GTX680 reall a 660Ti?

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Qbah

Diamond Member
Oct 18, 2005
3,754
10
81
Is the OP serious? Really? Since when was a (supposedly) mid-range card of new-gen FASTER than last gen's high-end? Sure, the jump is not as big as 7800GTX -> 8800GTX -> GTX280 -> GTX480, but come on...

You get more performance than last gen's fastest, with less power usage at the same price point ($499 MSRP).

nVidia not releasing a (rumored) card they could have but didn't is pure comedy.
 

Lepton87

Platinum Member
Jul 28, 2009
2,544
9
81
Is the OP serious? Really? Since when was a (supposedly) mid-range card of new-gen FASTER than last gen's high-end? Sure, the jump is not as big as 7800GTX -> 8800GTX -> GTX280 -> GTX480, but come on...

They could have easily done the same with GF104 if they went all out with clocks and shader count.(GF114) Fortunately G100 didn't flop as bad as bigK did. You really think nvidia proper high-end would be stripped of compute? Why would they make a high-end card just for the gaming market and leave HPC out? Does not make any sense.

http://www.anandtech.com/bench/Product/520?vs=547

Beats previous flagship just as much as GTX680 beats 580
 
Last edited:

Rvenger

Elite Member <br> Super Moderator <br> Video Cards
Apr 6, 2004
6,283
5
81
The only thing I see Bigk having is much higher power consumption and good compute performance. I bet the gaming performance improvement is minimal except for more gains in high resolutions.
 

Denithor

Diamond Member
Apr 11, 2004
6,298
23
81
The only thing I see Bigk having is much higher power consumption and good compute performance. I bet the gaming performance improvement is minimal except for more gains in high resolutions.

I tend to agree with this, and would note that once these cards arrive in 4GB variants the high end picture should improve noticeably.

:)
 

Rvenger

Elite Member <br> Super Moderator <br> Video Cards
Apr 6, 2004
6,283
5
81
I tend to agree with this, and would note that once these cards arrive in 4GB variants the high end picture should improve noticeably.

:)


Yep and we will see if the 256bit bus width actually bottlenecks the GPU. I really think not.
 

StrangerGuy

Diamond Member
May 9, 2004
8,443
124
106
A supposedly midrange GPU destroying the whole AMD lineup?

Personally that is amazing, even it is $500.
 

rusina

Member
Mar 20, 2012
31
0
66
Is the OP serious? Really? Since when was a (supposedly) mid-range card of new-gen FASTER than last gen's high-end? Sure, the jump is not as big as 7800GTX -> 8800GTX -> GTX280 -> GTX480, but come on...

You get more performance than last gen's fastest, with less power usage at the same price point ($499 MSRP).

nVidia not releasing a (rumored) card they could have but didn't is pure comedy.
Well last time when Nvidia launched next gen architecture with new manufacturing method mid range-gpu was faster than anything from last generation. This was with ultra perfect Geforce FX-line up.

If you look at GK104..what makes you think that it's high end? It's 20% smaller than GF114 (Geforce GTX 560 ti), 256-bit bus, crippled dp performance..

If you look at PCB of GTX 680 you can see that Nvidia could have gone further with it's performance. You can see that they were preparing for 6pin + 8pin and much heavier power supply..obviously for higher clocks.

..but yeah rumor was that GK100 was fundamentally flawed and Nvidia decided to concentrate on GK110. Several news sites are suggesting that Nvidia would demonstrate GK110 in May
---
On the other hand the whole Kepler-architecture seems like it's been developed from GF104/GF114 structure.. not GF100/GF110
 
Last edited:

evilspoons

Senior member
Oct 17, 2005
321
0
76
If nvidia had a more powerful card ready they would have released it. I can start talking about 2020 Corvettes right now to prevent buyers from buying current gen Vipers under the same theory.

Hi, I'm here to be pedantic. There is no current-generation Viper, it is currently out of production. :D Incidentally, it will be a "SRT Viper" when it comes back, not a "Dodge Viper"

Carry on...

To continue the car analogies, actually, let's say we have something reasonably fast like, say, a Mustang GT, traditionally powered by a V8.

Chevy steps in with some new Camaro, originally intending for it to be powered by a V8 as well. The V8 is delayed, and instead they have to cram a V6 from a different car in there. But they do a lot of work on it because it's their only engine. They manage to come out with a Camaro that's faster than the Mustang even though it's "just" a V6. Is that a problem? It's still faster. Then the V8 finally comes out. Hell, it might end up being worse, because the V8 is heavier. Why bother releasing the V8 at all at this point?
 

Paul98

Diamond Member
Jan 31, 2010
3,732
199
106
We won't know for sure what part this was expected to target till we see what comes out in the next 6 months to a year. It would have been nice to see this part come out before AMD then they would have put it where they expected it to go with respect to AMD. Then AMD would have had to reprice their chips. I think the lack of huge bump in AMD performance along with the high prices really is going to screw up these prices for the short term.

