review nvidias geforce gtx 690 graphics card

Review of Nvidia GeForce GTX 690 Graphics Card


This article is a review of the Nvidia GeForce GTX 690 graphics card. It includes photos, tables, graphs, information, costs, game tests, and more.

The Review of Nvidia GeForce GTX 690

Man, I have to say, thank goodness for tablet computers, smart phones, Xboxes, and whatever else is distracting the masses from PC hardware. The rise of those other devices is supposed to be siphoning off interest from the personal computer, and on some levels, that must be true. There was a time when, as a guy who wrote for TR, I could make conversation with friends and relations about the latest Pentium or Radeon or whatever. These days, those conversations are all about iPads and such instead.

At the same time, it appears those of us still paying attention to, you know, the most powerful consumer computing platform are living in some sort of a magical future-land where our most persistent gripes have been replaced by difficult choices between multiple amazing options. Want a quiet case? Easily done. Want a case capable of housing powerful hardware? Easily had, as well. Want a case that’s both at once? Also readily available. Need a power supply? This one’s modular and makes zero noise at idle. Lousy keyboard got you down? Here, take your pick from ten different mechanical offerings with four different switch types. This one will massage your fingertips as you type.

Decent Computer Parts

Decent computer parts can still be had for cheap, but if you want to pay more in order to get something that’s higher quality, the choices now are better than ever. Component makers are increasingly catering to the desires of PC hobbyists, and frankly, we could get used to it. Already are used to it, really, except when we’re occasionally surprised by another nifty example of the trend.

We were a little startled recently when we received a package containing nothing but the prybar you see above. No, as far as we know, this isn’t a new motion controller for Left 4 Dead, although that would be awesome. Instead, it’s just a big, metal implement. We weren’t sure what to make of it until several days later, when the following arrived:

Without the prybar, I would have surely chipped a tooth trying to get that crate open, so thank goodness.

Inside was the subject of our attention today, the GeForce GTX 690. Yes, this is the new uber-high-end graphics card from Nvidia that packs two of the GK104 graphics processors found in the GeForce GTX 680. You can probably tell from the picture above that the GTX 690 looks a little different from your typical graphics card. Here are a couple more shots, closer up, to give you a better sense of things.

Yes, the GTX 690 looks distinctive. What may not be obvious from the pictures is that the card’s sleek lines and metallic color palette are not a plasticky imitation of something better, as one might expect given the history here. Instead, this premium graphics card is built with premium materials. The silver-colored portions of the cooling shroud are chromium-plated aluminum, and the black piece between them is made of magnesium. Beneath the (still plastic) windows on either side of the cooling fan, you can see the nickel-plated fins of the card’s dual heatsinks. Oh, and there’s a bit of a light show, too, since the green “GeForce GTX” lettering is LED-illuminated.

Touch a fingertip to the cool, solid surfaces of the GTX 690, and you can feel the extra expense involved. Use that fingertip to give the cooling fan a spin, and you’ll feel it there, too, in a motion that’s perceptibly smoother than most. No expense was spared on the materials for this card, and it shows in little ways that, we’ll admit, not everyone will appreciate. We can’t help but like it, though. In terms of look and feel, if the GTX 690 has a rival among current video cards, it may be XFX’s aluminum-shrouded Radeons. But you’ll need two of those in order to approach the GTX 690’s performance.

Nvidia tells us it has invested heavily in tuning the acoustics of the GTX 690’s cooler, as well. Beyond the fancy fan mechanism, the base plate beneath the fan features ducted air channels. Mounted on the board are very low profile capacitors, intended to reduce turbulence in the air flowing across the heatsinks. Time constraints have kept us from disassembling our GTX 690 yet, but below are a couple of stock pictures of the areas in question. As you can see, the GTX 690’s cooler is designed to send air flowing in two directions: half toward the back of the card and outside of the case, and half toward the front of the card, into the PC enclosure.

Source: Nvidia.

Source: Nvidia.

All told, Nvidia expects the GTX 690’s cooler to be very quiet for what it is—quieter even than some of the firm’s single-GPU cards. We’ll test that claim shortly, of course.

Specs like EyeMasters

GeForce GTX 680 1006 1058 1536 128 32 6 GT/s 256 195W
GeForce GTX 690 915 1019 3072 256 64 6 GT/s 2 x 256 300W

The GeForce GTX 690’s specifications are eye-popping, which is mostly what you’d expect from an SLI-on-a-stick graphics card. All of the GK104’s units are enabled, so many of the key rates are twice the GTX 680’s. Since the GTX 690 is a dual-GPU pairing, of course, the peak graphics rates shown in the table above are somewhat less connected to reality than usual. Applications may or may not take advantage of all of that power depending on many things, some of which we’ll discuss shortly.

The GTX 690 does have loads of bandwidth on tap, though. Between the two GPUs is a PCI Express bridge chip supplied by PLX; it has 48 lanes of PCIe Gen3 connectivity, 16 lanes routed to each GPU and 16 lanes connected to the host system.

Although the prior-generation GeForce GTX 590 performed more like a couple of down-spec GTX 570s, Nvidia has been able to reach relatively higher with this new card. Some of the credit goes to the Kepler generation’s new GPU Boost dynamic voltage and frequency scaling feature, which raises clock speeds to take advantage of any available thermal headroom. The GTX 690’s “base” clock is lower than the GTX 680’s by quite a bit, but the 690 has more range built into it. The 690’s “boost” clock of 1019MHz isn’t far from the GTX 680’s boost clock of 1058MHz. If the workload and the ambient conditions allow enough headroom, the GTX 690 should operate at something close to its boost clock rate—sometimes at even higher frequencies than that. As a result, Nvidia expects the GTX 690 to perform very similarly to a pair of GTX 680s in SLI.

That’s pretty impressive in the grand scheme, since the GTX 680 can claim to be the world’s fastest GPU. However, the GK104 graphics processor isn’t exactly a heavyweight, by most other standards; it’s a mid-sized chip with a modest power envelope and a 256-bit memory interface. In many ways, the GK104 is more like the GF104 chip that powers the GeForce GTX 560 Ti than it is like the GF110 chip that powers the GTX 580. That lineage is probably why the GTX 690 has “only” 2GB of memory per GPU—4GB in total, but effectively 2GB for all intents. (The Radeon HD 7970, by contrast, has 3GB for a single GPU.) The GK104’s middleweight status was no doubt helpful when Nvidia was attempting to cram two GPUs onto a single card in a reasonable power envelope. In fact, the GTX 690’s max power rating of 300W is 65W lower than the GTX 590’s.

The GeForce GTX 590 (left) versus the GTX 690 (right)

In many other respects, the GTX 690 mirrors its predecessor. The board is 11″ long, occupies two expansion slots, and requires a pair of 8-pin aux power inputs. The display outputs include a trio of dual-link DVI ports and a single mini-DisplayPort connector. The GTX 590 is just very, uh, plasticky by comparison.

Peak pixel
fill rate
Peak bilinear
texel filtering
rate int/FP16
Peak shader
GeForce GTX 580 37 49/49 1.6 3.1 192
GeForce GTX 680 34 135/135 3.3 4.2 192
GeForce GTX 590 58 78/78 2.5 4.9 328
GeForce GTX 690 65 261/261 6.5 8.2 384
Radeon HD 7970 30 118/59 3.8 1.9 264
Radeon HD 6990 53 159/80 5.1 3.3 320
Radeon HD 6990 AUSUM 56 169/85 5.4 3.5 320

The GTX 590 is also quite a bit slower than the 690, in theory, as is nearly every other video card out there. Only the Radeon HD 6990 with its special “AUSUM” overclocking switch thrown really competes on any of the key rates, like memory bandwidth and texture filtering—and in other important respects, it’s not even close. We’re not likely to see a true competitor for the GTX 690 until AMD takes the wraps off of its dual-Tahiti card. Frankly, we’re kind of surprised to have made it this far into 2012 without seeing AMD’s dual-GPU entry for this generation, since they’ve been talking about it for some time. Until that product arrives, the GTX 690 is pretty much in a class by itself.

That brings us, inevitably, to the question of price. Given the GTX 690’s premium materials and performance, Nvidia has decided to slap a price tag on this puppy that reads: $999.99, one penny short of a grand. I believe that makes the GTX 690 the most expensive consumer graphics card ever. The one-grand sticker essentially doubles the GTX 680’s list price, so it makes a sort of sense. Still, you’d kind of hope for some sort of volume discount when buying two GPUs together, wouldn’t you?

I dunno. I’m not sure the folks who would pony up for this sort of card will care that much.

