No GK110. Instead GK114 will be the GTX 780

spoffle · 6 Nov 2012 at 00:40

Why is there no doubt? Planned designs, and what they can manufacture I'd say are two very different things.

Meaker · 6 Nov 2012 at 00:42

humbug said:
True, there are other variables. like bOrking the TIM to use your example.

Same goes for notebooks too, the same heat output on a smaller die has made some cooling systems struggle as they can't take the heat packed into a smaller area.

IC3 · 6 Nov 2012 at 00:42

TheRealDeal said:
There is no doubt the gk104 (gtx680) was meant to be for mid-high end. Its not been a secret that nvidia have pushed the boundaries of what's been possible to manufacter in there last few generations. The gtx680 is a step back in a few ways comparing with gtx4-5 and obviously not in gaming performance. The fact that gk102 was said to be cancelled and gk112 might not be heading our way points towards nv struggling to manufacture it.

I will add though i am just saying what i read on different sites and in no way understand it fully.

This is like Politics :rolleyes:

You just know, what they want you to know.
I think 5xx series was one of the best bang for buck that Nvidia have ever produced. Its only my personal opinion.

TheRealDeal · 6 Nov 2012 at 00:46

spoffle said:
Why is there no doubt? Planned designs, and what they can manufacture I'd say are two very different things.

Look at gtx4-5 series which were named gf102 and gf112. They were the top end parts gf102 was gtx480 and gf112 was gtx580. So from my logic gk102 should have been gtx680 and gk112 should or still might be gtx780. Gf102 (gtx480) was meant to be a fully fledged gtx580 maybe with gtx480 clocks but nvidia had troubles manufacting it, if what was said at the time was true. If it did come out in all of its glory it would have been a monster part at the time yet it came out a flop in most peoples eyes.

TheRealDeal · 6 Nov 2012 at 00:48

IC3 said:
This is like Politics
You just know, what they want you to know.
I think 5xx series was one of the best bang for buck that Nvidia have ever produced. Its only my personal opinion.

The 5 series was a good part but you can bet nvidia hated the publicity the 4 series got.

spoffle · 6 Nov 2012 at 00:57

TheRealDeal said:
Look at gtx4-5 series which were named gf102 and gf112. They were the top end parts gf102 was gtx480 and gf112 was gtx580. So from my logic gk102 should have been gtx680 and gk112 should or still might be gtx780. Gf102 (gtx480) was meant to be a fully fledged gtx580 maybe with gtx480 clocks but nvidia had troubles manufacting it, if what was said at the time was true. If it did come out in all of its glory it would have been a monster part at the time yet it came out a flop in most peoples eyes.

I couldn't say that was more than conjecture though despite the fact that I see people reporting it as fact.

TheRealDeal · 6 Nov 2012 at 01:06

spoffle said:
I couldn't say that was more than conjecture though despite the fact that I see people reporting it as fact.

I just look at whats out there and try to make some sense of it. Nvidia moving backwards in my opinion is not one of there policy's. Nvidia in my opinion would love to do what intel are doing to amd on the cpu side but what amd have on there side is nvidia have to work in the same manufacturing process as amd so its more of a fair head to head. If nvidia could have with there gk102 stuck the boot into amd i am pretty sure they would have. I just don't think it was viable to. Business is a war at the end off the day.

Gregster · 6 Nov 2012 at 11:09

Ever since NVIDIA first showed us Kepler in its full form as the Tesla K20, people have been asking about a GeForce card based on this silicon, GTX 780. However it was then speculated by OBR-Hardware that there would most likely not be a GeForce GTX 780 GK110 based card. Since then we have heard little, if anything about what GK11x derivative would see the light of day.

SemiAccurate is citing two sources as naming the “big gaming GPU”, meaning the 780, from NVIDIA as the GK114 and the GK114-GX. Similar to AMD, this chip is expected to be late and not see a release until the earliest being March 2013, but S.A. believes April or May is much more likely. Also like AMD, the jump in performance is not supposed to be mind boggling; partially due to die size restraints and in my opinion this is to be expected with Maxwell now on the horizon. If NVIDIAs next GPU release completely demolished the GTX 680, then the market enthusiasm for Maxwell would shrink dramatically when the time comes. So you have Maxwell playing a role most likely in expected performance and you also have AMD, which we wrote on yesterday, expected to jump only by 15%.

All of these factors play into the expected performance of GK114, however unlike AMD, NVIDIA has more room to play in the TDP department. This is due to the highly efficient architecture of Kepler. According to S.A. “those who have been briefed” are also saying NVIDIA is expected to have the same 15% jump in performance over the GTX 680 as was reported on AMD. So basically the next generation of silicon from both NVIDIA and AMD is only expected to gain 15% performance increase over their predecessors. However according to our sources the increase for the GTX 780 would be around 25% over the GTX 680, instead of the 15% S.A. is citing. This may not be the series launch you will tell your grandkids about, but may remind you more of the G92 vs G92b period. The GTX 600 Series was restrained heavily by its small 256-bit Bus, I believe NVIDIA could achieve this 15% increase simply by giving it the bus width seen on Fermi cards, instead of the Bus we saw on the FX5900 Ultra.

Keith H.

http://videocardz.com/35441/nvidia-geforce-gtx-780-to-be-based-on-gk114-gpu

This guy talks sense

Sea Islands Performance Hit By Tropical Storm or Hurricane.