But then again look at Nvidia's old offerings, they have come way down in price.
 

notty22

Diamond Member
Jan 1, 2010
3,375
0
0
Hi, I'm here to be pedantic. There is no current-generation Viper, it is currently out of production. :D Incidentally, it will be a "SRT Viper" when it comes back, not a "Dodge Viper"

Carry on...

To continue the car analogies, actually, let's say we have something reasonably fast like, say, a Mustang GT, traditionally powered by a V8.

Chevy steps in with some new Camaro, originally intending for it to be powered by a V8 as well. The V8 is delayed, and instead they have to cram a V6 from a different car in there. But they do a lot of work on it because it's their only engine. They manage to come out with a Camaro that's faster than the Mustang even though it's "just" a V6. Is that a problem? It's still faster. Then the V8 finally comes out. Hell, it might end up being worse, because the V8 is heavier. Why bother releasing the V8 at all at this point?

LOL, I had thought of a similar car analogy. And it actually happened. When the Pontiac made some Trans Ams with a Buick V6 Turbo GN engine.
I'm pretty sure it was one of the fastest TA's that had been produced since the 455SD days when they built it.
 

OCGuy

Lifer
Jul 12, 2000
27,224
36
91
GK104 is about 20% smaller than GPU on Geforce GTX 560 ti. My personal opinion is that this doesn't indicate that it would be high end.

Huh? Since when is "high end" determined by die size? It is the fastest GPU available, that alone makes it "high end".

I think people are grasping at straws here. Probably the same people that complained about Fermi 1.0 being big, loud, and hot.
 

evilspoons

Senior member
Oct 17, 2005
321
0
76
GK104 is about 20% smaller than GPU on Geforce GTX 560 ti. My personal opinion is that this doesn't indicate that it would be high end.

Well... we're now seeing a 28 nm GPU vs a 40 nm GPU. If the design were exactly the same as the 560 ti, it would still be significantly smaller due to feature size change... on the order of 50%. So... you've got a chip that SHOULD be half as big but is only 20% smaller, meaning there's a whole bunch of extra stuff.
 

Lepton87

Platinum Member
Jul 28, 2009
2,544
9
81
Well... we're now seeing a 28 nm GPU vs a 40 nm GPU. If the design were exactly the same as the 560 ti, it would still be significantly smaller due to feature size change... on the order of 50%. So... you've got a chip that SHOULD be half as big but is only 20% smaller, meaning there's a whole bunch of extra stuff.

Your argument is ridiculous. Successor doesn't mean a straight die-shrink. We always get more transistors in each market segment but according to your twisted logic we shouldn't?
 

BallaTheFeared

Diamond Member
Nov 15, 2010
8,115
0
71
Well... we're now seeing a 28 nm GPU vs a 40 nm GPU. If the design were exactly the same as the 560 ti, it would still be significantly smaller due to feature size change... on the order of 50%. So... you've got a chip that SHOULD be half as big but is only 20% smaller, meaning there's a whole bunch of extra stuff.

I'm not following the reasoning...

It's a 560ti x2, not just a 560 ti shrunk.

http://www.anandtech.com/show/5699/nvidia-geforce-gtx-680-review/2

There are some changes to Fermi with Kepler, but Kepler is still based on Fermi.
 

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
We agree with everything you are saying...all we are trying to say is this card was not supposed to be nv top end card.

What the heck are they going to call the real 680 when it comes out?if i was nv i would of called it a 660ti just to bust amds ballz and to let them know they have a monster waiting.

The op asked if its a relabled gpu and it is 100% not a real 680 just called one because it beats amds top end gpu.

Dont forget its on a 256bit bus also

Why do you keep trolling this point over and over? The current 680 IS the 'real 680' because it IS the 680. Clear enough?

The 680 is whatever NV wants it to be. Maybe they wanted a teddy bear to be the 680, and that didnt work out? What does that mean to you or me? Absolutely nothing.

The fact is that the current 680 IS the fastest card you can buy out of the box, period. When NV releases a new card later this year (presumably) maybe that is the 690? Maybe the 685? Maybe the 780? Who knows.

I don't get the people here. Constant complaints about NV's 'big die' strategy and power usage vs. AMD. NV responds and releases a MASSIVELY more efficient GPU and then folks complain it was 'never meant to be this good'. Who the F cares? Look at the measurable features; power usage, performance, and cost. It wins on all fronts.

I expect the 8970 (or whatever gets released by AMD next) to probably match-up with the next NV GPU. I hope NV produces the 'big die' kepler because many folks here love all-out performance. I like having choices. Choices are GOOD. Small, efficient (yet still fast) or larger and less efficient (but really fast).

NV has finally offered a pure gaming GPU that is efficient for the masses (hint: most people DONT use compute functions) and will later offer a great option for both gaming and compute.

AMD's offerings likely fall in-between, which is good too.

The new Breakdown:
Just for gaming - buy NV
Gaming plus some compute - buy AMD
Lots of compute + some gaming - NV/AMD (depending on what compute functions you need)

I love that breakdown. Looks like a variety of choices that hopefully will encourage competition.
 

