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!

EVGA GTX 780 Classified Review

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
Location
Montreal
EVGA’s GTX 780 lineup may hold an incredible eight different variations but their GTX 780 Classified is without a doubt the flagship. This card’s sole focus is to offer the highest possible amount of overclocking headroom to beginners and enthusiasts alike, which meshes well with the series’ pedigree. Fromm one generation to the next, it has always been Classified cards which headlined the overclocking charts and this time is no different.

Some may be wondering how EVGA can talk about this card’s overclocking prowess after their GTX 680 Classified was curtailed after launch by limiting headroom and effectively castrating EVBot support. We’ve often lamented that NVIDIA’s own policies restrict their board partners from exploring the Kepler architecture’s boundaries and the previous Classified couldn’t be a better example. Naturally, a situation like this could have left the GTX 780 version in a precarious position since much of its potential would have been inaccessible to the general buying public, rendering the numerous high-end features pointless.

The unenviable situation NVIDIA put their board partners into didn’t stop EVGA from approaching the issues from a different perspective. Instead of creating a piece of software meant only for overclockers and locking it behind iron-clad NDAs like MSI does, EVGA opened their Precision tool and the Classified’s BIOS to the public and developers. As a result, a vibrant, passionate community has been built up around the GTX 780 Classified and they’ve created a series of custom BIOS files and a number of unique applications. The modified software and BIOSes aren’t officially endorsed by EVGA but we have a feeling they’re sitting back, nodding in approval as their card runs away from the completion.

So what do these modified elements bring to the table? There’s a BIOS floating around which boosts the Power Target to 200% which Precision can take full advantage of while a custom designed Classified voltage tool offers a nearly unlimited amount of additional current. If that isn’t enough to get an overclocker salivating, we don’t know what would.

While the community involvement gives EVGA some plausible deniability, provided the PCB isn’t touched, the various community-developed additions won’t void your 3 year warranty. EVGA can offer this unheard-of support structure since there is a series of failsafes built into the card which should ensure safe operation, even in extreme conditions.

GTX-780-CLASSIFIED-47.jpg

For those who aren’t willing to take the GTX 780 Classified to the next level, EVGA has implemented a robust set of specifications. With the core operating at 993MHz and Boost frequencies topping out at about 1046MHz, this is actually one of the fastest GTX 780 cards available. More importantly, due to its efficient cooling design and high efficiency components (more on those two aspects a bit later), the average frequency we saw was a staggering 1150MHz, over 200MHz faster than the reference card. These frequencies max the Classified one of the fastest GTX 780s available, one upping Gigabyte's WindForce, ASUS' DirectCU II OC and EVGA's own ACX SC.

As with ever other GTX 780 we’ve reviewed to date, memory allotment and GDDR5 frequencies remain at reference levels. This is due to the challenge in sourcing and binning modules that will reliably overclock to higher speeds. However, we can assure you that EVGA’s chosen memory has plenty of headroom.

GTX-780-CLASSIFIED-1.jpg

At first glance, the Classified looks like an oversized GTX 780 ACX but there’s more here than what first meets the eye. Not only are there countless features hidden under the sleek exterior but a length of 11” and a titanic width of 5” should overcome any mistaken identity problems.

GTX-780-CLASSIFIED-2.jpg
GTX-780-CLASSIFIED-3.jpg

EVGA’s ACX (or Active Cooling Xtreme) cooler makes a comeback on this card but in an expanded setup with larger fin arrays for additional cooling capacity. This version is based off of EVGA’s first in-house heatsink design and like its smaller sibling, utilizes a large dual chambered heatsink layout with five chrome-plated copper heatpipes and a secondary reinforcement baseplate. This baseplate is supposed to increase rigidity while also lowering memory and VRM temperatures.

The two 80mm fans boast some impressive specifications as well. They are engineered to be extremely light, thus decreasing their acoustical signature and ensuring less power is needed for their operation. They are equipped with double ball bearings which grant a 12 year lifespan, a significant improvement over competing solutions according to EVGA.

GTX-780-CLASSIFIED-4.jpg

Along the Classified’s side are a number of different connectors and other add-ons. There’s a port for EVGA’s unique tuning device, the EVBot (which is sold separately for $50) and a secondary hub for the $10 ProbeIT connector. ProbeIT is a unique take on the typical voltage read points found on many of these high-end, overclocker-friendly GPUs since it includes cabled multimeter adapters that make access infinitely easier.

EVGA has also included a dual BIOS switch on their card to change between the default setup and a preloaded custom BIOS which allows for an enhanced Power Target of 115%. Finally, there’s a number of LEDs which are supposed to glow when all components are operating with norms.

