Anthony M. Dos Santos Escreveu:Muito bem, a nVidia está à frente na corrida de desempenho e de tecnologia com as 8800GTX. Também é oficial que a AMD está bastante atrasada em relação à sua opositora em termos de mercado. Mas a resposta vem em força, em maior força que todos esperavam.
Começando pelas entrada de gama RV610, apenas se sabe que irão usar 64bits de bus e memória GDDR, GDDR2 e GDDR3. Fabricadas a 65nm, deverão ser muito poupadas, consumindo entre 25 e 35W em full. Eis um quote do Dailytech:
AMD currently has four RV610 reference designs based on two RV610 variants – Antelope FH, Antelope LP, Falcon FH and Falcon LP reference boards and RV610LE and RV610PRO GPUs. Antelope FH and Antelope LP are similar; however, Antelope LP is the low-profile variant. Both reference boards feature 128MB or 256MB of DDR2 video memory clocked at 400 MHz. Antelope boards employ the RV610LE, feature passive cooling and consume less than 25-watts of power.
AMD’s Falcon LP reference board is another low-profile model with 256MB of GDDR3 memory clocked at 700 MHz. Falcon LP takes advantage of a DMS-59 connector for dual video outputs while maintaining a low profile form factor. The Falcon LP reference board employs active cooling to cool the RV610LE or RV610PRO GPU.
AMD Antelope FH, Antelope LP and Falcon LP only support software CrossFire – all lack support for the CrossFire bridge connector. HKEPC confirmed this CrossFire setup in a recent report last week.
The Falcon FH reference board is the performance variant and designed for the RV610PRO ASIC with 256MB of GDDR3 video memory. AMD estimates board power consumption at approximately 35-watts, though it is unknown if Falcon FH boards will feature active or passive cooling. Falcon FH is the only RV610 reference board to support AMD’s CrossFire bridge connector for hardware CrossFire support. Falcon FH also features VIVO capabilities.
Sobre as RV630, começam as surpresas:
As our sources revealed that RV630XT will score over 12K in 3DMark05 which is in fact faster than a X1900 XTX card. As for RV610, you can check out our previous article for its details and performance.
Para quem não sabe, isso significa que as novas RV630XT (X1600XT) são mais performantes que as anteriores topo de gama Geforce 7900GTX e X1950XTX. Muito bom, quando consideramos que é uma gráfica apontada para os 200€.
Sobre caracteristicas:
RV630 has three reference board configurations – Kohinoor, Orloff and Sefadu. Kohinoor is the high-performance RV630 variant and features 256MB or 512MB of GDDR4 memory. It also features VIVO and dual dual-link DVI outputs. However, it consumes the most power out of the three RV630 reference boards, requiring 121-watts for 256MB models and 128-watts for 512MB models.
Orloff falls in the middle with 256MB of GDDR3 video memory. Orloff lacks the video input features of Kohinoor but supports HDMI output. AMD estimates Orloff to consume less than 93-watts of power. Kohinoor and Orloff support PCIe 2.0 and native CrossFire. Kohinoor and Orloff require additional power via PCIe power connector though.
Sefadu falls at the bottom of the RV630 lineup and features 256MB or 512MB of DDR2 video memory. HDMI remains supported, as with Orloff though. Power consumption is estimated at less than 75-watts, and does not require the additional power supplied by a PCIe power connector. All RV630 boards feature 128-bit memory interfaces and occupy a single-slot.
[url=http://www.dailytech.com/article.aspx?newsid=6451
]http://www.dailytech.com/article.aspx?newsid=6451
Ora bem, GDDR2, 3 ou 4 à escolha. Nada mal para uma média gama, e a presença dessa última pode ajudar a baixar os preços das GDDR. A potência é um bocado elevada, mas actualmente não é nada de surpreendente. Pena continuam a ser os 128bits apenas, algo que se deveria mudar, mas visto a nVidia também não se preocupa muito com isso. O desempenho é surpreendente, acima das antigas topo de gama.
