What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

NVIDIA GeForce GTX 780 Review

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
Onscreen Frame Times w/FCAT

Onscreen Frame Times w/FCAT


When capturing output frames in real-time, there are a number of eccentricities which wouldn’t normally be picked up by FRAPS but are nonetheless important to take into account. For example, some graphics solutions can either partially display a frame or drop it altogether. While both situations may sound horrible, these so-called “runts” and dropped frames will be completely invisible to someone sitting in front of a monitor. However, since these are counted by its software as full frames, FRAPS tends to factor them into the equation nonetheless, potentially giving results that don’t reflect what’s actually being displayed.

With certain frame types being non-threatening to the overall gaming experience, we’re presented with a simple question: should the fine-grain details of these invisible runts and dropped frames be displayed outright or should we show a more realistic representation of what you’ll see on the screen? Since Hardware Canucks is striving to evaluate cards based upon and end-user experience rather than from a purely scientific standpoint, we decided on the latter of these two methods.

With this in mind, we’ve used the FCAT tools to add the timing of partially rendered frames to the latency of successive frames. Dropped frames meanwhile are ignored as their value is zero. This provides a more realistic snapshot of visible fluidity.


GTX-780-NV-44.jpg

GTX-780-NV-45.jpg

GTX-780-NV-46.jpg

GTX-780-NV-47.jpg

Here and on other sites, there has been a monumental amount of discussion about AMD’s frame timing issues when using Crossfire. As we can see, the problems haven’t cascaded down into single card configurations as in most cases the HD 7970 GHz easily matches the fluidity of NVIDIA’s GTX 780. As usual, AMD’s problems in Far Cry 3 persist with extremely distracting intermittent stutters but that was its only real misstep.

The 780 does hold an advantage in some situations but from an overall fluidity perspective, there really isn’t that much to distinguish one card from another.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
Onscreen Frame Times w/FCAT (pg.2)

Onscreen Frame Times w/FCAT (pg.2)


When capturing output frames in real-time, there are a number of eccentricities which wouldn’t normally be picked up by FRAPS but are nonetheless important to take into account. For example, some graphics solutions can either partially display a frame or drop it altogether. While both situations may sound horrible, these so-called “runts” and dropped frames will be completely invisible to someone sitting in front of a monitor. However, since these are counted by its software as full frames, FRAPS tends to factor them into the equation nonetheless, potentially giving results that don’t reflect what’s actually being displayed.

With certain frame types being non-threatening to the overall gaming experience, we’re presented with a simple question: should the fine-grain details of these invisible runts and dropped frames be displayed outright or should we show a more realistic representation of what you’ll see on the screen? Since Hardware Canucks is striving to evaluate cards based upon and end-user experience rather than from a purely scientific standpoint, we decided on the latter of these two methods. With this in mind, we’ve used the FCAT tools to add the timing of runted to the latency of successive frames. Dropped frames meanwhile are ignored as their value is zero. This provides a more realistic snapshot of visible fluidity.


GTX-780-NV-48.jpg

GTX-780-NV-49.jpg

GTX-780-NV-50.jpg

The final FCAT tests repeat the results seen in the previous round with neither card holding an advantage. Both provide a completely smooth gaming experience without any noticeable stutters.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
Temperatures & Acoustics / Power Consumption

Temperature Analysis


For all temperature testing, the cards were placed on an open test bench with a single 120mm 1200RPM fan placed ~8” away from the heatsink. The ambient temperature was kept at a constant 22°C (+/- 0.5°C). If the ambient temperatures rose above 23°C at any time throughout the test, all benchmarking was stopped..

For Idle tests, we let the system idle at the Windows 7 desktop for 15 minutes and recorded the peak temperature.


GTX-780-NV-55.jpg

NVIDIA has set their default temperature limit on the GTX 780 to 80°C so the second its GK110 core hits a degree above that, fan speeds are gradually increased to compensate. As a result, the maximum temperature doesn’t get above 81°C regardless of ambient conditions or load.


Acoustical Testing


What you see below are the baseline idle dB(A) results attained for a relatively quiet open-case system (specs are in the Methodology section) sans GPU along with the attained results for each individual card in idle and load scenarios. The meter we use has been calibrated and is placed at seated ear-level exactly 12” away from the GPU’s fan. For the load scenarios, a loop of Unigine Valley is used in order to generate a constant load on the GPU(s) over the course of 15 minutes.

GTX-780-NV-52.jpg

With the new fan speed profile in full effect, the GTX 780 is one of the quietest cards on the market and is nearly inaudible when playing a game. What this chart doesn’t show is how the lower number of rotational speed fluctuations impacts the overall acoustical experience. Since any fan speed increases are done gradually, you’ll never be distracted by it.


System Power Consumption


For this test we hooked up our power supply to a UPM power meter that will log the power consumption of the whole system twice every second. In order to stress the GPU as much as possible we used 15 minutes of Unigine Valley running on a loop while letting the card sit at a stable Windows desktop for 15 minutes to determine the peak idle power consumption.

Please note that after extensive testing, we have found that simply plugging in a power meter to a wall outlet or UPS will NOT give you accurate power consumption numbers due to slight changes in the input voltage. Thus we use a Tripp-Lite 1800W line conditioner between the 120V outlet and the power meter.