GTX-780-CLASSIFIED-5.jpg

The GTX 780 Classified’s ultra widebody PCB is there for a reason: it houses an immense 14+3 phase all-digital PWM that features upgraded components. It also gives that ACX heatsink some much-needed expansion room so the fin array could be optimized for the core’s increased heat output.

GTX-780-CLASSIFIED-6.jpg
GTX-780-CLASSIFIED-7.jpg

Moving around to the card’s side, we have a pair of 8-pin power connectors with a total input rating of 300W. Meanwhile, the backplate may house a basic layout of HDMI, DisplayPort and DVI connectors but EVGA has expanded the cooling capabilities by adding larger ventilation slots.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
Location
Montreal
Test System & Setup

Main Test System

Processor: Intel i7 3930K @ 4.5GHz
Memory: Corsair Vengeance 32GB @ 1866MHz
Motherboard: ASUS P9X79 WS
Cooling: Corsair H80
SSD: 2x Corsair Performance Pro 256GB
Power Supply: Corsair AX1200
Monitor: Samsung 305T / 3x Acer 235Hz
OS: Windows 7 Ultimate N x64 SP1


Acoustical Test System

Processor: Intel 2600K @ stock
Memory: G.Skill Ripjaws 8GB 1600MHz
Motherboard: Gigabyte Z68X-UD3H-B3
Cooling: Thermalright TRUE Passive
SSD: Corsair Performance Pro 256GB
Power Supply: Seasonic X-Series Gold 800W


Drivers:
AMD 13.8 BETA
NVIDIA 326.80


*Notes:

- All games tested have been patched to their latest version

- The OS has had all the latest hotfixes and updates installed

- All scores you see are the averages after 3 benchmark runs

All IQ settings were adjusted in-game and all GPU control panels were set to use application settings


The Methodology of Frame Testing, Distilled


How do you benchmark an onscreen experience? That question has plagued graphics card evaluations for years. While framerates give an accurate measurement of raw performance , there’s a lot more going on behind the scenes which a basic frames per second measurement by FRAPS or a similar application just can’t show. A good example of this is how “stuttering” can occur but may not be picked up by typical min/max/average benchmarking.

Before we go on, a basic explanation of FRAPS’ frames per second benchmarking method is important. FRAPS determines FPS rates by simply logging and averaging out how many frames are rendered within a single second. The average framerate measurement is taken by dividing the total number of rendered frames by the length of the benchmark being run. For example, if a 60 second sequence is used and the GPU renders 4,000 frames over the course of that time, the average result will be 66.67FPS. The minimum and maximum values meanwhile are simply two data points representing single second intervals which took the longest and shortest amount of time to render. Combining these values together gives an accurate, albeit very narrow snapshot of graphics subsystem performance and it isn’t quite representative of what you’ll actually see on the screen.

FCAT on the other hand has the capability to log onscreen average framerates for each second of a benchmark sequence, resulting in the “FPS over time” graphs. It does this by simply logging the reported framerate result once per second. However, in real world applications, a single second is actually a long period of time, meaning the human eye can pick up on onscreen deviations much quicker than this method can actually report them. So what can actually happens within each second of time? A whole lot since each second of gameplay time can consist of dozens or even hundreds (if your graphics card is fast enough) of frames. This brings us to frame time testing and where the Frame Time Analysis Tool gets factored into this equation.

Frame times simply represent the length of time (in milliseconds) it takes the graphics card to render and display each individual frame. Measuring the interval between frames allows for a detailed millisecond by millisecond evaluation of frame times rather than averaging things out over a full second. The larger the amount of time, the longer each frame takes to render. This detailed reporting just isn’t possible with standard benchmark methods.

We are now using FCAT for ALL benchmark results.


Frame Time Testing & FCAT

To put a meaningful spin on frame times, we can equate them directly to framerates. A constant 60 frames across a single second would lead to an individual frame time of 1/60th of a second or about 17 milliseconds, 33ms equals 30 FPS, 50ms is about 20FPS and so on. Contrary to framerate evaluation results, in this case higher frame times are actually worse since they would represent a longer interim “waiting” period between each frame.

With the milliseconds to frames per second conversion in mind, the “magical” maximum number we’re looking for is 28ms or about 35FPS. If too much time spent above that point, performance suffers and the in game experience will begin to degrade.

