Geforce Titan - GK110 will be a consumer part

[emertX]

Ooops...your right!

[Tha Last Meal]

Ooops...your right!

np mate

[GoldenTiger]

guys come on it is the same Result from Obr-Hardware That appeared two days before



http://www.obr-hardware.com/2013/02/...l-numbers.html
http://www.xtremesystems.org/forums/...=1#post5169787

Yep, but read the article... it claims they have one in hand and their results were basically the same as OBR's. Is it true? Who knows yet...

[Tha Last Meal]

Yep, but read the article... it claims they have one in hand and their results were basically the same as OBR's. Is it true? Who knows yet...

yeah Who knows, but I do not believe it! Very difficult to achieve the same result is 100%! with another device will appear to have different results by 0.1/1 More or less....Points or FPS.right?

[Handrox]

They may be the source where the OBR took this results ...

[Tha Last Meal]

They may be the source where the OBR took this results ...

obr-hardware put the Result in 2/10/2013
hardware-infos put the Result in 2.11.2013
What do you think?

[Handrox]

obr-hardware put the Result in 2/10/2013
hardware-infos put the Result in 2.11.2013
What do you think?

I think the leak date is indifferent, it can happen if the owner of this material impossible to publish it before ...


Excuse me for my bad english, is not my natural language. I hope you understand me.

[Tha Last Meal]

I think the leak date is indifferent, it can happen if the owner of this material impossible to publish it before ...


Excuse me for my bad english, is not my natural language. I hope you understand me.

yeah maybe.
I understand you 100% np mate

[Olivon]

Special Report: Nvidia GeForce GTX official winning features of Titan - Donanimhaber

First let me say, contrary to the news in the last few days, not 512-bit GeForce GTX Titan will support 384-bit memory bus. GK110 28nm Kepler architecture, parallel processing unit owned 2688 graphics chip graphics card, Nvidia's official documents as the world's fastest single-GPU graphics card goes. So, contrary to some claims GeForce GTX Titan, dual-GPU GeForce GTX 690 is faster, but performance is very close to the values of the information.
GeForce GTX 690 is the same as in the high-tech arena, and thermal conductivity, high strength metal used in the used graphics card equipped with vapor chamber-based cooling is accompanied by 6GB of GDDR5 memory. The performance of texture fill 288 billion / sec (GeForce GTX 690 is 234) we learn that the graphics card 5.4 teraflops of computing power will at the same time.

PCIe 3.0 compliant graphics card with Nvidia's recommended minimum power supply for the 600 Watt.

Astronomical video card is expected to be launched with a retail price, GK110 graphics chip performance / watt performance benchmarks that will capture the single-GPU graphics card for a very long time as is expected to remain as

288Gtexels/s and 5.4Tflops is really something !

[AffenJack]

Yeah, it's a fake. 288Gtex doesn't fit with 5,4 Tflops.

They should learn at least calculating.

[boxleitnerb]

They probably mean 288GB/s bandwidth. Lost in translation most likely.

[prava]

prava 83 -> 118 = 35

but there does not speak of 118fps vs 83fps but 118% - 83%, so I think that you guys are talking about should be used if we speak of frames per second.

Even so 42% is very little ..

1.42 != 42%.

Cmon guys! Math should be strong in enthusiasts!

[GoldenTiger]

Cmon guys! Math should be strong in enthusiasts!

Agreed... and when you're unsure of something, it's never wrong to look it up and learn how it's done.

[Handrox]

Well, I think that 42% said they are incorrect because we speak percent previously stipulated.

The basis of the account is the Asus Ares, all other percentage are made ​​from it and I repeat, do not speak of frames per second, but an estimate of pre conceived.

[Mabyboi]

http://www.techpowerup.com/gpudb/b14...GTX_Titan.html

GPU Name: GK110
Process Size: 28 nm
Transistors: 7,080 million
Die Size: 502 mm?

