AMD Radeon R9 300 series

[Winjer]

Fica aqui o tópico para discutirmos a próxima série de gráficas da AMD.

Começo com esta noticia das Radeon 380X

AMD To Launch Radeon R9 380X in February

AMD held their annual event over the weekend in Japan. At the event the company revealed to journalists a number of different exciting plans in a round table discussion. The most exciting of these was that AMD is prepping a Radeon R9 380X model that will be available as soon as February if plans go well.

4game-amd-event


AMD says that the Radeon R9 380X will compete head on with NVIDIA’s GeForce GTX 980 graphics card. The R9 380X will be based on the Pirates Islands micro-architecture and will be powered by the Fiji GPU. We all thought that the R9 390X would be based off the Fiji GPU, but apparently that is not the case.


AMD actually have three graphics cards in the works. The Radeon R9 390X will be based on the Bermuda GPU, the R9 380X will be based off the Fiji GPU and the R9 370X will be based off Treasure Island. AMD hopes to launch the R9 380X as soon as February if all goes well.


These new series of graphics cards are said to feature TSMC’s new 20 nm manufacturing technology as well as 3D stacked HBM memory. The hybrid “Hydra” cooling solution developed by Asetek will be used on the R9 390X, we are not sure about other cards.


AMD Radeon R9 390X


The R9 390X is expected to compete against the unreleased NVIDIA GeForce GTX 980 Ti / Titan X. The R9 380X against the GTX 980 / 970 and the R9 370X against the GTX 960.

[Winjer]

Se isto for verdade, a AMD poderá ter aqui uma boa vantagem em relação à nVidia.
No entanto, lembro que a fonte é o WCCFTech.......e estes gajos publicam tudo o que vem à rede.
HBM 3D Stacked Memory is up to 9X Faster Than GDDR5 – Coming With AMD R9 390X and R9 380X 20nm Pirate Islands

[Leak] A recent presentation of the AMD and SK Hynix co-developed 3D stacked High Bandwidth Memory standard HBM has made its way to the web. This new leak gives us a very good idea of what to expect from the new technology in terms of performance and market availability.
HBM was originally developed in tandem by both SK Hynix and AMD to replace GDDR5 as the new standard for bandwidth hungry processors such as GPUs. GDDR5 performance scaling has slowed down dramatically and grown exponentially more expensive in the last few years. Faster GDDR5 modules and wider memory interfaces were only going to take us so far, a new standard had to replace the aging technology.
AMD Pirate Islands, R9 390X and 380X Will Feature HBM.

HBM will allegedly make its way to AMD’s next generation of Pirate Islands GPUs starting early next year. With the introduction of the R9 380X powered by the Fiji GPU core and competing against the GTX 980. Also the R9 390X flagship powered by the Bermuda GPU which will be competing against the Titan 2 / GTX 980 Ti. Which is also rumored to be fully water cooled right out of the factory.
HBM represents the revolutionary step that has been so badly needed in the evolution of memory standards. The first generation of HBM promises to deliver 4.5X the bandwidth of GDDR5 and a staggering 16 times the bandwidth of DDR3.
Note that single channel DDR3 controllers normally have 32bit wide I/O. Leading to 8GB/s bandwidth. But that number is still eclipsed by the 128 GB/s figure for first generation HBM.
The second generation promises to double the bandwidth by doubling the speed from 1Gbps to 2Gbps. While also quadrupling the memory capacity for 4-Hi stacks from 1GB to 4GB.

AMD and SK Hynix will continue to scale this technology well into the mid 2020s. In terms of performance, capacity and power efficiency. Availability of HBM memory to SK Hynix and its partners begins as of now Q4 2014.
HBM will not only make its way to next generation GPUs but also APUs. You can check our detailed coverage of the technology here.
The first graphics products to feature HBM are allegedly going to be AMD’s 20nm Pirate Island Rx 300 series GPUs in 2015. In addition to all the new architectural improvements coming with the Pirate Islands graphics architecture. The Rx 300 series is expected to be the first to feature TSMC’s 20nm manufacturing technology a full shrink from the current 28nm technology. And also be the first to be equipped with HBM. TSMC’s 20nm technology allows AMD to drastically reduce power consumption and heat as as well as drive clock speeds up. HBM also adds a healthy doze of power efficiency to the equation by reducing active power of the memory system to nearly half that of GDDR5.

Nvidia will be using 3D stacked memory on its GPUs starting with Pascal in 2016, a full year after AMD. Volta was Nvidia’s scheduled GPU for 2015 that was supposed to use the Hybrid Memory Cube or HMC for short which was s 3D stacked memory technology competing with HBM. However after HMC’s development fell behind the roadmap it put a lot of pressure on Nvidia to look for an alternative technology, which they did. What’s painfully ironic however is that Nvidia ended up using the AMD and SK Hynix co-developed HBM with Pascal. Which the company confirmed earlier this year. Nvidia very rarely uses a technology developed by their competition if ever.
AMD bet on HBM early on, while Nvidia bet on the competing HMC standard. It seems AMD bet on the right technology and they stand to gain an entire year’s worth of exclusive access to the technology. Nvidia will simply have to make due with GDDR5 for 2015.

