Jump to content
Geeks Nation Forums
Boulard83

Nvidia GTX600 / Kepler / En vente depuis le 22 mars.

Recommended Posts

Sur papier, la GTX670 semble plus performante que la Présente GTX580 .... Plus de core et plus de clock !

Pour les GTX680 et 90 .... plus de 700 cuda cores, clocks élevé ... OUFFF !

Posted Image

Edited by Boulard83

Share this post


Link to post
Share on other sites

Très intéressant! Des GTX660 à prévoir pour moi probablement! Peut-être des 670. On verra si les prix seront au niveau de ces MSRP aussi mais ça semble très intéressant. On a une idée de la date de sortie officielle et de quand on les verra en vente un peu partout?

Share this post


Link to post
Share on other sites

Si tout ceci est vrai.... ce sont effectivement de superbes nouvelles !

Pour les dates de sortie j'avais entendue parler de Q2 si ma mémoire est bonne. Les HD7000 il me semble que ce sera Q1, mais les moyenne et petites en premier. ( encore une fois si ma mémoire est bonne ).

Share this post


Link to post
Share on other sites

Un peu avant l'été. Bah, ça me laisse le temps d'économiser pour les GTX670 alors. Mais, moins chère qu'une GTX570 et plus forte qu'une GTX580? Bien hâte de voir en tout cas, c'est surprenant comme chiffres.

Share this post


Link to post
Share on other sites

Autre news plus fraiche.

GeForce Kepler 104 and 100 GPU Specifications Compiled

A quick stroll through our previous article about how the GeForce Kepler family of next-generation GPUs is laid out, would tell you that GeForce Kepler 104 (GK104), is going to be NVIDIA's answer to AMD's Tahiti. GK104 will be a high-performance (≠ high-end) GPU by NVIDIA that will have many of the features that were reserved for its previous high-end GPUs (such as a 384-bit wide GDDR5 memory interface), but will not be NVIDIA's most powerful GPU in the series. The throne will be kept empty for GK100, which will comply with NVIDIA's "go all in" design ideology for high-end GPUs.

3DCenter.org compiled a few specifications of the GK104 and GK100. They go like this:

GK104

•640 to 768 CUDA cores

•80 to 96 TMUs (depending on what the CUDA core count ends up being)

•384-bit GDDR5 memory interface, 48 ROPs

•Built on the 28 nm TSMC process

•Products based on this will launch in the first quarter of 2012

GK100

•1024 CUDA cores

•128 TMUs

•512-bit GDDR5 memory interface, 64 ROPs

If anything, this specifications disclosure is sure to have an impact on buyers saving up for AMD's HD 7900 series products that are heading toward launch on the 9th of January.

GK100 on parle ici de 1024core ? 2x ce qu'une GTX580 à ? DAMN ! si ça peut être vrai !

Edited by Boulard83

Share this post


Link to post
Share on other sites

NVIDIA Kepler To Do Away with Hotclocks

Since the days of NVIDIA's very first DirectX 10 GPUs, NVIDIA has been using different clock domains for the shaders and the rest of the GPU (geometry domain). Over the past few generations, the shader clock has been set 2x the geometry domain (the rest of the GPU). 3DCenter.org has learned that with the next-generation "Kepler" family of GPUs, NVIDIA will do away with this "Hotclock" principle. The heavy number-crunching parts of the GPU, the CUDA cores, will run at the same clock-speed as the rest of the GPU.

It is also learned that NVIDIA will have higher core speeds overall. The clock speed of the GK104, for example, is expected to be set "well above 1 GHz", yielding compute power "clearly over 2 TFLOPs" (3DCenter's words). It looks like NVIDIA too will have some significant architectural changes up its sleeve with Kepler.

D'autres news et spéculation. J'aime ça ! ! ! !

Shader égal au clock GPU et non x2 mais GPU clock plus élevé ? À suivre ! !