96Firebird

Diamond Member
Nov 8, 2010
5,714
316
126
Here's an idea... Stop thinking of video cards as generational products. Instead, just see them for what they are at the time they are released, and watch them trickle down in time.
 

rusina

Member
Mar 20, 2012
31
0
66
Well... we're now seeing a 28 nm GPU vs a 40 nm GPU. If the design were exactly the same as the 560 ti, it would still be significantly smaller due to feature size change... on the order of 50%. So... you've got a chip that SHOULD be half as big but is only 20% smaller, meaning there's a whole bunch of extra stuff.
Yes..that would be direct die shrink. Obviously GTX 660 was supposed to be more than that.
 

evilspoons

Senior member
Oct 17, 2005
321
0
76
I'm not following the reasoning...

It's a 560ti x2, not just a 560 ti shrunk.

http://www.anandtech.com/show/5699/nvidia-geforce-gtx-680-review/2

There are some changes to Fermi with Kepler, but Kepler is still based on Fermi.

I know that! :p

Someone else was complaining that it's a low-end part since its' 20% smaller than a 560ti. I'm saying that no, it isn't, because if it were just a 560 ti, it would be like 50% smaller. But it's only 20% smaller, which means there's a whole bunch of new transistors, meaning it can easily be a high-end part.
 

Ajay

Lifer
Jan 8, 2001
16,094
8,109
136

^This. But it may have started out as being an X60 class GPU. Basically the pricing rumors went from $300 to $400 and finally the card came out at $500.

Makes me wonder if the 'yield problem' then NV had wasn't so much the number of active SMXs, but the clock speed. So as production ramps we could see GPUs binned into lower clock groups and have our 670 Ti that way. Down the road as the process matures (and maybe NV does a re-spin: GTX 760 Ti maybe?) then some the number of SMXs may change. Just IMHO.
 

Zebo

Elite Member
Jul 29, 2001
39,398
19
81
It's bullshit. nV took their best newest most efficient design, the 560ti, and grew upon it. Just like AMD will take their best, newest , most efficient design, the 7870 and grow upon it. Are we going to say the 8900 is meant to be mid range chip when that happens?


Close.
 

Puppies04

Diamond Member
Apr 25, 2011
5,909
17
76
I do hope that everyone complaining that it should be called a 660 TI (and priced around the 560TI mark) actually realise that if nvidia did do this it would basically kill AMDs high end graphics division.

Look at what happens to the high end cpu market now AMD has basically pulled out if you need reasons as to why this would not be good for consumers.

IF, and that is a big if, NVIDIA actually had a faster part that was originally going to be called the 680 then so what? It wasn't released and I defy anyone to show me a big company that given this hypothetical situation wouldn't do exactly what NVIDIA is accused of doing.
 

Qbah

Diamond Member
Oct 18, 2005
3,754
10
81
They could have easily done the same with GF104 if they went all out with clocks and shader count.(GF114) Fortunately G100 didn't flop as bad as bigK did. You really think nvidia proper high-end would be stripped of compute? Why would they make a high-end card just for the gaming market and leave HPC out? Does not make any sense.

http://www.anandtech.com/bench/Product/520?vs=547

Beats previous flagship just as much as GTX680 beats 580

So they did a better refresh on the GTX460 than they did on the GTX280 - how does that prove anything related to a generation leap?

AMD/ATi:
X700 vs 9800XT - nope
X1600 vs X850XT - nope
HD2600XT vs X1950XTX - nope
HD4850 vs HD3870 - yes... here's the only one for AMD/ATi (HD48xx was a killer product)
HD5770 vs HD4890 - nope (no refresh here), same for HD5830
HD7850 vs HD6970 - nope - jacked up pricing too on the 7-series

nVidia:
6600GT vs FX5900Ultra - yes... here's the only one for nVidia (FX series was a total failure)
7600GT vs 6800Ultra - nope (no refresh here)
8600GT vs 7900GTX - nope
GTS250 vs 9800GTX - those are the same cards... D:
GTX460 vs GTX285 - nope
GTX660Ti(?) vs 580GTX - can't say yet - wouldn't hold my breath though!
 
Last edited:

Smoblikat

Diamond Member
Nov 19, 2011
5,184
107
106
Huh? Since when is "high end" determined by die size? It is the fastest GPU available, that alone makes it "high end".

I think people are grasping at straws here. Probably the same people that complained about Fermi 1.0 being big, loud, and hot.

I am not complinging about the 400 series. I love my 470, its quiet, cool, and uses minimal power when not gaming.
 

Smoblikat

Diamond Member
Nov 19, 2011
5,184
107
106
It's bullshit. nV took their best newest most efficient design, the 560ti, and grew upon it. Just like AMD will take their best, newest , most efficient design, the 7870 and grow upon it. Are we going to say the 8900 is meant to be mid range chip when that happens?


Close.

If their high end chip was origonally going to be their mid range chip......then yes i WILL say that the 8900 should be the mid range chip. That only makes sense