GTX-780-NV-56.jpg

The GTX 780 isn’t a particularly frugal card but its performance per watt factor is still head and shoulders above AMD’s current high end offerings. It also seems like the cut down SMX modules do tend to lower this card’s power needs regardless of its TDP similarity with that of the TITAN. With all things considered, GK110 seems to be a versatile core in a number of different ways and we doubt the GTX 780 will be its last hurrah. It is also interesting to see the progress which has been made since the GTX 580 was released.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
Overclocking Results

Overclocking Results


To overclock the GTX 780, we used EVGA’s Precision while manually setting the card’s fan speed to 60%, a level which was still nearly inaudible over the sounds of gaming but ensured consistent temperatures. Voltage was also set to the maximum allowable under Precision which is +35mV.

As you may have garnered from comments in other parts of this review, our experience overclocking the GTX 780 met with several roadblocks along the way. According to the handy “reasons” feature within Precision we were able to determine that even with both set to the max, core voltage and the card’s Power Limit kneecapped any serious overclocking attempts. Let’s be honest, an additional 35mV and 6% more power overhead won’t result in any clock speed miracles.

With that being said, both of our GTX 780 samples hit nearly identical levels and provided decent overclocks. Boost clocks remained steady at 1122MHz with spikes to 1168MHz while the memory leveled out at 6868MHz before error correction began throttling performance.

Throughout over overclock benchmarking, it became evident that NVIDIA’s GTX 780 certainly isn’t starved for memory bandwidth. Increasing GDDR5 clocks did little to nothing to improve performance until the core speeds were bumped up. Nonetheless, with both core and memory running at their final clock speeds, the GTX 780 has the capability to match and in some cases surpass the GTX TITAN’s performance.

GTX-780-NV-58.jpg

GTX-780-NV-59.jpg
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,840
Location
Montreal
Conclusion

What was supposed to be a slow year for the consumer graphics market has quickly turned into a free-for-all of epic proportions. The GTX 780 is NVIDIA’s latest broadside towards an AMD lineup that has begun feeling a bit dated so once again, we’re seeing a GeForce card catapulted into a segment without any direct competition.

The refresh direction between GTX 680 and GTX 780 is actually quite straightforward; think of it as this generation's GTX 480 / GTX 580 transition, but this time using the Kepler architecture. Cores have been added, clock speeds massaged and the feature set has received a shot of adrenalin. The result is a card which can nearly compete against the GTX TITAN yet hits a more “affordable” price point. More importantly, since AMD has decided to skip their own refresh, it will be a matter of months rather than weeks before a response can be issued. This will give NVIDIA’s GTX 780 and its upcoming siblings free reign over the high stakes enthusiast GPU market for the time being.

GTX-780-NV-51.jpg

For someone buying a $650 graphics card, performance concerns always come first and the GTX 780 delivers in spades. It thoroughly outpaces NVIDIA’s own GTX 680 by an average of 31% and manages to stay well ahead of AMD’s HD 7970 GHz Edition while also treading dangerously close to the TITAN. In many titles it was a dominant figure in terms of framerates and frame times. NVIDIA still needs to work on optimizations for Tomb Raider and Dirt Showdown but even in these two instances, the GTX 780 had no trouble delivering a playable, stutter-free experience.

The GTX 780’s primary goal is to provide a suitable step-up solution for the thousands of GTX 580 users who didn’t jump on NVIDIA’s GTX 680 bandwagon. If you’re one of the many gamers that skipped GK104, it will provide an immeasurably impressive gaming experience. That 70% performance improvement comes alongside some tangible fringe benefits as well; power consumption, noise and heat have all been significantly reduced since the last Fermi-based cards were introduced.

When it comes to price, like all ultra high end graphics cards, the GTX 780’s positioning becomes a bit less definitive. To ensure TITAN’s sales weren’t compromised, NVIDIA couldn’t institute a sub-$599 price, nor could their new card retail for more than $699 due to current market conditions. $649 meanwhile is a happy middle ground, giving them some wiggle room once AMD rolls out their response. However, while it outperforms the GTX 680 and HD 7970 GHz Edition by significant margins, performance gaps of 31% and 22% respectively won’t directly make up for a $200 premium. On the other hand, when compared directly against NVIDIA’s $999 TITAN, the GTX 780 starts looking like an astounding value.

But is the GTX 780 really worth nearly 45% more than a HD 7970 GHz Edition? To someone who cares about worry-free display overclocking, multi monitor support, multi card fame delivery, quieter fan speeds and properly implemented stereoscopic software, absolutely. The eventual evolution of the GeForce Experience platform will become yet another reason to choose an NVIDIA card in the future. But with all of those items being set aside, the GHz Edition still provides a phenomenal performance per dollar uppercut, making it an inexpensive alternative…for now. It can also overclock to a level that closes the gap between it and the GTX 780. Sure, the GTX 780 can overclock as well but it tends to feel a bit constrained in this respect, regardless of its limited voltage tuning abilities.

The GTX 780 is surprisingly efficient and provides an excellent performance per watt ratio, items which previous flagship GPUs couldn’t lay claim to. It requires the same amount of juice as AMD’s HD 7970 GHz Edition while offering much better framerates. This relative frugality is backstopped by one of the quietest fans we’ve come across on a reference card.

All in all, the GTX 780 is exactly the card NVIDIA needed to release at this juncture. They have effectively bridged the gap between the GTX 680 and TITAN with a card that thoroughly outperforms the former and retails for significantly less than the latter. While its value ratio against lower end cards may not perfectly line up with everyone’s expectations, the GTX 780’s proximity to TITAN ensures enthusiasts can now get bleeding edge performance without having to shell out a thousand bucks.


dam_good.jpg
 
Last edited:
Top