Nvidia 700 series Release date ?

[brandon888]

im on my GTX 560 TI 448 cores EVGA FTW .... and want to know when 700 series will come ? i want upgrade to 600 series and waiting for price drops

 

[f1sherman]

I think we recently heard from Ryan Smith "4 months" for 780

 

[brandon888]

I think we recently heard from Ryan Smith "4 months" for 780

oh and one more thin .. there will be same logic as for 600 series ? i mean 780 then 770 and 760 ti and ect .... ?


so 780 expected in april/may ?

 

[f1sherman]

Well 780 is either GK114(refreshed GK104), or mighty GK110. We don't know.

"Spring" he said...

 

[Borealis7]

NV just slashed a bit off the price of the GTX680, so must be clearing stock to make room for 7XX.

http://www.techpowerup.com/178795/NVIDIA-Lowers-Price-of-GeForce-GTX-680.html

 

[brandon888]

NV just slashed a bit off the price of the GTX680, so must be clearing stock to make room for 7XX.

http://www.techpowerup.com/178795/NVIDIA-Lowers-Price-of-GeForce-GTX-680.html

hehe yes but tooo late .... im focused on gtx 670 ^_^ difference is small .... but 670 > 660ti a lot ^_^

 

[boxleitnerb]

I don't think it will be too long now. Maybe not for the lower part refreshes of GK104/106/107, but the "BigK" should come rather sooner than later. Just speculation, though.

 

[brandon888]

I don't think it will be too long now. Maybe not for the lower part refreshes of GK104/106/107, but the "BigK" should come rather sooner than later. Just speculation, though.

and no one knows how good it will be yes ? 15% or 40% faster then 680 ? i think it will be 30% faster at max ;/

 

[boxleitnerb]

If they use GK110, then 40+% faster on average. Doubling GTX580 is just normal for a new generation, I would expect nothing less.

 

[Rvenger]

If AMD doesn't fix their drivers (cough cough FarCry 3) I will be getting a GTX 780.

 

[brandon888]

If AMD doesn't fix their drivers (cough cough FarCry 3) I will be getting a GTX 780.

how so ? i though it's amd's game ... amd optimized i mean ;/


i run on gtx 560 ti 448 cores on high/very high with mediumm post AA and no msaa 60 fps 99% time

 

[Rvenger]

12.11 beta 11 has known artifacting issues in FC3. Vsync doesn't work on all drivers it seems either. I get texture flickering in other games as well as some older titles I get latency issues. It seems that AMD optimized the 12.11 driver for BF3 because that works perfect.

 

[iMacmatician]

I think we recently heard from Ryan Smith "4 months" for 780

"Spring" he said...

Not that I don't believe you (or him), but do you have (a) link(s) to his claims?

 

[Shmee]

still running 6950 crossfire on 12.10, but I would be interested in the next gen series as well, as long as its not just an 8970 Not interested in a rebadge

 

[MisterMac]

I'm hoping for a GTX 770 - based on GK110.

As i am awaiting my 25x14 screen - and would need at minimum that to be keep satisfied for a few years.


Theorecticly - how much optimized could a GK114 be? and how much faster ?

GK104 vs GK114?

10% raw?
20% raw?

...and would it be in raw shaders\ipc or clockspeed upgrades?

 

[brandon888]

i hope 780 would be with at least 320 bits.....

 

[Homeles]

I'm hoping for a GTX 770 - based on GK110.

As i am awaiting my 25x14 screen - and would need at minimum that to be keep satisfied for a few years.


Theorecticly - how much optimized could a GK114 be? and how much faster ?

GK104 vs GK114?

10% raw?
20% raw?

...and would it be in raw shaders\ipc or clockspeed upgrades?

Well, what upgrades could be made? They could push for higher memory clocks, since the maximum GDDR5 operating speed is 7 Gbps; they're currently at 6 Gbps. I think the primary driving force behind this will be the 28nm process. I wouldn't expect big leaps here, but there is potential for improvement. Assuming Nvidia doesn't go out of spec, the best case improvement you'll see with memory clocks is 16.67%.

Let's take a look at memory clock bumps of the past:
5870 vs 4890: 33% increase
6970 vs 5870: 14.58% increase
7970 vs 6970: 0% increase
7970 GE vs 6970/7970: 9.09% increase

Looking at this, AMD's memory gains have stagnated.

580 vs 480: 8.44% increase
680 vs 580: 49.9% increase

And if we take a look at the GTX 680 review here at AnandTech, we can find that it looks like memory clock improvements will be likely be minimal, if they exist at all:

On a final tangent, the memory controllers ended up being an unexpected achievement for NVIDIA. As you may recall, Fermi’s memory controllers simply never reached their intended targets – NVIDIA hasn’t told us what those targets were, but ultimately Fermi was meant to reach memory clocks higher than 4GHz. With GK104 NVIDIA has improved their memory controllers and then some. GK104’s memory controllers can clock some 50% higher than GF114’s, leading to GTX 680 shipping at 6GHz.