AMD’s upcoming GPU line, Sea Islands (successor to Southern Islands), have been speculated on across many hardware sites since the end of the 7900 series. First, what we are almost certain of is this series of GPUs will not be released in 2012, rather it is being reported as March 2013 by SemiAccurate.com.

Now comes the interesting new information, AMD promised AIBs an increase of around 30% vs. their current Southern Islands generation. However SemiAccurate is reporting this amount of performance increase to be a rather radical claim as they are citing “more measured sources” as showing only a 15% increase in performance. Users with the top of the line AMD product, the 7970 Gigahertz Edition, have seen the large power draw Tahiti XT uses when OC’d. This can also be seen by the lack of any 7990 by AMD themselves; instead AIBs have created their own 7990, all of which require 3x 8 Pin power connections. This makes AMD’s next iteration most likely power bound (as S.A. says) and obviously, just like any silicon based hardware manufacturer, die size bound by cost.

These two pieces of information solidify the final the rumor, which is that the number of Stream Processors will only be increased slightly. As I’m sure many of you know, increase of SPs not only increase TDP but also increases the mm^2 of the die. As we get closer to the launch of the product I’m sure we will get more details on how solid these rumors are or are not; they do call themselves Semi Accurate for a reason.

Written by Keith H.

http://videocardz.com/35422/sea-islands-to-feature-only-15-performance-increase

Don't forget that no solid information is out yet, so rumours will be rife till release.

Jakus · 6 Nov 2012 at 13:13

What is all this TDP rubbish ?

From what I see it's a number magicked up by manufacturers, with a free hand to estimate their components design power limits....how high/low they wish to set that has little bearing regarding efficiency comparisons to any other manufacturers numbers unless the components are exactly like for like :rolleyes:

Gregster · 6 Nov 2012 at 13:27

JAKUS said:
What is all this TDP rubbish ?

From what I see it's a number magicked up by manufacturers, with a free hand to estimate their components design power limits....how high/low they wish to set that has little bearing regarding efficiency comparisons to any other manufacturers numbers unless the components are exactly like for like

I am sure you could be more accurate with your next GPU build. :rolleyes:

LeJosh · 6 Nov 2012 at 13:43

JAKUS said:
What is all this TDP rubbish ?

From what I see it's a number magicked up by manufacturers, with a free hand to estimate their components design power limits....how high/low they wish to set that has little bearing regarding efficiency comparisons to any other manufacturers numbers unless the components are exactly like for like

...

http://en.wikipedia.org/wiki/Thermal_design_power

CAT-THE-FIFTH · 6 Nov 2012 at 14:16

gregster said:
snip

Considering the other GCN cards are competitive with Kepler in regards to die size,performance and performance/watt,it will be interesting to see how well the Pitcairn successor does.

Jakus · 6 Nov 2012 at 15:52

LeJosh said:
...

http://en.wikipedia.org/wiki/Thermal_design_power

Quote from the wiki link.....

"TDP values between different manufacturers cannot be accurately compared. While a processor with a TDP of 100 W will almost certainly use more power at full load than a processor with a 10 W TDP, it may or may not use more power than a processor from a different manufacturer that has a 90 W TDP. "

Is that not basically what I said ?

I posted because there was inaccurate discussion

about it !

Marine-RX179 · 6 Nov 2012 at 15:57

Seriously...if a pitiful 15% increase in speed per gen is becoming the norm, then both companies are seriously taking a mick of the consumers...

flopper · 6 Nov 2012 at 16:02

Marine-RX179 said:
Seriously...if a pitiful 15% increase in speed per gen is becoming the norm, then both companies are seriously taking a mick of the consumers...

TDP and engineering sets its limits.

Gregster · 6 Nov 2012 at 16:04

Regardless of what will be better next gen for me is irrelevant. What worries me is all this volt locking that has crept into the current GPUs. I will have zero interest if this is the new standard. Me and soldering irons don't agree

but I will get some practice in if this is the way it will be.

Marine-RX179 · 6 Nov 2012 at 18:10

flopper said:
TDP and engineering sets its limits.

Then sell the cards at lower price rather than demanding the same premium like the old days on 80-100% speed increase.

Rather than at £400+, £320 should be max for the top-end single GPU card.

FoxEye · 6 Nov 2012 at 18:15

Marine-RX179 said:
Then sell the cards at lower price rather than demanding the same premium like the old days on 80-100% speed increase.

Rather than at £400+, £320 should be max for the top-end single GPU card.

The thing is, and we can't escape this, there are plenty of people willing to pay those prices. Even if it turns out to be a 15% increase, they will still sell loads at those premium prices.

Stanners · 6 Nov 2012 at 18:17

flopper said:
TDP and engineering sets its limits.

Is it not the engineering that has set the limits for TDP and speed increase at every new generation!

Tonester0011 · 6 Nov 2012 at 18:20

Marine-RX179 said:
Then sell the cards at lower price rather than demanding the same premium like the old days on 80-100% speed increase.

Rather than at £400+, £320 should be max for the top-end single GPU card.

You have two choices, wait X amount of months for the prices to drop or don't upgrade....

This gen was only rumored to be 15% increases, it's more like 60%...until the cards drop and prices are announced, try to get a little less angry :rolleyes: