In a typical high-end GPU launch we’ll see the process take place in phases over a couple of months if not longer. The new GPU will be launched in the form of one or two single-GPU cards, with additional cards coming to market in the following months and culminating in the launch of a dual-GPU behemoth. This is the typical process as it allows manufacturers and board partners time to increase production, stockpile chips, and work on custom designs.

But this year things aren’t so typical. GK104 wasn’t the typical high-end GPU from NVIDIA, and neither it seems is there anything typical about its launch.

NVIDIA has not been wasting any time in getting their complete GK104 based product lineup out the door. Just 6 weeks after the launch of the GeForce GTX 680, NVIDIA launched the GeForce GTX 690, their dual-GK104 monster. Now only a week after that NVIDIA is at it again, launching the GK104 based GeForce GTX 670 this morning.

Like its predecessors, GTX 670 will fill in the obligatory role as a cheaper, slower, and less power-hungry version of NVIDIA’s leading video card. This is a process that allows NVIDIA to not only put otherwise underperforming GPUs to use, but to satisfy buyers at lower price points at the same time. Throughout this entire process the trick to successfully launching any second-tier card is to try to balance performance, prices, and yields, and as we’ll see NVIDIA has managed to turn all of the knobs just right to launch a very strong product.

  GTX 680 GTX 670 GTX 580 GTX 570
Stream Processors 1536 1344 512 480
Texture Units 128 112 64 60
ROPs 32 32 48 40
Core Clock 1006MHz 915MHz 772MHz 732MHz
Shader Clock N/A N/A 1544MHz 1464MHz
Boost Clock 1058MHz 980MHz N/A N/A
Memory Clock 6.008GHz GDDR5 6.008GHz GDDR5 4.008GHz GDDR5 3.8GHz GDDR5
Memory Bus Width 256-bit 256-bit 384-bit 320-bit
VRAM 2GB 2GB 1.5GB 1.25GB
FP64 1/24 FP32 1/24 FP32 1/8 FP32 1/8 FP32
TDP 195W 170W 244W 219W
Transistor Count 3.5B 3.5B 3B 3B
Manufacturing Process TSMC 28nm TSMC 28nm TSMC 40nm TSMC 40nm
Launch Price $499 $399 $499 $349

Like GeForce GTX 680, GeForce GTX 670 is based on NVIDIA’s GK104 GPU. So we’re looking at the same Kepler design and the same Kepler features, just at a lower level of performance. As always the difference is that since this is a second-tier card, NVIDIA is achieving that by harvesting otherwise defective GPUs.

In a very unusual move for NVIDIA, for GTX 670 they’re disabling one of the eight SMXes on GK104 and lowering the core clock a bit, and that’s it. GTX 670 will ship with 7 active SMXes, all 32 of GK104’s ROPs, and all 4 GDDR5 memory controllers. Typically we’d see NVIDIA hit every aspect of the GPU at once in order to create a larger performance gap and to maximize the number of GPUs they can harvest – such as with the GTX 570 and its 15 SMs & 40 ROPs – but not in this case.

Meanwhile clockspeeds turn out to be equally interesting. Officially, both the base clock and the boost clock are a fair bit lower than GTX 680. GTX 670 will ship at 915MHz for the base clock and 980MHz for the boost clock, which is 91MHz (9%) and 78MHz (7%) lower than the GTX 680 respectively. However as we’ve seen with GTX 680 GK104 will spend most of its time boosting and not necessarily just at the official boost clock. Taken altogether, depending on the game and the specific GPU GTX 670 has the capability to boost within 40MHz or so of GTX 680, or about 3.5% of the clockspeed of its more powerful sibling.

As for the memory subsystem, like the ROPs they have not been touched at all. GTX 670 will ship at the same 6.008GHz memory clockspeed of GTX 680 with the same 256-bit memory bus, giving it the same 192GB/sec of memory bandwidth. This is particularly interesting as NVIDIA has always turned down their memory clocks in the past, and typically taken out a memory controller/ROP combination in the past. Given that GK104 is an xx4 GPU rather than a full successor to GF110 and its 48 ROPs, it would seem that NVIDIA is concerned about their ROP and memory performance and will not sacrifice performance there for GTX 670.

Taken altogether, this means at base clocks GTX 670 has 100% of the memory bandwidth, 91% of the ROP performance, and 80% of the shader performance of GTX 680. This puts GTX 670’s specs notably closer to GTX 680 than GTX 570 was to GTX 580, or GTX 470 before it. In order words the GTX 670 won’t trail the GTX 680 by as much as the GTX 570 trailed the GTX 580 – or conversely the GTX 680 won’t have quite the same lead as the GTX 580 did.

As for power consumption, the gap between the two is going to be about the same as we saw between the GTX 580 and GTX 570. The official TDP of the GT 670 is 170W, 25W lower than the GTX 680. Unofficially, NVIDIA’s GPU Boost power target for GTX 670 is 141W, 29W lower than the GTX 680. Thus like the GTX 680 the GTX 670 has the lowest TDP for a part of its class that we’ve seen out of NVIDIA in quite some time.