Consistency is a major factor here as well. Too much variation in adjacent frames could induce stutter or slowdowns. For example, spiking up and down from 13ms (75 FPS) to 28ms (35 FPS) several times over the course of a second would lead to an experience which is anything but fluid. However, even though deviations between slightly lower frame times (say 10ms and 25ms) wouldn’t be as noticeable, some sensitive individuals may still pick up a slight amount of stuttering. As such, the less variation the better the experience.

In order to determine accurate onscreen frame times, a decision has been made to move away from FRAPS and instead implement real-time frame capture into our testing. This involves the use of a secondary system with a capture card and an ultra-fast storage subsystem (in our case five SanDisk Extreme 240GB drives hooked up to an internal PCI-E RAID card) hooked up to our primary test rig via a DVI splitter. Essentially, the capture card records a high bitrate video of whatever is displayed from the primary system’s graphics card, allowing us to get a real-time snapshot of what would normally be sent directly to the monitor. By using NVIDIA’s Frame Capture Analysis Tool (FCAT), each and every frame is dissected and then processed in an effort to accurately determine latencies, frame rates and other aspects.

We've also now transitioned all testing to FCAT which means standard frame rates are also being logged and charted through the tool. This means all of our frame rate (FPS) charts use onscreen data rather than the software-centric data from FRAPS, ensuring dropped frames are taken into account in our global equation.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
Location
Montreal
Assassin’s Creed III / Crysis 3

Assassin’s Creed III (DX11)


<iframe width="560" height="315" src="http://www.youtube.com/embed/RvFXKwDCpBI?rel=0" frameborder="0" allowfullscreen></iframe>​

The third iteration of the Assassin’s Creed franchise is the first to make extensive use of DX11 graphics technology. In this benchmark sequence, we proceed through a run-through of the Boston area which features plenty of NPCs, distant views and high levels of detail.


2560x1440

GTX-780-CLASSIFIED-37.jpg

GTX-780-CLASSIFIED-30.jpg


Crysis 3 (DX11)


<iframe width="560" height="315" src="http://www.youtube.com/embed/zENXVbmroNo?rel=0" frameborder="0" allowfullscreen></iframe>​

Simply put, Crysis 3 is one of the best looking PC games of all time and it demands a heavy system investment before even trying to enable higher detail settings. Our benchmark sequence for this one replicates a typical gameplay condition within the New York dome and consists of a run-through interspersed with a few explosions for good measure Due to the hefty system resource needs of this game, post-process FXAA was used in the place of MSAA.


2560x1440

GTX-780-CLASSIFIED-38.jpg

GTX-780-CLASSIFIED-31.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
Location
Montreal
Dirt: Showdown / Far Cry 3

Dirt: Showdown (DX11)


<iframe width="560" height="315" src="http://www.youtube.com/embed/IFeuOhk14h0?rel=0" frameborder="0" allowfullscreen></iframe>​

Among racing games, Dirt: Showdown is somewhat unique since it deals with demolition-derby type racing where the player is actually rewarded for wrecking other cars. It is also one of the many titles which falls under the Gaming Evolved umbrella so the development team has worked hard with AMD to implement DX11 features. In this case, we set up a custom 1-lap circuit using the in-game benchmark tool within the Nevada level.


2560x1440

GTX-780-CLASSIFIED-39.jpg

GTX-780-CLASSIFIED-32.jpg



Far Cry 3 (DX11)


<iframe width="560" height="315" src="http://www.youtube.com/embed/mGvwWHzn6qY?rel=0" frameborder="0" allowfullscreen></iframe>​

One of the best looking games in recent memory, Far Cry 3 has the capability to bring even the fastest systems to their knees. Its use of nearly the entire repertoire of DX11’s tricks may come at a high cost but with the proper GPU, the visuals will be absolutely stunning.

To benchmark Far Cry 3, we used a typical run-through which includes several in-game environments such as a jungle, in-vehicle and in-town areas.



2560x1440

GTX-780-CLASSIFIED-40.jpg

GTX-780-CLASSIFIED-33.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
Location
Montreal
Hitman Absolution / Max Payne 3

Hitman Absolution (DX11)


<iframe width="560" height="315" src="http://www.youtube.com/embed/8UXx0gbkUl0?rel=0" frameborder="0" allowfullscreen></iframe>​

Hitman is arguably one of the most popular FPS (first person “sneaking”) franchises around and this time around Agent 47 goes rogue so mayhem soon follows. Our benchmark sequence is taken from the beginning of the Terminus level which is one of the most graphically-intensive areas of the entire game. It features an environment virtually bathed in rain and puddles making for numerous reflections and complicated lighting effects.