One thing that this, er, formidable price tag could do is keep demand in line with the limited supply of these cards. Most folks are keenly aware that the supply of GK104 chips is rather tight right now, since the GTX 680 is tough to find in stock anywhere. Furthermore, the dual-GPU cards of the last generation, the Radeon HD 6990 and the GeForce GTX 590, seem to have been in short supply throughout their model runs. We expect the GTX 690 to reach online store shelves this week, but we have few illusions about them being plentiful, at least initially.

A Testing Conundrum

As you might recall, we’ve been skeptical about the merits of multi-GPU solutions like the GeForce GTX 690 since we published this article last fall. That piece introduced some new ways to think about gaming performance, and the methods we proposed immediately highlighted some problems with SLI and CrossFire.

Multi-GPU schemes generally divide the work by asking a pair of GPUs to render frames in alternating fashion—frame 1 to GPU 0, frame 2 to GPU 1, frame 3 to GPU 0, and so on. The trouble is, the two GPUs aren’t always in sync with one another. Instead of producing a series of relatively consistent frame delivery times, a pair of GPUs using alternate frame rendering will sometimes oscillate between low-latency frames and high-latency frames.

To illustrate, we can zoom in on a very small chunk of one of our test runs for this review. First, here’s how the frame times look on a single-GPU solution:

Although frame times vary slightly on the single-GPU setup, the differences are pretty small during this short window of time. Meanwhile, look what happens on a CrossFire setup using two of the same GPU:

You can see that alternating pattern, with a short frame time followed by a long one. That’s micro-stuttering, and it’s a potentially serious performance issue. If you were simply to measure this solution’s performance in average frames per second, of course, it would look pretty good. Lots of frames are being produced. However, our sense is that the smoothness of the game’s animation will be limited by those longer frame times. In this short window, adding a second GPU appears to reduce long-latency frames from about 29 ms to about 23 ms. Although the FPS average might be nearly doubled by the presence of all of those low-latency frames, the real, perceived impact of adding a second card would be much less than a doubling of performance.

This problem affects both SLI and CrossFire, including multi-GPU graphics cards like the GTX 690. How much micro-stuttering you find can vary from one moment to the next. In this example, we can see a little bit of jitter from the GTX 690, but it’s fairly minimal.

However, it appears that the degree of jitter tends to grow as multi-GPU solutions become more performance-constrained. That’s bad news in our example for the older dual-GPU graphics cards:

Ouch. If this trend holds up, the more you need higher performance from a multi-GPU solution, the less likely it is to deliver. Kind of calls the value proposition into question, eh?

Things get even trickier from here, for several reasons. Both AMD and Nvidia acknowledged the multi-GPU micro-stuttering problem when we asked them about it, but Nvidia’s Tom Petersen threw us for a loop by asserting that Nvidia’s GPUs have had, since “at least” the G80, a built-in provision called frame metering that attempts to counteract the problem.

Source: Nvidia

The diagram above shows the frame rendering pipeline, from the game engine through to the display. Frame metering attempts to smooth out the delivery of frames by monitoring frame times and, as necessary, adding a slight delay between a couple of points on the timeline above, T_render and T_display. In other words, the GPU may try to dampen the oscillating pattern characteristic of micro-stuttering by delaying the display of completed frames that come “early” in the sequence.

We think frame metering could work, in theory, with a couple of caveats. One obvious trade-off is the slight increase in input lag caused by delaying roughly half of the frames being rendered, although the impact of that should be relatively tiny. The other problem is the actual content of the delayed frames, which is timing-dependent. The question here is how a game engine decides what time is “now.” When it dispatches a frame, the game engine will create the content of that image—the underlying geometry and such—based on its sense of time in the game world. If the game engine simply uses the present time, then delaying every other frame via metering will cause visual discontinuities, resulting in animation that is less smooth than it should be. However, Petersen tells us some game engines use a moving average of the last several frame times in order to determine the “current” time for each frame. If so, then it’s possible frame metering at the other end of the graphics pipeline could work well.

A further complication: we can’t yet measure the impact of frame metering—or, really of any multi-GPU solution—with any precision. The tool we use to capture our performance data, Fraps, writes a timestamp for each frame at a relatively early point in the pipeline, when the game hands off a frame to the Direct3D software layer (T_ready in the diagram above). A huge portion of the work, both in software and on the GPU, happens after that point.

We’re comfortable with using Fraps for single-GPU solutions because it captures frame times at a fixed point in what is essentially a feedback loop. When one frame is dispatched, the system continues through the process and moves on to the next, stopping at the same point in the loop each time to record a timestamp.

That feedback loop loses its integrity when two GPUs handle the work in alternating fashion, and things become particularly tricky with other potential delays in play. Fraps has no way of knowing when a buffer flip has happened at the other end of the pipeline, especially if there’s a variable metering wait involved—so frame delivery could be much smoother in reality than it looks in our Fraps data. By the same token, multi-GPU schemes tend to have some additional latency built into them. With alternate frame rendering, for instance, a frame completed on the secondary GPU must be transferred to the primary GPU before it can be displayed. As a result, it’s possible that the disparity between frame display times could be much worse than our Fraps data show, as well.

So, what to do if you’re us, and you have a multi-GPU video card to review? The best we can say for our Fraps data is that we believe it’s accurate for what it measures, the point when the game engine presents a frame to Direct3D, and that we believe the frames times it captures are at least loosely correlated to the actual display times at the other end of the pipeline. We can also say with confidence that any analysis of multi-GPU performance based solely on FPS averages is at least as wrong as what we’re about to show you. We had hoped to have some new tools at our disposal for this article, including a high-speed camera we ordered, but the camera didn’t arrive in time for this review, unfortunately. We will have to follow up with it at a later date. For now, we’ll have to march ahead with some big, hairy caveats attached to all of our performance results. Please keep those caveats in mind as you read the following pages.

Testing the Nvidia GeForce GTX 590

Test notes

In order to take full advantage of high-end graphics cards these days, you’ve got to ask a lot of ’em. That’s why we decided to conduct our testing for this review with a trio of monitors, all Dell U2410s, each with a display resolution of 1920×1200.

Together, they have a collective resolution of about six megapixels, roughly 50% more pixels than the 30″ monitor we usually use for GPU testing. The increased resolution and complexity made it fairly easy to push the limits of these multi-GPU setups. We even had to go easy on the image quality settings in some cases to maintain playable frame rates.

Most of our multi-GPU pairings were built from cards we’ve tested before, but our GTX 680 team had one brand-new member: Zotac’s GeForce GTX 680 AMP!, a product just announced today. Obviously, that’s not a stock cooler, but it is very swanky. This is an AMP! edition, so its default clock speeds are quite a bit higher than a stock GTX 680’s. The base and boost frequencies are 1111MHz and 1176MHz, well above the stock 1006/1071MHz speeds. Even more impressively, perhaps, the Zotac card’s memory speed is 1652MHz, up from 1502MHz stock. We suspect memory bandwidth may be an important performance limiter on the GTX 680, so the higher RAM speeds are noteworthy. Zotac is asking $549 for this card, 50 bucks above the stock GTX 680’s list price.

For the purposes of this review, we committed the heinous crime of dialing back the Zotac GTX 680 card’s base and memory clock speeds to match the other card in the SLI pairing, which was a standard-issue GTX 680. We’re worried about GPUs being out of sync, after all, and we didn’t want to make matters worse with a mismatch. (We did the same with the XFX Radeon HD 7970, bringing it back to stock clocks to match the other card.) The thing is, the utilities we had on hand wouldn’t let us straightforwardly control the Zotac card’s boost clock, so perfect symmetry eluded us.

With the GTX 680, that is kind of the way of things, though. Nvidia expects slightly variant performance from every GTX 680 card thanks to GPU Boost, which will adjust to the particulars of a card’s thermals, the individual chip’s properties, and such. Two GTX 680s in SLI aren’t likely to run at exactly the same speed, since the thermal conditions at one spot in a system will vary from those at another. Nvidia anticipates that the frame metering capabilities in the GK104 will keep frame delivery consistent, regardless.

Oh, and please note that we tested the Radeon HD 6990 with its “AUSUM” switch enabled, raising its clock speed and PowerTune limits. We saw no reason not to test it in that configuration, given what it is.

Our testing methods

As ever, we did our best to deliver clean benchmark numbers. Tests were run at least three times, and we’ve reported the median result.