Moving on, unlike the GTX 680 launch NVIDIA is letting their partners customize right off the bat. GTX 670 will launch with a mix of reference, semi-custom, and fully custom designs with a range of coolers, clockspeeds, and prices. There are a number of cards to cover over the coming weeks, but today we’ll be looking at EVGA’s GeForce GTX 670 Superclocked alongside our reference GTX 670.

As we’ve typically seen in the past, custom cards tend to appear when GPU manufacturers and their board partners feel more comfortable about GPU availability and this launch is no different. The GTX 670 launch is being helped by the fact that NVIDIA has had an additional 7 weeks to collect suitable GPUs compared to the GTX 680 launch, on top of the fact that these are harvested GPUs. With that said NVIDIA is still in the same situation they were in last week with the launch of the GTX 690: they already can’t keep GK104 in stock.

Due to binning GTX 670 isn’t drawn from GTX 680 inventory, so it’s not a matter of these parts coming out of the same pool, but realistically we don’t expect NVIDIA to be able to keep GTX 670 in stock any better than they can GTX 680. The best case scenario is that GTX 680 supplies improve as some demand shifts down to the GTX 670. In other words Auto-Notify is going to continue to be the best way to get a GTX 600 series card.

Finally, let’s talk pricing. If you were expecting GTX 570 pricing for GTX 670 you’re going to come away disappointed. Because NVIDIA is designing GTX 670 to perform closer to GTX 680 than with past video cards they’re also setting the prices higher. GTX 670 will have an MSRP of $399 ($50 higher than GTX 570 at launch), with custom cards going for higher yet. This should dampen demand some, but we don’t expect it will be enough.

Given its $399 MSRP, the GTX 670 will primarily be competing with the $399 Radeon HD 7950. However from a performance perspective the $479 7970 will also be close competition depending on the game at hand. AMD’s Three For Free promo has finally gone live, so they’re countering NVIDIA in part based on the inclusion of Deus Ex, Nexuiz, and DiRT Showdown with most 7900 series cards.

Below that we have AMD’s Radeon HD 7870 at $350, while the GTX 570 will be NVIDIA’s next card down at around $299. The fact that NVIDIA is even bothering to mention the GTX 570 is an interesting move, since it means they expect it to remain as part of their product stack for some time yet.

Update 5/11: NVIDIA said GTX 670 supply would be better than GTX 680 and it looks like they were right. As of this writing Newegg still has 5 of 7 models still in stock, which is far better than the GTX 680 and GTX 690 launches. We're glad to see that NVIDIA is finally able to keep a GTX 600 series card in stock, particularly a higher volume part like GTX 670.

Spring 2012 GPU Pricing Comparison
AMD Price NVIDIA
  $999 GeForce GTX 690
  $499 GeForce GTX 680
Radeon HD 7970 $479  
Radeon HD 7950 $399 GeForce GTX 670
Radeon HD 7870 $349  
  $299 GeForce GTX 570
Radeon HD 7850 $249  
  $199 GeForce GTX 560 Ti
  $169 GeForce GTX 560
Radeon HD 7770 $139  

 

Meet The GeForce GTX 670
Comments Locked

414 Comments

View All Comments

  • chizow - Thursday, May 10, 2012 - link

    Hehe ya exactly.

    It seems as if many of the apologists willing to give AMD and Nvidia a pass on 28nm pricing are new to the GPU game, or tech toy game for that matter. They just have no historical perspective at all which I'm sure thrills the marketing/finance guys over in Silicon Valley....they can't sink their meathooks into these guys fast enough.

    But yeah its not about being able to afford it, its about being able to buy them and actually feel good about the purchase looking back, a week, a month, a year from now. Most people only need to be burned once to learn their lesson, hopefully those early adopters who bought 7970/7950 and GTX 680/690 have learned theirs.
  • CeriseCogburn - Friday, May 11, 2012 - link

    Hehe exactly read above, you're wrong. Prove otherwise or shut up. Calling everyone else stupid when you have ZERO EVIDENCE presented doesn't cut it.
  • chizow - Friday, May 11, 2012 - link

    Zero evidence? Try 10 years of GPU benchmarks. Seriously, try looking at some before commenting because its obvious you haven't paid close enough attention in the past....
  • SlyNine - Saturday, May 12, 2012 - link

    You never provide any valid evidence. But this topic has been debated and historical data is all the proof you need.
  • SlyNine - Saturday, May 12, 2012 - link

    I didn't want to spend 500$, but I did want something 2x as fast as my 5870. So the 680GTX got the bill.

    But honestly I wouldn't expect cards to keep evolving at the same rate. Cards used more slots and more power to keep doubling and tripling in performance. That trend cannot go on for long because their is not enough slots and power to do so.