2560x1440

GTX-780-CLASSIFIED-41.jpg

GTX-780-CLASSIFIED-34.jpg


Max Payne 3 (DX11)


<iframe width="560" height="315" src="http://www.youtube.com/embed/ZdiYTGHhG-k?rel=0" frameborder="0" allowfullscreen></iframe>​

When Rockstar released Max Payne 3, it quickly became known as a resource hog and that isn’t surprising considering its top-shelf graphics quality. This benchmark sequence is taken from Chapter 2, Scene 14 and includes a run-through of a rooftop level featuring expansive views. Due to its random nature, combat is kept to a minimum so as to not overly impact the final result.


2560x1440

GTX-780-CLASSIFIED-42.jpg

GTX-780-CLASSIFIED-35.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
Location
Montreal
Tomb Raider

Tomb Raider (DX11)


<iframe width="560" height="315" src="http://www.youtube.com/embed/okFRgtsbPWE" frameborder="0" allowfullscreen></iframe>​

Tomb Raider is one of the most iconic brands in PC gaming and this iteration brings Lara Croft back in DX11 glory. This happens to not only be one of the most popular games around but it is also one of the best looking by using the entire bag of DX11 tricks to properly deliver an atmospheric gaming experience.

In this run-through we use a section of the Shanty Town level. While it may not represent the caves, tunnels and tombs of many other levels, it is one of the most demanding sequences in Tomb Raider.


2560x1440

GTX-780-CLASSIFIED-43.jpg

GTX-780-CLASSIFIED-36.jpg
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
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-CLASSIFIED-45.jpg

With the Classified’s core running at such high speeds, it ran the very real possibility of being held back by its 81°C temperature limit. Luckily, that didn’t happen as the expanded ACX heatsink proved itself more that worthy of dispersing the excess heat.


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-CLASSIFIED-44.jpg

Even though it keeps pace with the overclocked core, the ACX heatsink doesn’t produce all that much noise. In all likelihood, its two fans will be easily drowned out by the intake and exhaust fans on your case. However, if you want to overclock the Classified, we’d suggest upping the rotational speed since as it stands, it is still quite close to thermal throttle territory.


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-CLASSIFIED-46.jpg

The Classified is the highest clocked GTX 780 we’ve come across so it goes without saying that power consumption should hit levels that rise above a TITAN.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
Location
Montreal
Overclocking Results

Overclocking Results


Overclocking a GTX 780 usually boils down to increasing clock speeds, maxing out the pitiful voltage and Power Limit sliders and then smashing face first into NVIDIA’s predetermined ceiling. It was an infinitely frustrating process that dragged down even the most expensive cards. The challenges were of course multi faceted: on one hand there’s a Voltage Limit which can only be increase by a pitiful 38mV. If that helps out, the Power Limit will usually step in and cap clock speeds since higher voltage typically leads to greater core power needs. It goes without saying that NVIDIA’s maximum of 110% simply isn’t enough.

As we mentioned in the introduction, EVGA’s community of intrepid overclockers has found a number of ways around these hurdles. May of these community developed tools we’ll be discussing can be found in this thread over at Overclock.net. Basically, any GTX 780 Classified experience (not to mention review) would be incomplete without delving into the finer points of pushing past NVIDIA’s barriers.

GTX-780-CLASSIFIED-10.png

The cornerstone of EVGA’s overclocking suite is their Precision tool, a utility that continues to evolve and still provides one of the best software monitoring and tuning platforms around. Here, you are able to push the Classified but only within the usual boundaries even though EVGA has kindly provided a preloaded secondary BIOS which allows for a Power Target of 115%.

Unfortunately, this higher Power Target really won’t do all that much simply because the Classified’s clock speeds are held back by NVIDIA’s maximum voltage. This means your overclocks will hit a voltage bottleneck long before the core gets starved for power.

GTX-780-CLASSIFIED-11.png

The first step towards overcoming the voltage limitation is to install the aptly-named GTX Classified Tuner. This lightweight tool allows for tuning voltages on the core (NVVDD) memory (FBVDD) and PCI Express (PEXVDD), all items which are typically unlocked with the EVBot. Before delving too far into these, remember that the voltages listed aren’t additive but rather represent the true current going to each of those components.

GTX-780-CLASSIFIED-9.jpg

There is an art to overclocking the Classified since it may be a forgiving card with an unbelievable amount of frequency overhead but its still governed by the laws of GPU engineering. Basically, voltage plus more power equals higher temperatures. The trick is balancing all three in order to get optimal clock speeds. This is where NVIDIA’s Reasons come into play since they tell you exactly what’s holding back a given overclock. Typically, this means maximizing the Power Target and temperatures before moving onto voltage since it will have a drastic impact upon the other two metrics.