Share this post


Link to post
Share on other sites

GeForce Kepler 104 (GK104) 256-bit GDDR5 Memory Bus - 225W TDP

By Hilbert Hagedoorn, January 20, 2012 - 10:37 PM

Another report has revealed more details on the NVIDIA GK104 GPU (graphics processing unit) based graphico card.This is the chip that will power the Santa Clara, California-based company's mainstream cards, also known as mid-range models.

That means that the GK104 will land on the card, or cards, that will succeed the GeForce GTX 560 Ti.

That is to say, the successor to the GTX 560 Ti will have a performance around the level of AMD's Radeon HD 7900 series. The GK104 has a memory interface of 256 bits and will be backed by up to 2GB of GDDR5 VRAM.

Meanwhile, the TDP (thermal design power) is of 225W.

Needless to say, the implications of this are massive, but one should not be too quick to jump headlong into excitement just yet. But when looking back at the GeForce 400 series being prepared, there was a similar level of hype, with many hoping for a product series that would clearly best AMD's Radeon.

The GTX 680 could arrive as early as February.

Des rumeurs que même la "petite" futur GTX660 approcherait les performances des HD79xx .... J'ai hate de voir ça et surtout de moir les highend !

Edited by Boulard83

Share this post


Link to post
Share on other sites

autre news dans le meme sens.

Geforce GTX 660 / Ti

Nvidia's next-generation 28nm GPU architecture, codenamed Kepler, is officially expected to launch in early Q2 2012 according to the latest schedule we have seen from the company. Although the company will technically be launching one quarter behind AMD, we can still expect to see a few noteworthy design wins in the enthusiast and performance segments, beginning with GK110 in April and following up with GK104, GK107 and GK108.

The latest report from sources overseas claims that Nvidia's GK104 GPU, the successor to 40nm GF114 (see: Geforce GTX 560, Geforce GTX 560 Ti), will feature a 256-bit memory interface, will pack 2GB of memory capacity and should have a 225W TDP. Nevertheless, the GK104 will most likely be branded as Geforce GTX 660 and may very well have a Geforce GTX 660 Ti variant later down the release roadmap.

Many analysts are expecting 28nm GK104 to get a big boost in CUDA cores. Some estimates are projecting as many as 768 CUDA cores or "well above 2 teraflops" of raw performance, which roughly equates to 50-percent more compute power than the current flagship Geforce GTX 580 single-GPU card with its 1.56 teraflops. While we don't have exact information on these specifications yet, we expect more information on the exact details of CUDA cores, texture units and ROPs in the very near future.

Share this post


Link to post
Share on other sites

débile!

très ... les prochaines GTX660ti (mid end) devrait théoriquement être près de 100% plus performante que les 560TI ... !!!!

Share this post


Link to post
Share on other sites

débile!

très ... les prochaines GTX660ti (mid end) devrait théoriquement être près de 100% plus performante que les 560TI ... !!!!

Par curiosité, le renouvellement des gammes se fait une fois par 1-2 ans chez Nvidia si je me souviens ? (Fermi > Kepler)

Merci pour le tableau, le choix de mon ivy et de ma nvidia kepler sera déjà fait et le budget prévisible quoi que ça ne soit pas si nécessaire ... ;oP

Share this post


Link to post
Share on other sites

En mars ça va faire 2 ans que les GTX480 ont vu le jour!

2 ans c'est pas si pire. Chez intel ça fait + de 3ans pour le high end et on parle d'un gain beaucoup + faible que celui qu'on "aurait" chez Nvidia entre le fermi et kepler. Et chez amd CPU, on parle d'une perte de gain avec le bulldoezer, mais ça c'est autre chose. ^^ Ça reste de la spéculation, j'ai hâte de voir si on va s'approcher des chiffres/% qu'on nous à dit dans les rumeurs.

Share this post


Link to post
Share on other sites

le double de Cudacore ne veut pas dire le double de performance mais c'est quand même très représentatif. Une GTX580 512sp est ~45% plus performante qu'une GTX 560 (non TI) de 336sp. J'imagine qu'une carte de 768sp sera logiquement 2x mieux qu'une GTX560 non TI. God damn juste de l'écrire j'en veux 2 !

Share this post


Link to post
Share on other sites

Three cores, seven cards

Nvidia Kepler specs are finally out, thanks to a leak dug up by Lenzfire. We already spoke about some upcoming cards, but the leaked chards now paint a more or less full picture of Nvidia’s upcoming architecture.

The Kepler series will consist of three cores, which will be the building blocks for seven cards, including a dual-GPU card.

The GK106 is a mainstream part and it will power the GTX 640 and GTX 650. The GTX 640 packs 192 stream processors, 16 ROPs, 128-bit memory bus and a memory bandwidth of 88GB/s. The GTX 650 promises to deliver quite a bit more performance, as it has 256 cores and a 192-bit memory bus, enough for 132GB/s worth of memory bandwidth. Interestingly, the GTX 640 will feature 2GB of memory, while the GTX 650 will end up with 1.5GB. The clocks are 850MHz and 900MHz respectively. In terms of performance, Nvidia reckons the new cards will be on par with GTX 550Ti and GTX 560 cards. The price points for GK106 series cards are $139 for the 640 and $179 for the GTX 650

The GK104 will probably be the card of choice for most consumers. This mid range part will also come in two flavours, the GTX 560Ti with a 224-bit bus and the GTX 660 with a 256-bit bus. The GTX 560Ti ends up with 448 cores, 850MHz clock and 154GB/s of bandwidth. It will pack a rather awkward 1.75GB of memory and the price stands at $249. The GTX 660 will feature 2GB of memory and a 900MHz clock, but it will also be significantly pricier at $319. However, Nvidia expects the latter to match the GTX 580 in the performance department, so the price looks pretty good. The GTX 560Ti should be neck and neck with the GTX 570, which sounds pretty good for $249.

As for the big guns, the GK110 will end up in four cards, including the dual-GPU GTX 690. It will cost a whopping $999 and it will launch in Q3, so let’s focus on single GPU cards for the time being. The GTX 660Ti is said to feature an 850MHz clock, 768 stream processors, 1.5GB of memory and a 384-bit bus. It should cost €399 when it launches in Q2/Q3 and Nvidia claims it will be about 10 percent faster than the HD 7950. The GTX 670 ends up with 1.75GB of memory, 896 cores and a 448-bit bus, all for $499. Nvidia pits this one against the HD 7970 and it claims it is about 20 percent faster. The flagship GTX 680 packs 2GB of memory, 1024 cores and a 512-bit bus. Nvidia wants $649 for it, but then again it claims the GTX 680 should beat the GTX 7970 by 45 percent.

Too optimistic? Well only time will tell. The GTX 670 and GTX 680, along with the GTX 660 should launch on April 12, giving AMD a comfortable lead in the meantime. Of course, in-house performance projections should always be taken with a grain of salt, but at this point it seems AMD will lose the performance crown come April. However, with some price cuts AMD should be able to compete with Kepler in terms of value for money. Also, bear in mind that AMD will also introduce its dual-GPU card in early Q2.

http://en.expreview.com/2012/02/06/entire-nvidia-kepler-line-up-unearthed/20836.html

Posted Image

Share this post


Link to post
Share on other sites

NVIDIA GeForce Kepler Packs Radically Different Number Crunching Machinery

NVIDIA is bound to kickstart its competitive graphics processor lineup to AMD's Southern Islands Radeon HD 7000 series with GeForce Kepler 104 (GK104). We are learning through reliable sources that NVIDIA will implement a radically different design (by NVIDIA's standards anyway) for its CUDA core machinery, while retaining the basic hierarchy of components in its GPU similar to Fermi. The new design would ensure greater parallelism. The latest version of GK104's specifications looks like this:

SIMD Hierarchy

•4 Graphics Processing Clusters (GPC)

•4 Streaming Multiprocessors (SM) per GPC = 16 SM

•96 Stream Processors (SP) per SM = 1536 CUDA cores

TMU / Geometry Domain

•8 Texture Units (TMU) per SM = 128 TMUs

•32 Raster OPeration Units (ROPs)

Memory

•256-bit wide GDDR5 memory interface

•2048 MB (2 GB) memory amount standard

Clocks/Other

•950 MHz core/CUDA core (no hot-clocks)

•1250 MHz actual (5.00 GHz effective) memory, 160 GB/s memory bandwidth

•2.9 TFLOP/s single-precision floating point compute power

•486 GFLOP/s double-precision floating point compute power

•Estimated die-area 340mm²

On aime ça les rumeurs.

Share this post


Link to post
Share on other sites

Sheeewiiiiinnng !

GK104 PCB Pictured in Full

Here is the first true-color picture of the GeForce Kepler 104 (GK104) reference PCB shot in full (well, almost, excluding the uneventful PCIe bus connector). The picture provides a panoramic view of the card's VRM as shown in a drawing posted earlier this day, and reveals the strange double-decker power connector. The card is loaded with a 5-phase NVVDC configuration, as detailed in an older article. It also confirms that the GK104 has a 256-bit wide memory interface, with likely 2 GB standard memory amount. This is also the first picture of the GK104 ASIC, which has square package, and somewhat square die. While the PCB is green in color, it's most likely an engineering sample. The final product (branded GeForce GTX 680 / GTX 670 Ti), could have a black-colored one.

Posted Image

Posted Image

Posted Image

Here is the first x-ray drawing of NVIDIA's GeForce Kepler 104 (GK104) reference board, outlining the VRM area. The GPU and memory areas are blanked out for some very obvious reasons. Nevertheless, there's plenty of fascinating stuff going on in these pictures. To begin with, the picture confirms that the board will have 5 NVVDD phases, and up to three miscellaneous power domains. The PCB has provisions for two 6-pin and one 8-pin connector.

The funny part here is a strange new plug that has two 6-pin (or 8-pin+6-pin) stacked, while one of the two 6-pin connector leads are blanked. Some of our sources also report having seen a similar connector with 8-pin and 6-pin on samples of this card (refer to the last picture below). It's not just this, that makes the card incapable of single-slot operation, the DVI connectors over at the display IO also are stacked like on previous-generation AMD Radeon cards. Other connectors on the card are HDMI and DisplayPort. There are two SLI bridge connectors, giving it 3-way and 4-way SLI support.

Edited by Boulard83

Share this post


Link to post
Share on other sites

GeForce GTX 680 Final Clocks Exposed, Allegedly

Waiting on Kepler before making a new GPU purchase? Well, you have to wait a little longer. Thankfully, this wait can be eased with the latest leaks about NVIDIA's 28 nm chip and the GeForce GTX 680 it powers.

According to VR-Zone, the GTX 680 does indeed feature 1536 CUDA Cores and a 256-bit memory interface, but it also has hotclocks, meaning the GPU is set to 705 MHz but the shaders operate at 1411 MHz. The memory (2 GB most likely) is supposed to be clocked at 6000 MHz giving a total memory bandwidth of 192 GB/s.

NVIDIA's incoming card is 10 inches long and also has 3-way SLI support, and four display outputs - two DVI, one HDMI and one DisplayPort. The GeForce GTX 680 is expected to be revealed on March 12 and should become available on March 23rd

1536core @ 705mhz .... damn .... c'est du processing power ça !

Une GTX580 à 512core @ ~800mhz, si Kepler est aussi efficace core pour core on va avoir droit à un boost de 200 à 300% pour à peu près tous les GPU Nvidia ?

Share this post


Link to post
Share on other sites

really nice! j'attends juste de voir les GTX660 un peu plus tard.

les HD7850/7870 sont assez intéressantes de leur côté aussi pour le prix.

Share this post


Link to post
Share on other sites

Si c'est vraiment aussi performant clock pour clock, même la "petite" 660 accotera une 580 actuelle. Si les pris sont correct, perso je crois aller vers des 670.

Share this post


Link to post
Share on other sites

×
×
  • Create New...