Our test systems were configured like so:

Processor Core i7-3820
Motherboard Gigabyte
Chipset Intel X79
Memory size 16GB (4 DIMMs)
Memory type Corsair
Vengeance CMZ16GX3M4X1600C9
DDR3 SDRAM at 1600MHz
Memory timings 9-9-11-24
Chipset drivers INF update
Rapid Storage Technology Enterprise
Audio Integrated
with Realtek drivers
Hard drive Corsair
F240 240GB SATA
Power supply Corsair
OS Windows 7 Ultimate x64 Edition
Service Pack 1
DirectX 11 June 2010 Update
core clock
GTX 590
608 854 3072
GTX 680
1006 1502 2048
GTX 680 + Zotac GTX 680
1006 1502 2048
GeForce GTX
915 1502 4096
12.4 + 12.3 CAP 1
880 1250 4096
HD 7970
12.4 + 12.3 CAP 1
925 1375 3072
HD 7970 + XFX HD 7970
12.4 + 12.3 CAP 1
925 1375 3072

Thanks to Intel, Corsair, and Gigabyte for helping to outfit our test rigs with some of the finest hardware available. AMD, Nvidia, and the makers of the various products supplied the graphics cards for testing, as well.

Unless otherwise specified, image quality settings for the graphics cards were left at the control panel defaults. Vertical refresh sync (vsync) was disabled for all tests.

We used the following test applications:

Some further notes on our methods:

  • We used the Fraps utility to record frame rates while playing a 90-second sequence from the game. Although capturing frame rates while playing isn’t precisely repeatable, we tried to make each run as similar as possible to all of the others. We tested each Fraps sequence five times per video card in order to counteract any variability. We’ve included frame-by-frame results from Fraps for each game, and in those plots, you’re seeing the results from a single, representative pass through the test sequence.
  • We measured total system power consumption at the wall socket using a Yokogawa WT210 digital power meter. The monitor was plugged into a separate outlet, so its power draw was not part of our measurement. The cards were plugged into a motherboard on an open test bench.

    The idle measurements were taken at the Windows desktop with the Aero theme enabled. The cards were tested under load running Skyrim at its Ultra quality settings with FXAA enabled.

  • We measured noise levels on our test system, sitting on an open test bench, using an Extech 407738 digital sound level meter. The meter was mounted on a tripod approximately 10″ from the test system at a height even with the top of the video card.

    You can think of these noise level measurements much like our system power consumption tests, because the entire systems’ noise levels were measured. Of course, noise levels will vary greatly in the real world along with the acoustic properties of the PC enclosure used, whether the enclosure provides adequate cooling to avoid a card’s highest fan speeds, placement of the enclosure in the room, and a whole range of other variables. These results should give a reasonably good picture of comparative fan noise, though.

  • We used GPU-Z to log GPU temperatures during our load testing.

The tests and methods we employ are generally publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

The Elder Scrolls V: Skyrim

Our test run for Skyrim was a lap around the town of Whiterun, starting up high at the castle entrance, descending down the stairs into the main part of town, and then doing a figure-eight around the main drag.

Since these are pretty capable graphics cards, we set the game to its “Ultra” presets, which turns on 4X multisampled antialiasing. We then layered on FXAA post-process anti-aliasing, as well, for the best possible image quality without editing an .ini file. We also had the high-res texture pack installed, of course. Although it’s not pictured above, the total display resolution was 5760×1200.

Frame time
in milliseconds
8.3 120
16.7 60
20 50
25 40
33.3 30
50 20

These first three plots show the raw data from a single test run, the rendering times for each individual frame, shown in milliseconds. Notice that because we’re thinking in terms of frame latency, lower numbers are better. For reference, we’ve included a table on the right showing the conversions from frame times to FPS.

As you can see, the GTX 690 performs essentially identically to two GTX 680s in SLI. Throughout the test run, the GTX 690’s frame latencies remain below 22 milliseconds or so, and they’re often under the magical 16.7 millisecond mark that, if it’s steady, translates into 60Hz or 60 FPS. Some of the other cards don’t fare as well, especially the Radeon HD 6990, whose spike-riddled plot reveals frame times that are often rather high. The GTX 590’s plot looks more like a cloud than a line, suggesting that it has some jitter going on, as well.

Just looking at the FPS average, the GTX 690 ties the GTX 680 SLI team, well ahead of anything else. Two Radeon HD 7970s in CrossFire, surprisingly enough, aren’t any faster than a GeForce GTX 590.

Of course, we’ve established that FPS averages don’t tell the whole story. We can get a better sense of the overall frame latency picture by looking at the 99th percentile frame time. Simply put, for each card, this number means that 99% of all frames were rendered in x milliseconds or less. Since we’re looking at a point where the vast majority of frames have been completed, the effects of any micro-stuttering oscillations will be reflected in this result.

Switching to this more latency-centered indicator does some interesting things for us. First and foremost, it brings the GTX 690 and the GTX 680 SLI back to the pack. Those two are only a couple of milliseconds ahead of a single GTX 680 in this measurement. Oddly enough, the Radeon HD 7970 CrossFire config looks to have higher latencies at the 99th percentile than a single 7970 card does. Worst of all for AMD, the Radeon HD 6990 looks like a basket case. Going by FPS alone, the 6990 would appear to be just a few ticks behind the 7970. A look at the latency picture reveals the gulf between the 6990 and everything else.

Then again, 99th percentile frame times are just one point along a whole latency curve, and we can show you how that curve looks.

With multi-GPU products in the mix, these latency curves are more interesting than ever. You can see that the Nvidia Geforce GTX 690 and the 680 SLI config are evenly matched throughout the test run, with no real weaknesses. Both solutions deliver frames quickly throughout, although their frame latencies rise, nearly to meet the single GTX 680’s, in the last 5% of frames.

The curve for the Radeon HD 7970 CrossFire setup tells quite the story. Although the dual 7970s deliver half of their frames much more quickly than a single card, their frame times rise at a sharper angle beyond 50%, eventually crossing over at around 82-83%. For the last 16% or so of frames delivered, the single Radeon HD 7970 is quicker. We’re likely seeing two halves of a multi-GPU jitter pattern illustrated in the 7970 CrossFire’s more rapidly ascending curve, and in the final analysis, the single 7970 may be the better of the two solutions.

We can also quantify “badness,” the slowdowns and delays one encounters while playing a game, by looking at the amount of time spent rendering frames above a certain threshold. The theory here is that the more time spent on long-latency frames, the more interruption you’re likely to perceive while playing a game.

We’ve chosen several noteworthy thresholds. The first, 50 milliseconds, equates to 20 FPS. We figure if the frame rate drops below 20 FPS for any length of time, most folks are likely to perceive a slowdown. The next two, 33.3 ms and 16.7 ms, equate to 30 and 60 FPS, respectively, which are traditionally important performance thresholds for gamers. Our three thresholds also equate to 60Hz, 30Hz, and 20Hz, the first three quantization points for a 60Hz display with vsync enabled. If you go beyond any of these points, you’ll be waiting at least one more vertical refresh interval before updating the screen.

As you might have expected, only the Radeon HD 6990 suffers any really substantial slowdowns, and even it doesn’t waste too much time working on frames above 50 milliseconds.

When we ratchet the threshold down to 16.7 ms, the GTX 690 and 680 SLI really separate themselves from the pack. A single GTX 680 card spends about three times as long as a GTX 690 or two GTX 680s in SLI above 16.7 milliseconds—so in a really picky way, the GTX 690 is measurably better at minimizing wait times for those worst-case frames.

Notably, the Radeon HD 7970 single-card and CrossFire configs are essentially tied here. Adding a second Radeon doesn’t appear to help at all in the most difficult cases.

Batman: Arkham City

We did a little Batman-style free running through the rooftops of Gotham for this one.

All of the plots for this game show lots of spikes, or occasional long frame times, something we’re used to seeing from Arkham City—doesn’t seem to matter much whether you’re dealing with one GPU or two.

The AMD and Nvidia offerings are much more competitive with one another here than they were in Skyrim, at least in the FPS sweeps.

Switching to the 99th percentile frame times whittles away the gap between the multi-GPU solutions and their single-chip equivalents. The GTX 690 still looks good here, but it’s no quicker than a pair of 7970s.

Although all of the solutions had spiky lines in the initial plots of frame times, the latency curve illustrates how two of the solutions, the single Radeon HD 7970 and the Radeon HD 6990, produce a higher proportion of long-latency frames than everything else. Again, the 6990’s sharper rise from the halfway point to about 88% suggests some longer frame times as part of a micro-stuttering pattern. However, the single 7970 struggles mightily with that last 7% of frames, all of which take longer than 60 milliseconds to render. Interestingly, in this case, adding a second card for CrossFire essentially eliminates those struggles.