Take the picture above for example. In it, we can see the Classified under full load (Unigine Valley running on a second screen) at an incredible and relatively constant 1360MHz.

That’s an achievement worthy of note on air cooling but Precision’s readout tells us both voltage and power are our limiting factors, subsequently dragging down clock speeds. Increasing both leads to increased heat output and that wouldn’t be optimal since the Classified was already running at almost 90°C with the fans set at 85%. Unfortunately, that meant it was time to scale things back a bit.

GTX-780-CLASSIFIED-8.jpg

By moving to a core speed of 1.32GHz, we were able to all but eliminate the ultra high temperatures and ensure a continual clock speed, regardless of the situation. Temperatures normalized as well. After 20 minutes of constant load (see the image above), the core frequency didn’t budge one iota, though we were still hitting the Classified’s Voltage Limit. This goes to show that just a moderate voltage boost could actually reduce clock speeds rather than improve overhead.

Did we forget to mention memory? Well, that topped out at 7426MHz before the GDDR5’s error correction stepped in, which is nothing short of incredible.

EVGA has obviously created a monster and unlike its competition, the Classified has the tools necessary to push the very limits of Kepler overclocking. Granted, we were held back by temperatures even without resorting to the custom BIOS which grants a Power Target of 200% but with better cooling, the sky’s the limit.

GTX-780-CLASSIFIED-48.jpg
GTX-780-CLASSIFIED-49.jpg
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,841
Location
Montreal
Conclusion

Conclusion


Up until this point, overclocker-oriented Kepler cards have been disappointments due to their distinct lack of overclocking headroom. While much of the blame for this can be placed squarely on NVIDIA’s shoulders, board partners have been strangely hesitant to unlock their cards’ potential even though they’re more than willing to charge a premium for features that will go largely unused. EVGA’s approach this time around is unique and it ultimately allows the GTX 780 Classified to become THE card to have for would-be overclockers.

Before we get too far into the GTX 780’s clock speed mastery, its out-of-box performance shouldn’t be overlooked. Even without flexing its substantial muscles, its default Boost frequency is the highest we’ve come across on a GTX 780. This leads to GTX TITAN-matching framerates which is an accomplishment considering the Classified costs $300 less than NVIDIA’s single core flagship. As usual, default performance is capped by the Voltage Limit in most situations but that’s easily taken care of by EVGA’s excellent Precision utility.

This card was built from the ground up for enthusiasts and in that respect, it succeeds in spectacular fashion. EVGA’s approach of community-based development should be a model for other board partners whose efforts have been stymied by NVIDIA’s restrictive Green Light program. It is this initiative which spawned the handy Classified voltage utility and a BIOS that pushes the card’s Power Target to a ridiculous 200%. With these two tools in hand, the GTX 780 Classified can reach unheard-of frequency levels without resorting to exotic cooling methods.

Even though we didn’t actually use the modified Power Limit BIOS, clock speeds hit levels which far surpassed competitors’ solutions. EVGA’s secondary pre-installed BIOS already includes an additional 5% of power overhead which helped immeasurably when trying to find the Classified’s air cooled limits. And those limits are astronomical. With that being said, the most important aspect of this card is the fact that it can take full advantage of water cooling or LN2 whereas MSI’s Lightning and ASUS’ DirectCU II OC both top out well within the limits of their respective heatsinks. EVGA has designed a monster by simple community engagement and perseverance; they deserve credit for that.

Much of the GTX 780 Classified’s overclocking headroom is accessible to everyday gamers because of EVGA’s excellent ACX cooling solution. It is both quiet and offers bucketloads of performance, keeping the core within acceptable temperature limits all the way past the 1.3GHz mark. Above that, you may want to look at EVGA’s Classified Hydro Copper water block.

Historically, EVGA has priced their Classified cards at levels that were out of reach to pretty much everyone but this iteration is blazing a new trail here as well. At $700, it still costs $50 more than the reference version and $20 more than alternatives from ASUS and Gigabyte but for a few bucks more you get what we believe is the best GTX 780 currently available.

The EVGA Classified may be one of the most expensive GTX 780 cards around but it is the first to truly push the outer limits of NVIDIA’s Kepler architecture. Not only does this make it a highly valuable commodity for overclockers and gamers alike but it points to a new direction for custom graphics card development. If you have $700 set aside for a GPU upgrade, this is the card you’ll want. Period.

240463923aca1f6b.jpg
240486eb8252c5e5.png
 
Top