Mas claro que toda a gente se interessa mais pelas R600. E sobre estas, a AMD diz:
One spoke about a problem of PCB with memory GDDR4. You confirm?
Well on that not. There is no concern, these rumours are nonfounded. R600 functions very well.
Compared to the solution of NVIDIA, which will give the R600 in terms of performances?
You will be surprised…
E não é bluff:
he big daddy of them all, the Radeon X2900XTX (codenamed Dragons Head 2 in 9.5" size XTX configuration with GDDR4 memory), is due to hit shop shelves on 30th of March 2007 but it's still unsure if they will actually make that date.
Samples were already delivered to AMD's partners last week and many companies will likely have R600 live demos on display at CeBIT.
The Radeon X2900XT (codenamed Cats Eye also measuring 9.5" but with GDDR3 memory and reduced clock speeds about 15% below XTX) will hit shop shelves on 19th of April.
Dragons Head 2 is actually already well into production and product will be delivered to AIB's in just a few days from now.
After many revisions, it seems like AMD have finally got things right and are satisfied with production output - Rev90 it is.
The original and monster Dragons Head measuring 12.4" long will still be sold but recommended just for OEM and system integrators.
AMD will also release the Radeon X2900XT and X2900XL sometime in April with reduced clock speeds and only 512MB GDDR3 memory. The Radeon X2900XTX will use faster 1GB GDDR4 memory operating at over 2000MHz DDR using a 1024-bit internal memory bus and 512-bit physical memory interface.
E também
The R600 is momentary on the CeBIT in all mouth, even if so far only one handful of people got it to face. After we already yesterday in detail reported coming flag ship on AMDs, there were further pikante details on the today's press conference of the processor manufacturer.
The California chip chipschmiede did not only confirm opposite that the R600 will come into 65 Nm, but also communicated that AMDs will have first DirectX 10 GPU a dedicated video processor. „Universal the video in such a way specified Decoding “unit should be trimmed particularly on the rendition of Full HP material (1920x1080). Thus the ATI GPU gets for the first time a dedicated video unit.
With past maps the H.264 acceleration ran over the Shader units of the GPU, which brought a dependence to acceleration with the equipment of the chip with itself. While a X1300 can accelerate only 480p content, it is enough with X1600 maps already for 720p and with X1800/X1900 maps to 1080p - more in addition in our Avivo test. With the UVD however one integrates its own unit, which dedicates itself alone to the H.264 Decoding. Thus the actual GPU is not loaded at all and is available other 3D uses (e.g. the aero surface of Vista). Besides can be processed in such a way up to two 40MBit/s Streams. Thus whole R (V) comes 600 Lineup to comprehensive video acceleration.
The remainder of the presentation held then however no more surprises ready, but was cold coffee. One looked for information for the number of the Unified Shader units or clock rates in vain. Rather „“AMD on well-known details rum eierte. All further information to the press conference can be inferred from the attached Slides.
65nm. Mais pequeno que isso não há em gráficas. Isso significa que a AMD vai lançar um GPU teoricamente mais capaz em termos de velocidades, mais pequeno e mais poupado que o G80. Na prática, ainda não se sabe.... Ninguém esperava um lançamento dessa magnitude, a AMD não mentiu quando disse que iria ser uma surpresa.
E interessante ainda é:
R6xx chips still compatible to Rialto AGP-bridge
ATI's Vijay Sharma (Director of ATI's Desktop Discrete Products) was interviewed by PC Games Hardware and he said the whole R600 lineup will be compatible to the PCIe-to-AGP bridge 'Rialto'.
NVIDIA's Product PR Manager Jens Neuschäfer on the other hand told them there will be no future AGP cards made by NVIDIA. At least as long as there isn't an OEM with a sufficient demand.
PC Games Hardware 1 and PC Games Hardware 2
Para quem não sabe, o Rialto é o chip que converte de PCI-e para AGP. Ou seja, provavelmente iremos ver RV630 e RV610 AGP.
Peço desculpa por ter feito um resumido assim como notícia, deveria ter sido colocado como artigo e irei faze-lo com mais tempo.
Comentem ai