The GTX 690 again looks excellent throughout, even surpassing the 7970 CrossFire config.

In this test scenario, with either Radeons or GeForces, you can substantially reduce slowdowns by adding a second GPU. That’s true for the GTX 680/690, and it’s true for the Radeon HD 7970, as well. The multi-GPU options look pretty darned good in that light.

Battlefield 3

We tested Battlefield 3 with all of its DX11 goodness cranked up, including the “Ultra” quality settings with both 4X MSAA and the high-quality version of the post-process FXAA. Our test was conducted in the “Operation Guillotine” level, for 60 seconds starting at the third checkpoint.

The multi-GPU plots look… cloudy. Could prove interesting.

Notice the disparity between the FPS average and the 99th percentile frame times. Although the 7970 CrossFire config is far and away the at the top of the FPS charts, it’s only slightly quicker than the GTX 690 at the 99th percentile. Why is that?

Well, both the 7970 CrossFire and the Radeon HD 6990 show a classic micro-stuttering split, as their frame times rise sharply at around the 50th percentile mark. The disparity is greater for the 6990, which looks to be more performance constrained. Even with that handicap, the 7970 CrossFire config at least matches the GTX 690 and 680 SLI across the latter half of the latency curve, which is why it still has a slight advantage at the 99th percentile. In other words, the 7970 CrossFire config is still a very competitive performer, even though it doesn’t quite live up to its towering advantage in the FPS results.

The worst performer here is the GeForce GTX 590, whose last 10% of frames take between 60 and 120 milliseconds to render.

I like the filtering effect we get from these three thresholds. If you’re looking to avoid real slowdowns, consider the 50-ms results, where only the 6990 and the GTX 590 really show any notable issues. At 33.3 ms, the three multi-GPU solutions from the current generation are in a class by themselves, while the Radeon HD 7970 carves out a clear advantage over the GeForce GTX 680. You’re better off with either current-gen single GPU than you are with a dual-GPU card from the prior generation, though. Crank the limit down to 16.7 ms, and the ranks of the top three remain the same, but all of the other solutions look to be pretty similar.

Crysis 2

Our cavalcade of punishing but pretty DirectX 11 games continues with Crysis 2, which we patched with both the DX11 and high-res texture updates.

Notice that we left object image quality at “extreme” rather than “ultra,” in order to avoid the insane over-tessellation of flat surfaces that somehow found its way into the DX11 patch. We tested 60 seconds of gameplay in the level pictured above, where we gunned down several bad guys, making our way across a skywalk to another rooftop.

You’ll see a couple of unexpected things in these results. First, the GeForce GTX 680 SLI is quite a bit slower than the GTX 690, which is unusual. Second, neither of those solutions performs very well compared to the Radeon HD 7970 CrossFire team. Why?

Well, looks like we found a bug in Nvidia’s graphics drivers. Here are the FPS averages for the GTX 680 SLI rig, runs one to five: 62, 44, 43, 44, 44. And here are the averages for the GTX 690: 59, 52, 51, 52, 50. In both cases, the first test run is faster than all of the rest. The GTX 590 suffered from the same problem, but none of the single-card configs did, nor did any of the Radeon setups. Looks like something bad happens when you exit and load a saved game on the Nvidia multi-GPU setups. After you’ve done that once, performance drops and doesn’t come back until you exit the game completely and start it up again. For whatever reason, the GTX 680 SLI setup suffers more from this issue than the GTX 690 does.

I briefly considered re-testing the Nvidia Geforce GTX 690 and company by exiting the game between runs, but I figure this problem is one that anybody who plays Crysis 2 will encounter. Seems like fair game to include it. Multi-GPU solutions tend to be prone to these sorts of issues, anyhow.

While we’re at it, these results are affected by a problem with the Radeons, as well. Let’s zoom in on the very beginning of the test runs to get a closer look.

One of the first motions of our test run, after loading a saved game, is to turn around 180° or so and face another direction. When we do so on the Radeon cards, we get some long delays, multiple frames that take 60 milliseconds or more. I didn’t know whether to blame the game engine or the Radeons, and I was considering doing a quick spin-around move to warm the GPU caches before starting the test run—until I started testing the GeForces, and the slowdowns became vastly less pronounced, almost imperceptible in most cases. You can see a single 70-millisecond frame from the GTX 690 above, but the following frames clock in at under 50 ms. There’s a lot more white area under those two Radeon lines, which means more time spent waiting. Again, I figured this problem was fair game to include, in the grand scheme of things.

Notice, also, that you can see the multi-GPU jitter patterns in both the GTX 690 and the 7970 CrossFire plots above. The GTX 690’s is less pronounced, but it’s still unmistakable.

Even with these two issues affecting the different camps, we still have some clear takeaways from these results. Of course, the Radeons all pay for that slowdown at the beginning of the test run in our 50-ms threshold results. There’s no escaping that.

Beyond that, the prior-generation multi-GPU cards look really quite poor, with the two worst latency curves from the 50th percentile on up and, thus, the most time spent beyond each of our three thresholds. You’re obviously better off with a single GTX 680 or 7970 than with a GTX 590 or 6990.

Finally, even with the Nvidia driver issue, the GTX 690 comes out of this test looking quite good in terms of the overall latency picture and its ability to avoid slowdowns.

Power consumption

These power consumption numbers aren’t quite what we’d expected, based on prior experience. Driving three monitors appears to change the math in some cases. For example, the Radeon HD 7970’s ZeroCore power feature evidently doesn’t work with three displays attached. Both the single 7970 and the primary card in our CrossFire team refused to enter the ZeroCore state when the display dropped into power-save mode. Their fans never stopped spinning. Not only that, but the idle power consumption numbers for a single 7970 are quite a bit higher than we saw with a single display attached. (And we’re not measuring the display power consumption here, just the PC’s power draw at the wall socket.) We’re also not sure why the GeForce GTX 680 SLI rig pulled more power with the display in power-save mode, but it did.

Given everything, the GTX 690’s power consumption is remarkably low, both when idle and when running Skyrim across three displays, which is how we tested under load. Our Nvidia geforce GTX 690 based test system pulled fully 100W less than the same system with a GTX 590 installed.

Noise levels and GPU temperatures

The NVIDIA GeForce GTX 690’s acoustic profile is, if anything, even more impressive than its power draw numbers. Nvidia’s new dual-GPU card achieves noise levels very similar to a single GeForce GTX 680, which is one of the quieter cards on the market. The 690 doesn’t come by its low decibel readings cheaply, either—it maintains lower GPU temperatures under load than almost anything else we tested.

If you’re wondering about why the single Radeon HD 7970 produced higher noise levels and higher temperatures than the 7970 CrossFire config, well, so am I. Some of the issue, I think, is that we have an asymmetrical CrossFire team, with different coolers on the two cards. Somehow, using them together produces a different fan speed policy, it seems. Also, of course, noise is not additive, so putting in a second card doesn’t always lead straightforwardly to higher decibel readings. Another contributor may be relatively higher GPU utilization in the single-card config, since 7970 CrossFire performance doesn’t appear to scale well in Skyrim. We may have to try testing with a different game next time.

Final Conclusions and Closing Thoughts

Well, now we have some performance numbers for the GeForce GTX 690. How correct they are, we’re not entirely sure. I will say this: regardless of the fact that we’ve not accounted for the potentially positive effects of frame metering, the GeForce GTX 690 looks to be the fastest overall graphics card on the planet. The GTX 690 even does well in our latency-sensitive metrics. Although it’s rarely twice as fast as a GeForce GTX 680 in terms of 99th percentile frame times, the GTX 690’s overall frame latency picture, as measured in Fraps, is generally superior to the GTX 680 by a nice margin. The GTX 690 also does a nice job of minimizing those painful instances of long-latency frames, improving on the performance of the GTX 680 in that regard in solid fashion.

Since we’re not entirely confident in our numbers, I’ll offer a few subjective impressions from my testing, as well. I started the process with the dual 7970 CrossFire team, and I chose the image quality settings for testing based on how well that config could run each game. My goal was to stress the multi-GPU solutions appropriately without cranking up the image quality options so far the games would be unplayable on the single-GPU cards.

I was initially surprised to see how easily the 7970 CrossFire config could be pushed to its limits at a six-megapixel resolution. The settings I chose yielded playable but not entirely fluid animation on the 7970 CrossFire rig (with the exception of BF3, which was stellar on the 7970s with the quality maxed out.) I was fearful of whether these games would truly be workable on a single 7970, but it turns out that I shouldn’t have worried. For the most part, playability wasn’t substantially compromised on a single card. However, playability was compromised when I switched over to the Radeon HD 6990. Although its FPS averages were generally higher than a single 7970’s, the experience of playing games with it was instantly, obviously worse. You might have guessed that by looking at our latency-focused numbers from Fraps, but the subjective experience backed that up.