    I fully expect all performance increases now to be from architecture improvements and node changes.
  • CeriseCogburn - Friday, May 11, 2012 - link

    You guys can claim anything you want with your bland, data absent talking point, so let's examine just how far out of sane bounds you two are ( you and chizow ) - and BTW I'd appreciate the reviewers talking point as well. A full quote will be fine.

    Let's skip any insane retentiveness with fancy specific wording you've used as a ruse taken absolutely literally in the hopes that those not noticing a perfectly literal and absolutely strict translation would be fooled by the idea presented, and do a cursory examination:

    We can start with the G80 - it morphed into the G92 and G92b which all you slam artists screamed was a rebranded absolute clone.

    So we'll take the 9800GTX+ vs- the next released card, the GTX280.
    GTX280 morphed into GTX285
    We can move from the GTX285 to the GTX480 - the GTX480 morphed into the GTX580.
    So we move from GTX580 to GTX680.

    Although I have not strictly gone insane talking point ruse literal and used a sort of CHEAT you people espouse with your lousy nm + new die move talking point, what I have is what people actually EXPERIENCED AS CARD RELEASES - so we'll have to go with those.

    9800GTX+ to GTX280 (wow that gigantic upgrade )

    GTX285 to GTX480 ( wow that gigantic upgrade )

    GTX580 to GTX680 ( wow that gigantic upgrade )

    Yes, you people are full of it. That's why you keep AVOIDING any real information and figured if you could spew on just the talking point, no one would have to notice what lying crap it is.
  • chizow - Friday, May 11, 2012 - link

    Once again, your arguments are full of fail or you simply don't know how to read simple benchmarks. Using your own, flawed comparisons, you would see:

    9800GTX+ to GTX280 (wow that gigantic upgrade) +70% OR MORE

    GTX285 to GTX480 ( wow that gigantic upgrade ) +60% OR MORE

    GTX580 to GTX680 ( wow that gigantic upgrade ) +30%......

    The reason your comparison is flawed however is because you are comparing half-generations when you compare a refresh to a new generation, so the gap in both time and performance is diminished which decreases value for your $$$.

    Correct comparisons are as follows, and when you look at it that way, GTX 680 and all other 28nm parts look EVEN worst in retrospect:

    8800GTX to GTX 280: +75% OR MORE
    GTX 280 to GTX 480: +80% OR MORE
    GTX 480 to GTX 680: +40%.....

    or if you prefer refresh to refresh but a full generation between them:

    9800GTX+ to GTX 285: +75% or MORE
    GTX 285 to GTX 580: +80% or MORE
    GTX 580 to GTX 685???: ???

    Seriously just read some benchmarks then come back because it seems you're the only one who doesn't seem to get it.
  • CeriseCogburn - Saturday, May 12, 2012 - link

    For shame for shame - more lies... no wonder you're yelling and you NEVER used benchmarks....
    Let's use anand's historical data !

    And let's do it correctly. We go from the card we have now, to the card they release. People now have the GTX580 - and that's what they see in the charts as you whine, bitch and moan and spread your Charlie D FUD. Likewise in former tier jumps/releases.
    So we will use the TRUTH, not some anal retentive abject cheat worded just so, as you two fools suggest we should, to spin your lies ever more "in your favor".

    9800GTX+ to GTX280 , crysis, 25fps to 34fps

    http://www.anandtech.com/show/2549/11

    There it is chizow and it ain't no 75% ! NOT EVEN CLOSE

    GTX285 to GTX480 , crysisw, 30fps to 44fps

    http://www.anandtech.com/show/2977/nvidia-s-geforc...

    Guess you lost on that 80% lie there, too.

    GTX580 to GTX680, 41fps to 48fps

    http://www.anandtech.com/show/5699/nvidia-geforce-...

    NOPE. Certainly not half of the former two moves, with NONE at any 80%,let alone 75%, not even 50%, can't even say 33% increase, EVER.

    Sorry chizow, your only lies, anf big ones at that, won't do.
  • SlyNine - Saturday, May 12, 2012 - link

    You're cherry picking. A huge fallacy. Some benchmarks do show 75%+

    Plus we are talking about 8800GTX to GTX280. We are not talking about rebaged products with very minor changes.
  • CeriseCogburn - Sunday, May 13, 2012 - link

    ROFL - I used what was first in line, I provided the information - I BROKE THE GIANT LIE you the amd fanboy have used with ZERO EVIDENCE.

    Let's face it, I'm 100% more accurate than you ever attempted to be, merely spewing your talking point in as big a fat fib fashion you could muster.

    Of course that's the usual crap from the liars.
    Now you'll just whine the facts I persented vs the no facts you ever presented or even linked to, "don't count".

    R O F L - loser ( what else do you expect me to do man - you're making it very difficult to support your opinion guy)

Log in

Don't have an account? Sign up now