[Jorge-Vieira]

Esperemos que isso seja verdade, são excelentes noticias para a AMD e mantém as coisas equilibradas na questão da competitividade.
Essa noticia também pode justificar o porquê de estarmos tanto tempo sem novidades dos GPU´s da AMD.

[MAXLD]

What’s painfully ironic however is that Nvidia ended up using the AMD and SK Hynix co-developed HBM with Pascal. Which the company confirmed earlier this year. Nvidia very rarely uses a technology developed by their competition if ever.

Lol, seria, ou será interessante... AMD a renovar mais uma vez os standards e outros a torcer o braço e ir atrás. Não quer dizer que a nVidia não vá fazer melhor com essa tech quando a tiver, mas o facto da adoptar já seria bom para o mercado avançar.

O que saíra das 380X é uma incógnita, mas -42% de consumo de energia nesse aspecto já deve ajudar bem naquela "inconveniente" que nós sabemos...

[Winjer]

A nVidia até pode ter de dar o braço a torcer com outro standard, o Free-sync. Não me admirava que daqui a dois anos todos os monitores tivessem suporte para este padrão VeSA e apenas meia dúzia de monitores tivessem o G-sync.

[MAXLD]

A nVidia até pode ter de dar o braço a torcer com outro standard, o Free-sync. Não me admirava que daqui a dois anos todos os monitores tivessem suporte para este padrão VeSA e apenas meia dúzia de monitores tivessem o G-sync.

Esse vai demorar bem mais tempo, se é que vai acontecer. Todo o investimento no desenvolvimento e promoção da novidade foi da nVidia, assim como os milhões gastos em contractos e afins. Houve muito guito investido nisto, há que ter o retorno. Aceitar o freesync seria deitar isso ao lixo. Portanto não vai ser nesta década. O mais certo é atirarem ainda mais dinheiro para o balde para aumentar o marketing, contractos e afins. Coisa que a AMD não irá fazer, já que é algo que não está 100% mas suas mãos, nem tem largo investimento a recuperar.

Neste caso das memórias, é um pouco o contrário. Não foi a nVidia a ter o trabalho de fazer a tech HMC, portanto apenas é uma questão de adoptar o que houver no mercado vindo de terceiros.

[Jorge-Vieira]

A nVidia até pode ter de dar o braço a torcer com outro standard, o Free-sync. Não me admirava que daqui a dois anos todos os monitores tivessem suporte para este padrão VeSA e apenas meia dúzia de monitores tivessem o G-sync.

Não me parece que a nVidia vá nesse sentido nos proximos anos, com o forte investimento e todo o marketing associado ao G-Sync, enquanto não existir um retorno finaceiro a nVidia vai continuar na sua.

[Winjer]

Ou então torna-se um buraco sem fundo como o physx. Onde temos a nvidia a andar continuamente a financiar alguns estúdios para estes implementarem o physx por gpu nos seus jogos.

[Jorge-Vieira]

Physx... meh. Quantos jogos é que utilizam physx por gpu que saiem durante um ano?
Acho que se não fosse a nVidia a injectar dinheiro o Physx já tinha morrido à muito.
E duvido muito que a nVidia tenha algum retorno com o physx, pelo menos a nível de jogos.

[MTPS]

Marketing...puro marketing.

Nem 0.00001% dos jogos suportam Physx, mas muito boa gente que vai buscar Nvidia vai iludida com isso.

"Leva Nvidia...é melhor em tudo e ainda tem Physx...", depois lá vão buscar um video do Batman para terem razão

[Jorge-Vieira]

O melhor depende do ponto de vista de cada um, ambos tem coisas boas e más.
Agora neste momento com as GTX 970 e 980 nVidia é melhor que AMD, ambos repartem as lideranças e afinal é isso que interessa.
Physx, é um lixo... e ir buscar videos todos pipis do Batman, mehhh, não é por ter physx ou deixar de ter que as placas são melhores ou piores, physx é mais uma tech para constar no roll das coisas praticamente inuteis a nível de jogos.

[LPC]

Boas!
Vamos esperar para ver...

A AMD deveria ter já lançado alguma coisa, pelo menos para vender no natal...
Assim a Nvidia vai andar na boa vidinha a vender 970 e 980 a preços acima do que na realidade são os chips (a história repete-se como a geração passada).

O real chip high-end está guardado na gaveta para o high-end da AMD.