From there, I switched to the green camp and the GeForce GTX 590, which was a bit of an upgrade from the 6990 in terms of overall smoothness—and it wasn’t such a basket case in Skyrim. When I swapped in a single GTX 680, though, the experence changed. The GTX 680 felt like a clear improvement in playability, after having tested the 6990 and GTX 590 back to back before it. The power of a single, fast GPU is something to be respected. I remember seeing the GTX 680’s FPS average at the end of the first Arkham City test run and being shocked at how low the number was (the card averaged 32 FPS) given the quality of the seat-of-the-pants experience.

Then again, I wasn’t getting much sleep during this period, and I’d overclocked my entire nervous system via copious amounts of tasty Brazilian coffee. Cerrado Gold, baby. Breakfast of champions.

The GTX 680 SLI config and the GTX 690 came next, and subjectively, the experience offered by the two was indistinguishable. Both were obviously faster in places where the GTX 680 felt strained, and I’d say they offered a better experience overall—and thus the best of any config I tested. However, it seemed like they’d still run into occasional, brief episodes of sluggishness that one didn’t experience on a single GTX 680.

You can make of those subjective impressions what you will. They’re in no way scientific, although I did try to throw in a big word here and there.

Given our seat-of-the-pants impressions and our test results, I’m pretty confident in offering a generally positive recommendation for the GeForce GTX 690. No, it’s not “twice as fast” as a GeForce GTX 680 in any meaningful sense, and a coldly rational buyer probably wouldn’t want to pay twice the GTX 680’s price for it. However, it is as quick as two GTX 680s in SLI, which makes it the highest-performance video card we’ve ever used. Furthermore, Nvidia has gone above and beyond with the GTX 690’s industrial design, materials, acoustics, and power efficiency, all of which are exemplary, outstanding, and most other positive words you might wish to use.

Cost Justification?

If you’re serious about buying one of these cards, you probably understand the logic of such a purchase better than I do. I’m not sure how one would justify the price, but Nvidia has given folks lots of shiny little excuses—and they’ve muted the drawbacks like excess noise, too. There’s not much not to like here, other than that fantastic, breathtaking, prodigious price tag. I suspect some folks will overcome that one obstacle without much trouble.

As for the lingering questions about multi-GPU micro-stuttering and the effectiveness of frame metering, we have several things in the works. There’s a tremendous amount of work still to be done, and we have a lot of other projects on our plate, so be patient with us, folks. Shortly before the completion of this review, we did finally receive that high-speed camera we’d ordered. We’ve already managed to capture a serviceable video at 240 FPS, four times our display’s refresh rate. The resolution isn’t too high, but it’s enough to show whether the animation is smooth. Have a look:

You can easily see the strobe of the monitor’s CCFL backlight, and since we had vsync disabled, several instances of tearing are clearly visible. We think this tool should allow for some worthwhile comparisons, eventually.

All the cool kids follow me on Twitter.

0 responses to “Review of Nvidia GeForce GTX 690 Graphics Card

  1. You can check out the GPU power usage with GPUZ and HWinfo. My 7950 seems limited to ~130W with the ‘power’ setting at +20%. The Heaven benchmark seems to draw ~70W, with MSI Kombustor it shoots to ~130W, at which point the GPU clock gets throttled.

  2. Well, GK104 is designed for efficient gaming performance, like the 560 TI, the 7870, or the 6870. It isn’t designed to be a DP compute card.

    GK 100 was canned, so GK 110 will probably be the next major Tesla compute card / GTX 780. GK 110 will be like the 280 or 580: a beast of a GPU compute card with a huge die and an even bigger price tag, making it relatively inefficient for gaming compared to more cost effective cards.

    The 7970 is odd: it’s somewhere in between a full blown compute card like the 580 and a gaming focused card like a 680. GK 110 will definitely have the gpu computation edge, but by then the Radeon 8000 series may be on it’s way.

  3. It is a shame that 120Hz source material is a rarity.

    Most games are capped at 60-85Hz for timing, animation and networking(multiplayer) reasons.

  4. I’m running the Asus 23″ 120hz with a Core i7 3770 and Nvidia 680 GTX. Even at native res + everything on ultra, this sucker is 80+ FPS. I will not go back to any monitor less than 120hz — its very noticeable even using the standard Windows desktop.

  5. I forgot to comment on his post earlier- but I think you’re right. If heat causes throttling, then yeah, any micro-stuttering would be exacerbated.

  6. Yeah, it’d be nice if there were more rigorous games available to give a trend like this more factual basis. Sadly the industry isn’t delivering anything worthwhile.

    What I took away from the from Damages 680 review is that it is designed for better general performance, but not for specific applications or when it really comes down to taxing the card hard. Unfortunately there is really nothing available to show off the 7970, so maybe Nvidia was smart to take such an approach and even if it’s not faster in all scenarios it’s better designed around todays workloads (including gaming).

    If that trend holds out into the future remains to be seen. Given current levels of consolization, that may be the case though.

  7. Yeah, things have slowed down as far as desireability goes. At one time games were keeping up and even pushing hardware so any amount of precious FPS you could get was great.

    The 680 isn’t as fast as the 7970 in some tests and vice versa. I don’t know why some people on here make it seem like the 680 hands down trumps 7970. I can only imagine its hate spilling over from BD.

  8. No doubt this is true!

    The real test, of course, is a detailed look at the fan and noise profiles of two comparable cards when overclocked, taking into account average reported overclocking results.

  9. Main reason to go with GTX670 or GTX680 SLi instead is for the fully exhausting coolers. My case is set up with massive positive airflow, with the H60 and two GPUs being the only exhausts. A GTX690 would probably work of course, but the GPU to the right of the fan may not cool nearly as well as the one on the left. Also, I don’t want to spend that much right now :).

  10. I agree- got one off the ‘Egg, should be here by Tuesday next week (I’m gone for the weekend anyway). Watch for 2x HD6950 2GB reference unlockable cards on sale shortly :).

  11. I agree. If I were looking at a going dual GPU, especially if only really concerned with 2560×1600, I’d be looking at hd 7950s or GTX 670s. Considering you’re inclined to stick it out with one card for a while, then I’d lean toward the GTX 670 as the single card solutions seem to definitely grant Nvidia the edge in most games.

    Or, if you can swallow the cost, waiting for a pair of 660 Tis will probably be the best bang/buck dual GPU setups for a 4 MP display.

    The only caveat to this with Nvidia cards right now is availability. Though the MSRPs of comparable cards across companies are similar, if they don’t improve supply, you may still end up with better value with the Radeons

  12. Interesting. That does not appear to be the case with my HD 7970. When I run the Unigine Heaven demo 3.0 at 2560X1600 with everything turned up, my FPS average drops to the low 30s at points. When running the Metro 2033 demo at 2560X1440 with everything turned up, the average FPS drops below 20 FPS and slowdowns become really apparent.

    This is with the HD 7970 clocked at 1175 Mhz, so power draw really starts to ramp up. With the Heaven demo, system power draw seems to remain under 400W. However, With the Metro 2033 demo, I get spikes hitting close to 500W at the points where the FPS drop < 20. After I’ve looped the demo a couple of times and the card’s nice and warm, these spikes easily surpass 500W.

    I suppose some of this could be the CPU ramping up higher in Metro 2033 too, but I would expect that most of this max power draw differential is due to the HD 7970

  13. 64 GB of ram will not be any faster than 4-8 because no consumer applications use that much memory.

  14. 2 x 670s in SLI should match the performance of this thing and save you $100. Nice sexy materials though. :/ Would buy a 670 with these materials for a $50 premium.

  15. QFT

    When I was in my late teens/early 20s, people thought I was crazy for dumping so much money into aluminum intakes, Holley double-pumpers, Hooker headers, cam and lifters, so on & so on. It’s just a different kind of money pit.

  16. I kinda found the whole Half-Life crowbar and crate gaming reference pretty cool, but YMMV as always. I’m sort of surprised that noone has mentioned it yet…

    Of course, using a 690 to play a Source based game is ridiculous overkill, but that only adds to the awesomeness, I suppose.

  17. Don’t forget that many enthusiasts don’t make a living depending on their hardware, and thus expensive/reliable/fast components don’t tend to pay for themselves over time in terms of less downtime or faster results, and then add to the equation that most companies can get tax breaks when investing in their production pipeline (at least where I live).

