipeline.
• ³ MTri/s = Million triangles per second, a measure of the core's geometric calculation capabilities. Related to core speed and the number of vertex shaders.

[编辑] 參考

1. ↑ Debevec, Paul. Rendering with Natural Light, Author's web page, 1998
2. ↑ Weinand, Lars. VGA Charts VII: AGP Update Summer 2005, Tom's Hardware, July 5, 2005.
3. ↑ Gasior, Geoff. NVIDIA's GeForce FX 5700 Ultra GPU: Third time's the charm?, The Tech Report, October 23, 2003.
4. ↑ Wasson, Scott. ATI's Radeon X850 XT graphics cards: Canadian double-wide?, The Tech Report, December 1, 2004.

• "3D Chip and Board Charts" by Beyond3D, retrieved January 10 2006
• "ATI’s Radeon 9700 (R300) – Crowning the New King" by Anand Lal Shimpi, Anandtech, July 18 2002, retrieved January 10 2006
• "ATI Radeon 9700 PRO Review" by Dave Baumann, Beyond3D, August 19 2002, retrieved January 10 2006
• "Matrox's Parhelia - A Performance Paradox" by Anand Lal Shimpi, Anandtech, June 25 2002, retrieved January 10 2006
• "A look at the Geforce 6600GT" by the Firing Squad, retrieved November 11 [[2006]
• "Infos zur ALDI Grafikkarte Radeon 9800 XXL(in German)"- Infos zur ALDI Grafikkarte Radeon 9800 XXL, retrieved November 21 2006

ATi顯示核心列表 2D Chips: Mach DirectX 5/6: 3D Rage • Rage Pro • Rage 128 DirectX 7.x: Radeon R100 DirectX 8.x: Radeon R200 DirectX 9.x: Radeon R300 • R420 • R520 DirectX 10: Radeon R600

ATi公司GPU外的產品、技術 晶片組: IGP3xx • 9100 IGP • 480X • 580X • 690G 多重GPU技術: AMR • CrossFire 專業級顯示晶片/介面卡: FireGL • FireMV 消費性电子產品: Imageon 其他技術: AVIVO • HyperMemory 用於zh-cn:遊戲主機;zh-hk:電動遊戲操縱臺;zh-tw:電視遊樂器的技術: Flipper/GameCube • Xenos/Xbox 360 • Hollywood/Wii