On a technical level, getting to 6GHz is hard, really hard. GDDR5 RAM can reach 7GHz and beyond on the right voltage, but memory controllers and the memory bus are another story. As we have mentioned a couple of times before, the memory bus tends to give out long before anything else does, which is what’s keeping actual shipping memory speeds well below 7GHz. With GK104 NVIDIA’s engineers managed to put together a chip and board that are good enough to run at 6GHz, and this alone is quite remarkable given how long GDDR5 has dogged NVIDIA and AMD.

Perhaps the icing on the cake for NVIDIA though is how many revisions it took them to get to 6GHz: one. NVIDIA was able to get 6GHz on the very first revision of GK104, which after Fermi’s lackluster performance is a remarkable turn of events. And ultimately while NVIDIA says that they’re most proud of the end result of GK104, the fact of the matter is that everyone seems just a bit prouder of their memory controller, and for good reason.

It'll be difficult for Nvidia to fix what isn't broken, but memory bandwidth does seem to be one of the bigger potential performance gains for Nvidia.

GK104 doesn't appear to be shader-bound. TMUs are tied to shaders, so I doubt they're texture-bound either.

For big gains, all fingers are pointing at increased ROPs, increased memory bus width, and bandwidth. Perhaps they avoided adding more ROPs by using a crossbar like AMD — unlikely, but it's worth a toss out there. As mentioned earlier, it's unlikely for Nvidia to make meaningful improvements with their memory speeds, but technically possible. Bus width is possible, but if we're looking at a GK114, I think it'd be difficult to position a 384 bit bus GK114 next to the 384 bit bus GK110. There'd be some significant differences — clock speeds, less die "wasted" on FP64, less shaders — but would the differences be enough to justify the jump?

There's one final possibility here: HBM, or high bandwidth memory, using stacked ICs and TSVs. AMD has been pushing this. There's potential for it to show up with HD 8000, and given that it's a third party (Hynix) solution, it's possible that Nvidia could use this as well. Nvidia would probably see the most gain out of it. I haven't the damndest idea if Nvidia will use it, but it does exist. Whether it ends up being cost effective or produced in high enough volume by the time GTX 700 or HD 8000 launches is very much up in the air.

Then there's the obligatory tweaks and tunings that go on that will probably make a few, small % difference. No one outside of Nvidia can really give numbers here.

These are just my observations... take them with a grain of salt.

 

[3DVagabond]

Fermi's memory controller was broke, but Kepler's is not. They won't see much improvement there. The best I've been able to figure is that GK110 will be used in a Kepler card if they need to. That depends on Sea Islands. If they can simply refresh GK104 and be competitive @ $500, I believe that's what they'll do. I don't see them coming out with a hugely changed GK114. GF100 had lots of room for improvement. Just getting the entire chip working, for one.

 

[Homeles]

Fermi's memory controller was broke, but Kepler's is not. They won't see much improvement there. The best I've been able to figure is that GK110 will be used in a Kepler card if they need to. That depends on Sea Islands. If they can simply refresh GK104 and be competitive @ $500, I believe that's what they'll do. I don't see them coming out with a hugely changed GK114. GF100 had lots of room for improvement. Just getting the entire chip working, for one.

That's how I see it, although I imagine we'll see GK110 consumer products eventually, regardless of whether or not AMD has anything interesting. They're going to have to do something with the dies that don't meet the cut.

 

[3DVagabond]

That's how I see it, although I imagine we'll see GK110 consumer products eventually, regardless of whether or not AMD has anything interesting. They're going to have to do something with the dies that don't meet the cut.

Depends on how much of those defective dies need to be fused off. Too much and it won't be appreciably faster than GK104R (GK114?). Might be useful for workstation cards though? AFAIK they haven't released any GK110 workstation cards yet, have they?

 

[Homeles]

Not yet. I think the HPC market is eating them up.

 

[brandon888]

if there will be jump like 400 series to 500 ... i will cry ;/

 

[Homeles]

if there will be jump like 400 series to 500 ... i will cry ;/

At an architectural level, you should be expecting less improvement. It's not hard to improve on garbage. Although... Nvidia has had more time to invest in engineering... GTX 500 launched essentially 8 months after GTX 400, and we're already at 10 months since GTX 600.

 

[boxleitnerb]

GK104 is still a midrange GPU. There is plenty of room for improvement. Just make everything wider

 

[Homeles]

GK104 is still a midrange GPU. There is plenty of room for improvement. Just make everything wider

That's already been addressed... how do you make it wider without encroaching on GK110?