    For many enthusiasts, I would venture that their PCs are basically luxury and/or hobby items. But in terms of the amount of $$ spent on hardware, I do concede that we enthusiasts can be hard to please from a price/performance standpoint compared to server customers.

  18. Relative to the time ofc. The best you could have in that moment in time……you were always able to do that, with the right funding *wink*.

  19. Not really surprising to me, it seems that nVidia just clearly used second mover advantage and knowing the competition, created the card to be just a little faster.

  20. Wouldn’t you rather have this card for the better construction, thermals, and power efficiency?

    I don’t see any reason to go with 680 SLI over this, period. It even has better frame times…

  21. On second thought, what you probably thought of might’ve been Lucid’s HyperFormance, because it works exactly like you say, to deliver extra-smooth temporal resolution at the cost of more power used.

  22. The cards only “run free” if the framerate doesn’t even meet the refresh rate. The cards use regular vsync if the framerate is indeed sufficient, thus there’s no extra work in that case.

    You’re confusing this with OpenGL’s vsync with triple buffering (if I recall correctly), which burns out the card regardless of whether you meet the refresh rate or not.

  23. Well not many enthusiasts own a company. Or wait, maybe they do…..

    But i did say desktop parts not servers.

  24. [quote<]When running a title that is only moderately demanding at moderate resolutions, the TDP limits of the GTX 680 allow the clock speed and voltage to ramp up - granting superior performance. However, in a more demanding title, the TDP is reached at lower clock speeds/voltages and the GTX 680 then performs slightly inferior to the statically-clocked HD 7970.[/quote<] From my experience, more demanding titles are rather LESS demanding on the videocard, in terms of power consumed. That is, I expect the GTX680 to clock itself higher in games where the average framerate is lower. Remember the Starcraft situation where some geforces fried themselves in the menu screen? That was a simple scene to render, and apparently it demanded dangerous amounts of power before it was fixed.

  25. Yes because quad SLi configs have never been built by home users……/sarcasm

  26. Not true. It is the cutting edge of today. Just cause we can’t afford it, doesn’t mean it’s not the fastest build you can make with desktop parts. Performance wise must be x2, x3 of what you would build if not more (for gaming ofc).

  27. Enthusiasts are amusing. I spend 2-3x that on servers regularly, and have spent more for over a decade.

  28. The Voodoo 2 didn’t use alternate-frame rendering to divvy up the work, so there was no chance for out-of-sync GPUs to deliver frames in odd patterns.

    SLI originally stood for “scan-line interleave,” which was the V2’s load-balancing method. When Nvidia brought back SLI, they said it now stands for “scalable link interconnect,” which is the connection between the two cards.

  29. One thing I don’t understand is the micro-stuttering. I have a SLI Voodoo2 setup for my “classic” gaming PC and I have never noticed it. Is it because the graphic fidelity is so much greater these days and the cards have to work harder? Also if I remember correctly there are several ways to do SLI. Did nVidia just not implement these different ones in favor of the even and odd frames?

    Again I just always wondered why my Voodoo SLI setup never seemed to have any of these issues running as a pair. Please let me know anyone, thanks.

  30. Like Meadows said, 0b means that the following is binary (base2), just as 0x means that the following is hexadecimal (base16). Sometimes, people use the prefix 0h to denote hex encoded numbers as well.

    [code<]0b1010110010 = 2^9 + 2^7 + 2^5 + 2^4 + 2^1 = 512 + 128 + 32 + 16 + 2 = 690[/code<]

  31. my experience with microstuttering is the more heat u make the worse it gets. id love to see a study done that tests the latency of frames delivered when a gpu is watercooled cause after i put some full cover blocks on my gpus, the microstuttering i had in some games dissapeared and ive yet to have it pop back up in any game.

  32. the adaptive vsync kinda worries me. its supposed to basically let the cards run free at 100% and it will just grab the frames that are needed and toss the extra ones away. this could really push the heat up and i know air cooling solutions are almost at there limit right now with gpus.

  33. I didn’t even notice that. Interesting omission on such a piece of hardware.

  34. Heck yeah, I’d love to see more Hyperformance stuff. But for a stop-gap solution adaptive V-Sync works for me (for now!).

  35. The only real advantage Nvidia seems to have that catches my attention is in their frame-time results- they’re just smoother. After 60FPS I could give a rats about AVG FPS, so Nvidia wins for a GPU pair. Also, my primary game is BF3 multi-player, which is not nearly as forgiving as the single-player game that is used in benchmarks. Here, smoothness wins :).

  36. Well, based on this review anyway, at 2560×1600 and above, I’m not sure that I’d agree. The Nvidia cards thrash AMD in Skyrim. Batman is essentially a wash (as far as perceptible differences go), with the Geforces eeking out a win. And in Battlefield and Crysis, the AMD cards do the thrashing.

    And this was with the AMD cards clocked at their lowly “stock” 925Mhz. I suppose, due to the GTX’s dynamic clocking feature, if you didn’t want to deal with overclocking, the GTX 680s might get the win, but otherwise, I’m not so certain.

    I might be biased as I own an HD 7970 – bought one a month before the GTX 680s came out. At first I was dismayed, but now, other than paying the inflated pre-price cut charge, I’m kind of glad I did. If having to decide between an HD 7970 and GTX 680 at about the same price point (your shoes), I’m not sure what I would do. This review kind of cements my cognitive dissonance and makes me think I’m better off with the AMD card.

    The HD 7970’s better compute performance and performance in games like Crysis, Metro 2033 and Battlefield 3 make it look not so bad after all, especially when you factor in the overclockability of these cards. Add the fact that you can find them more readily and get one for as little as $450 now without too much difficulty.

  37. I should have put more detail in- essentially, 680GTX SLi is the better setup, if you cannot afford a 690GTX outright, over HD7970 CFX. While gaming at 2560×1600, no single card is fast enough (for me), which is why I’m making the point :).

  38. From what I’ve read on TR about Nvidias implementation it introduces more latency, but not the same amount of a triple buffer. Compared to the Lucid Hyperfromance approach, which apparently doesn’t introduce any latency and even cleans up frame rates. Two different approaches for the same topic. I’m more interested in Lucids tech though. I’m surprised TR doesn’t cover it more, unless nothing new has come of the technology yet. Lucid has some really interesting stuff happening.

  39. Not meaning to be difficult, but I don’t see how your reply fits with this particular thread, which basically points out that the HD 7970 may actually have performance advantages over the GTX 680 in certain situations. The import being that this notion is sort of counter-intuitive to the seeming general consensus on the net that the GTX 680 is hands-down better than the HD 7970.

    But yeah, the GTX 680 is great card. Just maybe not the hands-down best for gaming at 2560×1440 or higher resolutions.

  40. Heres a related video of slow motion exposing stutter, and in this case low framerate
    [url<][/url<] And as mentioned earlier in this thread, testing this on a 120Hz monitor would expose the judder twice as much. Speaking of 120Hz... how about some testing on what GPU can run what games, at what settings, and get 100-120fps? Once the framerate hits 100+ the game gets an amazing solidity.

  41. V-Sync is a ‘broad-brush’ or ‘crude’ implementation that is archaic but still moderately useful. The adaptive solutions give the benefit of V-Sync without the lag-inducing effects seen in some games.

  42. I’m seriously looking at selling my HD6950 2GB CFX setup to grab a single GTX680. The plan would be to grab another one later on.

  43. Just to prove a point, he went overboard on everything. That system is 5x as expensive as it needs to be, for the same end performance (if not more than 5x).

  44. I also think this will happen. In the initial GK104 review we saw it was built for graphics rendering speed rather than compute, especially FP16/double precision.

    It will be interesting to see what both the dual 7970 on-a-stick and GK100 /GK110 will look like when/if they are released. To be competitive on compute I think nV will have to beef up the units specifically for compute power or disable whatever seems to be crippling the compute units in GK104.

  45. 64GB of ram is for enthusiasts? You went overboard with the RAM, SSD’s and video cards just to prove a point.

  46. To chime in here.. I’ve also been noticing that while the 680 GTX generally out “FPSs” the HD 7970 at 1080p, etc, in most games, while more demanding games (ie Battlefield 3, Crysis 2, Metro 2033 on Very High) tend to give the HD 7970 the nod.