Render Config
Shading Units: 2688
TMUs: 224
ROPs: 48
SMX: 14

Pixel Rate: 51.2 GPixel/s

Texture Rate: 205.0 GTexel/s

Floating-point: 4,919.04 GFLOPS

Graphics Features
DirectX: 11.1
OpenGL: 4.3
OpenCL: 1.2
Shader Model: 5.0

Released: February 18, 2013
Bus Interface: PCIe 3.0 x16 ?
ASUS Part #: GTXTITAN-6GD5

Clock Speeds
GPU Clock: 915 MHz
Boost Clock: 1019.5 MHz
Memory Clock: 1502 MHz (6008 MHz effective)


Memory
Memory Size: 6144 MB
Memory Type: GDDR5
Memory Bus: 384 bit
Bandwidth: 288.4 GB/s


Reference Board
Slot Width: Dual-slot
TDP: 235 W
Outputs: 2x DVI
1x HDMI
1x DisplayPort

The part that I'm happiest about is that it still shows 6GB of VRAM on there.. But I need a mini display port

[boxleitnerb]

Fake. 502mm2 is wrong, that gives it away. Also at these clocks, 235W TDP is outright impossible.

[zalbard]

The part that I'm happiest about is that it still shows 6GB of VRAM on there.. But I need a mini display port

Pretty sure it will be miniDP. Otherwise it will not fit. Besides, in either case it's just a matter of buying an inexpensive adapter...

[highoctane]

Fake. 502mm2 is wrong, that gives it away. Also at these clocks, 235W TDP is outright impossible.

Why would 502mm2 be so wrong, gtx580 was around 520mm2 and gtx680 is 294mm2. Going by gtx680 size Titan is not a doubling of 680 specs and neither should the size be so 502mm2 seems like a realistic assumption to me.

GTX580 with a 520mm2 die was a 244w TDP rated part stock clocked at 772mhz core, 235w TDP for around a 500mm2 Titan core shoehorns within the general dimensions and TDP.

[Iconyu]

Why would 502mm2 be so wrong, gtx580 was around 520mm2 and gtx680 is 294mm2. Going by gtx680 size Titan is not a doubling of 680 specs and neither should the size be so 502mm2 seems like a realistic assumption to me.

GTX580 with a 520mm2 die was a 244w TDP rated part stock clocked at 772mhz core, 235w TDP for around a 500mm2 Titan core shoehorns within the general dimensions and TDP.

The 780GTX might be 502mm2, but Titan is GK110, these chips have been available for nearly a year, there are xrays of them online and you don't miss that much difference in size. Either Titan is GK110 and the size on the chart is wrong, or it isn'tGK110 and the chip on the chart is wrong.

[PedantOne]

Info from TPU and Donanimhaber is correct. Launch date is correct too. I seen Three Titans today (it looks better then GTX 690, too sexy boards), few Magazines has cards for review.

Performance is few percents under GTX 690, numbers from OBR are correct.

[GoldenTiger]

Info from TPU and Donanimhaber is correct. Launch date is correct too. I seen Three Titans today (it looks better then GTX 690, too sexy boards), few Magazines has cards for review.

Performance is few percents under GTX 690, numbers from OBR are correct.

[Loque]

Is it a waste to buy that card for my system ? I can oc the cpu to a stable 4.4ghz if needed.. will use the card for 3 1080p monitors.

[[XC] Oj101]

The link I posted is two days older than OBR and has approximately the same score

[tajoh111]

Info from TPU and Donanimhaber is correct. Launch date is correct too. I seen Three Titans today (it looks better then GTX 690, too sexy boards), few Magazines has cards for review.

Performance is few percents under GTX 690, numbers from OBR are correct.

IF this is correct. This could scare AMD away from the high end or the GPU compute professional market, if the price of these cards start settling down. Unless AMD starts making monster cards like Nvidia, they might only just match this card at 20nm and at that point, that would only equal a disastrous launch for a next, next generation card(50% is nothing to scoff at as a improvement within the same manufacturing process). AMD could beat this card with a smaller card, but it will have to give up alot of GPU compute abilities thus surrender a lot of potential application performance, like they did in the old days.

AMD probably doesn't want to make a big die because AMD doesn't have enough reputation or sales in the pro market to make it worth it considering the R and D and the lost wafers on other products.

Although we don't have much data or nothing is certain at this point, paper specs really paint the geforce titan to be a monster card based on what we know how much die size and components it has over gk104. Unlike the ridiculous 6970 hype which had people saying 60 percent performance jumps on the same manufacturing process on a card that was just a little bigger.

http://www.xtremesystems.org/forums/...umor%29/page17

[Dimitriman]

Info from TPU and Donanimhaber is correct. Launch date is correct too. I seen Three Titans today (it looks better then GTX 690, too sexy boards), few Magazines has cards for review.

Performance is few percents under GTX 690, numbers from OBR are correct.

If this is true they might as well EOL Gtx 690.