Portanto as 970, seriam na verdade as 960 (a 250€ ou menos), as 980 são no fundo as 970 e as 980 serão as 980TI e possívelmente ainda deve existir outra versão com um bus acima para as 990 ou Titan X.
Lembro-me bem quando existia praticamente uma duplicação de performance entre a geração anterior e a seguinte... Sentia-mos o salto...

Assim são pequenos passos, pensados para fazer Cow Milking e pouco mais...
Seja do lado da AMD ou da NVIDIA.

Se bem que no lado da AMD, as 290 e 290x foram um bom passo em frente VS as 7950 e 7970 sejam elas as 280 ou 280x e até mesmo as 285.

Cumprimentos,

LPC

[Jorge-Vieira]

AMD’s Next Generation Fiji XT GPU Spotted In Shipping Manifesto – Radeon R9 300 Series Production Might Be Ramping Up

Read m

Zauba has become a source of many early graphics processing units spotting such as the GK110, GK210, GM204 and the flagship GM200 core. The site has also listed several AMD chips along the line and Videocardz has just recently spotted the Fiji XT GPU in a shipping manifesto which could be an early sample of what is likely to be a part of AMD’s next generation Radeon R9 300 series.
AMD’s Fiji XT GPU For Next Generation Radeon R9 300 Series Spotted

According to the data uploaded on the website, the specific card is labeled 102-C88001-00. From deeper insight it is revealed that the Radeon R9 290 and Radeon R9 290X which are AMD’s current flagship offerings based on the Hawaii chip feature the C671 board hence it is possible that the C880 board would be a part of the Radeon R9 390X and Radeon R9 390 (Fiji XT / Fiji Pro) chips. While we don’t know anything specific about Fiji in terms of specifications, we do have a hint of the design which will be featured on the upcoming cards which will be a much needed change over the current cards.
For starters, a modified version of the Hydra cooler that is featured on AMD’s Radeon R9 295X2 graphics card was leaked around 2 months ago. The leaked design was obviously built for a single chip card since it was short and had a large intake fan cut through which air was to be diverted out of the exhaust. The R9 295X2 has a fan placed in the center since it has to push air towards both sides because of the two Hawaii chips fused on it. But AMD had a surprise this time around by shipping their fastest card with an interesting design. The card included a AIO closed loop liquid cooler which turned out to be well for the hot Hawaii cores. AMD is going the same route with their next generation series and looks like at least the flagship offerings will have the option to get the liquid cooler bundle as seen from the vertical cut outs on the cooler shroud. As suggested by Videocardz, there could be four variants of this cooler which will be shipping to consumers early next year.
But that’s not all when it comes to AMD GPUs, there are three more chips which will become a part of the Radeon R9 300 series. The Tonga GPU which was recently announced was expected to launch in two variants, a Tonga Pro with 1792 stream processors plus 256-bit bus and a Tonga XT variant with 2048 stream processors with either 256-bit or 384-bit bus. The Radeon R9 285 and Radeon R9 M285X are the only two variants which were launched but the Radeon R9 285X which was expected to come out during last month never came out. While AMD never officially announced the card itself, AIB partners already had NDA for the card in plans and were expected to out it however the launch of NVIDIA’s GM204 disrupted the pattern. The expected cost of the Radeon R9 285X was supposed to be around $299 US but that would have made it less demanding over the GeForce GTX 970 which costs $20 US more but is a league of its own in terms of performance and efficiency.
The only route left for AMD was to keep the chip stocked and launch it during the next generation Radeon series. It is likely to be a Radeon R9 370 series or R9 380 series part. It is not known what AMD will do with Hawaii during the next generation since it is a new chip and they can rebrand it as the R9 380 series as they have done with Tahiti (Radeon R9 280 series). Regardless, AMD seems to be ramping up production with their next series probably hitting the market early next year. We can’t say for sure what kind of performance will the flagship cards feature but we do know how Tonga and Hawaii perform so we can make our own guesses.
On a side note, we all know that AMD’s AIB partners recently introduced new 8 GB variants of their cards which look attractive in this era of 4K and 8K display resolutions. In the long term, AMD has tough competition ahead of them with Maxwell selling well and have been around for two months now and NVIDIA is while AMD works its way to offer their Fiji core next year, NVIDIA already seems to be developing their next key performance and flagship chip codenamed GM200. The information on GM200 is grim at the moment but the first obvious target market for NVIDIA would be professional so we can expect a card based on that chip only until late 1H 2014.
AMD Next Generation Fiji and Tonga GPUs Listed:

Link da noticia:
http://wccftech.com/amds-generation-...ction-ramping/

[Winjer]

Venham elas depressa e bem.

[Jorge-Vieira]

Esperemos que venham aí umas grandes graficas, competitivas em todos os aspectos com o rival.