    Not to sound like an AMD fanboi or anything, but the Radeon seems to give you more performance when it matters most (also corroborated by the finding of higher minimum FPS, albeit lower average FPS). To me, after factoring in the somewhat generally better overclockability (and scaling of performance at higher overclocks) of the HD 7970, it seems that, if both the same price, anyone gaming at higher resolutions (ie, 2560×1440 and above) would be better off with the HD 7970 over a GTX 680. Perhaps it might even be fair to say the HD 7970 is a slightly more future proof-looking card? However the difference here is probably negligible given the other unknowns that dictate this. Furthermore, as shown in this review, crossfire seems to generally scale better in these more demanding environments.

    While the majority of this phenomenon can likely be explained by increased memory bandwidth of the HD 7970 as pointed out by Damage, I wonder if some of it isn’t attributable to the new dynamic clocking functionality (boost) of the GTX 6XX products. When running a title that is only moderately demanding at moderate resolutions, the TDP limits of the GTX 680 allow the clock speed and voltage to ramp up – granting superior performance. However, in a more demanding title, the TDP is reached at lower clock speeds/voltages and the GTX 680 then performs slightly inferior to the statically-clocked HD 7970.

    Of course, this could also be do to AMD playing games with their driver optimizations or application profiles for these specific titles, but I hardly think that is giving them the benefit of the doubt.

  47. There’s a big difference though as you can buy the i7-3930K for around 60% of the price and ~97% of the performance of the 3960X. You can’t do that with the GTX 690.

  48. You’re not the target market. I think it’s priced just fine. Also, I am semi-certain that some 5970 X2 cards went for $1200 or so.

  49. I love the idea of adaptive vsync and it makes Nvidia’s new stuff much more attractive to me.

    But seriously I hope it actually works properly. I simply don’t understand why v-sync doesn’t work with so many games.

  50. Sorry, but no. I’m not really spending any time looking at other sites right now.

  51. If that’s not sarcasm, i accept the challenge. I will list the most cutting edge desktop enthusiast components for a build. You do the same with components from, idk, 2007 lets’ say, you choose. Ofc don’t forget to adjust for inflation ^^.

    [url=<]Core i7-3960X Extreme Edition [/url<] $1,029.99 Mobo [url=<]ASUS Rampage IV Extreme LGA 2011 [/url<] $429.99 RAM [url=<]G.SKILL Ripjaws Z Series 64GB (8 x 8GB) [/url<] $949.99 Video Cards [url=<]2 x EVGA GeForce GTX 690 4GB [/url<] 2 x $999.99 PSU [url=<]SILVERSTONE ST1500 1500W ATX [/url<] $399.99 SSD [url=<]SAMSUNG 830 [/url<] $799.99 PCI-E SSD [url=<]OCZ RevoDrive 3 X2[/url<] $3,149.99 Mass storage HDD [url=<]Seagate Constellation ES 3TB [/url<] $459.99 CPU cooler [url=<]CORSAIR H100[/url<] $114.99 Case [url=<]Xigmatek Elysium Black Server Edition [/url<] $219.99 [b<]GTAND TOTAL: $ 9554.8[/b<]

  52. Very nice review, and you earn a lot of respect for being upfront that FRAPS frametimes may well not be telling us the whole story. Given the time and thought you’ve sunk into this new way of analyzing graphics performance, it would have been very tempting (for less scrupulous reviewers) to gloss over the inconvenient truth that the tool currently used might not be giving representative results. Hopefully something usable will come from the high speed camera work, or you’ll find another better way.

    As you rightly point out, even with the concerns about the current data, we’re still getting way more useful information than can be gleaned from reviews elsewhere. The subjective assessment at the end was especially interesting, and it’s good that it seems to tie in with the frametime consistency results far more than the raw FPS. It’s a nice validation of the whole approach.

    Just one thing that wasn’t entirely clear from your concluding text: subjectively was the single 680 noticeably better than either the single 7970 or the 7970 in Crossfire?

  53. “you can probably tell from the picture above that the GTX 690 looks a little different from your typical graphics card”

    yeah, NO HDMI! its not like it doesnt have space yah know…

  54. wow, you’re spending $1000 on a card and you’re complaining about $2 worth of box. Because $998 makes this card a whole lot easier to swallow.

    Also, a solid wooden box is much more useful than any paper/cardboard box. I could use it to [i<]put things inside[/i<].

  55. But people do shell that out. Not a lot, but enough that offering such a product is a profitable venture.

  56. Great article, I only wish there were more game tests to get a broader perspective.

    A few things that raised my eyebrow:

    In Skyrim, according to the graphs, 7970 looks like it stutters way more than 7970CF, yet the text seems to indicate the opposite? In any case I think you shoud maybe switch the benchmark to more GPU bound Riften woods rather than the CPU bound Whiterun. You can definetely see in the beginning of the benchmark GPU has nothing to do with the results and we’re CPU bound even at this extreme resolution.

    It would also be really nice to know what the deal is with 7970 in Arkham City. Maybe some setting bugged? Weird that CF could actually fix it though.

    AMD really need’s to fix their Zero core for Eyefinity. I haven’t been able to use any power saving options with my four displays connected, and in fact haven’t been able to trigger Zero Core even with one monitor connected.

    I still couldn’t make up my mind whether 690 was actually benefiting from frame metering after reading the article, which was mostly what I hoped to find out. Hopefully we’ll get some nice results from the new fast speed recording.

    What I would also like is for TR to test the effects of (adaptive) vsync and framerate limiting on multi GPU cards to limit stuttering. This proved effective for me, but it would be nice to see some deeper analysis on whether it is beneficial. For example in the BF3 graph 7970CF could use vsync to get rid of that awfully large variance.

    So far it still seems like it’s best to go with single 7970 and OC it to 1300MHz if you can. You might lose some performance against CF/SLI, but probably not a lot, and you’ll certainly save quite a bit of money and avoid driver issues. Multi GPU just isn’t as well supported as it should be. I just facepalmed when I read from CatalystMaker that Tera Online developers simply refused to add CF support for the game, and probably SLI too. I bet those mid range SLI/CF people are happy about their excellent value graphics…

  57. Damage, any comments on why you found in point (1) performance to be so similar while other websites peg the performance at: techpowerup ~5-10%, hardwarecanucks ~10%, etc?

  58. nah, I think it’s pretty dumb to plonk down a grand on any one component.

  59. The hockey-stick graphs & multiple “time spent beyond” charts are a HUGE improvement. This presentation finally makes the frame-time info useful, IMO.

    Well done, TR!

  60. Ah I didn’t catch the difference in resolution, that makes sense 🙂

  61. Any bets that “big Kepler” will only ever by a Tesla/Quadro part, and never appear as a GeForce?

  62. So, two things are up.

    1) If you look at *ahem* our GTX 680 review, we said at the time that the 680 and 7970 were essentially in a dead heat, performance-wise. The 7970 was a little ahead in the 99th percentile metric, while the GTX 680 had a slightly higher FPS average.

    2) We increased the display resolution tested here from a mix of 1920×1200 and 2560×1600 to 5760×1200. With that move, memory becomes more important than ever. The 7970 has a wider memory interface, more bandwidth, and 3GB of RAM versus the 680’s 2GB. I’d expect the bandwidth to be more of an advantage than the capacity, really, but either way, these really high resolutions are home turf for the bigger chip with the more robust memory subsystem.

  63. Yes and yes, and I’ve not seen anything about increased performance that’s relevant to the benchmarks in the Catalyst 12.3 or 12.4 release notes. Assuming the 680 didn’t magically slow down, what else could be up?

  64. It’s [u<]never been more affordable[/u<], when taking inflation into consideration.

  65. 3960x is kinda silly in its own way.

    Most of its power goes to waste if you use for gaming and other desktop tasks. It is really just a workstation part marketed towards gamers with more $$$$ than sense.

    Its Xeon counterparts are certainly worth the price-tage if you use your system for real work (Workstation). Time is $$$ here and the chips pay for themselves.

  66. [quote<]It's funny how some people think it's ok to shell out 1000 $ for a 3960x, because it's "high end", but not for a 690.[/quote<] But people don't. 1k for intel's EE is just as ridiculous.

  67. Just wanted to say I’m loving the quantile functions. Shows a lot of information, but it’s also very readable. I like picking three different latencies as well; you’re right, it helps make some distinctions between the cards much more apparent, and I like that all the chosen cutoffs make for a good point of reference between FPS and latency.

  68. The high-speed camera may be a big help for preliminary investigations like “does the frame metering actually work” and getting a feel for how the FRAPS data relates to what shows on the screen, but it’d be tough to use that as a primary means for analysis.

    (Since you won’t want to sift through tens of thousands of pictures looking for each time the display is updated you could try to write a script to automate the process of identifying screen updates by looking at a measure of difference between subsequent camera frames- SSIM, perhaps?- and translating spikes in that to frame times, but that could be error-prone esp. if there’s little motion between frames.)

    I’m not sure, but I would guess that the various vendor-specific profiling and benchmark tools which hook into the drivers and do lower-level stuff- nV PerfKit, AMD PerfStudio and PerfAPI, Intel GPA- might record frame times at a later point in the process than FRAPS and thus give you better quality data.

  69. The problem with that is that cases introduce a number of new variables. How many fans? Vents? Where is the GPU in relation to the power supply? How much else is in the case?

    Simulating a case would only simulate the cases that match the characteristics of the test. By doing an open test, you take out just the GPU in relation to heat and noise, so even though it might seem unrealistic, you know how hot your GPU is making things with respect to other GPUs. That can be more than enough to let you know what kind of case you need for your new build.

  70. [url=<]So...prediction confirmed?[/url<]

  71. As time goes by and drivers reform I expect 680 sli to preform notably better.

  72. Anyone else a little… surprised… at the 7970 results? I admit to not following the GPU Wars as closely as I once did, but when the 680 reviews first hit it definitely seemed to me like the 680 was a hands down winner. Now, these results seem to indicate not so much. In fact, the 7970 seems to handle BF3 and Crysis 2 a little better than the 680.

  73. Exactly. I made the same comment on the podcast and I was surprised nobody picked up on it. Have people really never heard of [url=<]the one and only objective metric for reviewing games[/url<]? It may be tongue-in-cheek, but it made people take a new approach to thinking about art, realism, creativity, and gameplay in level design.

  74. I wonder if they’ll use Adaptive Vsync on the 690/680 in the future?

  75. They could probably make GTX 685 a GTX 670/660 on a stick to fill the very large gap between the 680 and 690.

  76. Oh, I get it. You think this is the retail packaging. My guess is no, when the 5 or 6 people who want to blow the money buy this thing, it’ll come in a boring cardboard box.

  77. Scott: Try turning your monitor up to full brightness when recording the high-speed video. This should eliminate (or at least greatly reduce) the backlight flicker effects. It also gives you more light to get a better picture.

  78. It’s a new world record for Start-To-Crate! Before you even install the video card to play your game: Crate! *Cue oscar music*

    (If you don’t know: [url<][/url<] )

  79. Thumbs up if you decoded the 0b1010110010 on the wooden crate to 690.

    Nice touch, that.

  80. I’d rather they took some $ of the price, and didn’t bother with the stupid box. you’re paying for it, and wtf are you going to do with it when you’re done? it’s moronic.

  81. Cool high motion video. I agree about showing the difference between 60hz and 120hz monitors with it. Some people still poopoo on the idea of 120hz monitors making any sort of tangible difference as the human eye can’t see faster then 60fps *rolls eyes*.

    That said, it would also be interesting to see what Lucids new vsync technology does to these results. I forgot what it was called, but I remember you guys talking about it being used on a laptop demo. It would be interesting to see if that clears up the sli/xfire micro-stuttering problems.

  82. [quote<]We've already managed to capture a serviceable video at 240 FPS, four times our display's refresh rate.[/quote<] I wonder how a 120hz screen would change that video. It doesn't make a lot of sense for Skyrim (its engine doesn't like high framerates), but maybe it could bring to light some nuances in games like BF3. I guess most people who could really use that GTX 690 have multiple screens and/or high resolution screens, for which a 120hz frame rate would be very hard to attain. But, I'm sure I'm not the only one who bought a high end videocard specifically to go with a 120hz display. And personally, as long as there are still instances of frame-jitter, I wouldn't buy into SLI or Crossfire.

  83. Scott, that’s a very well done article; thank you. The high speed camera footage is super cool. And your testing methods are more solid and meaningful than any other site on the net, period.

    The 690 is pretty magnificent. The best dual GPU card ever made in every way. Very impressive. But it’s far beyond both my means and my needs. Heck, every card in this review is beyond my needs. But if the GTX 690 was on display in a graphics card museum somewhere, I’d like to go see it.

    What’s easier for me to appreciate is the GTX 680. That card is quite spectacular. Nvidia delivered with that card, now if Nvidia could just deliver the card in sufficient volumes to meet demand! I continue to be surprised that Nvidia delivered a card that beats the 7970 in every single metric. I expected it to be faster, but not quieter, and certainly not less power-consuming.

    p.s. – when are you going to get started building that graphics card museum? I know an architect or two.

  84. I was one of the lucky ones who snagged a 680 off of the Egg. In the 2 months+ since then, all cards are out 🙁

  85. It’s funny how some people think it’s ok to shell out 1000 $ for a 3960x, because it’s “high end”, but not for a 690.

    The 690 houses two state of the art GPUs and each can render stuff about 5-6 times faster than a single 3960x. Add memory, the PCB, VRMs and heatsinks incorporating more metal and copper than Intel uses in 5 of its shipped coolers (not that 3960x has any) and… Well, you get the picture 🙂

    I know it’s a bit of an apple to oranges comparison, but since GPUs these days are used by the big boys for stuff other than gaming, it was bound to happen.

  86. Seriously? It made me laugh in a positive way. Sure it’s over the top, but it’s funny. It’s a good thing to be amused once in a while 😉

  87. Sapphire has decorated the boxes and coolers of my last couple of cards with images of women with swords. My latest purchase from Gigabyte has only a stylized outline eye on the box and no artwork on the cooler at all.

  88. Wooing the press with a fancy crowbar and ridiculous packaging.

    That’s representative of a real product, isn’t it!

    Anyway, I’ve been looking forward to an “inside the second” article on this new flavour of SLI, I’ll hunker down for a good read….

  89. The price is too damn high (like rent). 1k for a vid card another for a CPU, another for an SSD….sooo expensive to be at the cutting edge.

  90. The only one that I saw in stock (for less than 5 minutes) was going for $1208 at Newegg.

  91. Considering how often Vsync doesn’t do anything I don’t think it matters which word he uses.

  92. Thanks for the great review, it provided more info than others’ even though you tested less games. While it is useful to have an adequately high FPS I find of more practical value to look in some detail at frame time, thus 99% percentile – my geek mind would say “if FPS > 60 then look at 99% frame-time”. The 60FPS number is variable too, I wouldn’t need 60 in Diablo for example.
    That said, I’m happy the GTX690 actually does something to smooth out the rough edges in 680’s gameplay and add some FPS in the process. Can’t wait for the high-speed camera footage, are those wears/tears repeatable? Anyway, many thumbs up for the review

  93. Not to nitpick, but there was another customer grade card that went for the outlandish $999 USD mark in its heyday.

    It was Quantium3D’s Voodoo 2 SLI on a single-card solution and that was for 1997-1998 USD.

    Anyway, this is probably the best SLI/CF on stick card so far in terms of scalability, power consumption and noise-level. 690 “beats” 680 SLI and eats less power while doing it. The best part that is 690 technically has less resources at disposal (lower clock speed) than the 680 SLI setup. That’s impressive in its own right. SLI/CF on stick solutions are usually a little worse than their SLI/CF counterparts.

    The only way you can justify the cost of this card is you are running some kind of 3D Vision setup or want to drive 4Megapixels with high doses of AA/AF on top at a somewhat smooth framerate. FYI, micro-shuttering prevents it from obtaining buttery-smoothness.

  94. Quote from article:

    “You can easily see the strobe of the monitor’s CCFL backlight, and since we had vsync enabled, several instances of tearing are clearly visible.”

    Did you mean to say “we had vsync disabled”?

  95. The wooden crate is [i<]a lot better[/i<] than those Korean anything-punk random illustrations which appear on retail boxes.

  96. I really don’t like this cooling setup. It’s optimized for open bench test -> The results in temp and noise are going to be different in closed cases -> They might be misleading for 99% of people.

    I understand open benches are more convenient when testing a lot of cards but I really think you should think about some sort of contraption to simulate a case. Maybe something as simple as some kind of cover with a couple fans for the air flow.

  97. cheapest I’ve seen one in Australia is $1600. I don’t care how good it is, I never let myself get shafted on prices like that.

  98. I’m pretty sure the crate is a PR gimmick only, not for end users

    If you think that is bad, you should hear the stories of what video game publishers send to editorial staff.

  99. It’s just as bad as when AMD/ATI tried to market the 6990 with the machine gun case. They are just stupid and gimmicky. Just give it a nice box and be done with it, do you really need a freaking wooden crate for it?

  100. Shipped it in a crate….ridiculous!

    Also, you need to take a regular clip of you in slow-mow, than the clip with crossfire/sli to really see a good comparison I feel.

    Otherwise, as expected I suppose.