NVIDIA GTX 680 ReviewBosco , airman -
» Discuss this article (308)
Testing of the NVIDIA GTX 680 will consist of running it and comparison cards through the OverclockersClub.com suite of games and synthetic benchmarks. This will test the performance against many popular competitors. Comparisons will be made to cards of a range of capabilities to show where each card falls on the performance ladder. The games used are some of today's newest and most popular titles, which should be able to provide an idea of how the cards perform relative to each other.
The system specifications will remain the same throughout the testing. No adjustment will be made to the respective control panels during the testing, with the exception of the 3DMark Vantage testing, where PhysX will be disabled in the NVIDIA Control Panel, if applicable. I will first test the cards at stock speeds, and then overclocked to see the effects of an increase in clock speed. The cards will be placed in order from highest to lowest performance in each graph to show where they fall by comparison. The latest press release driver will be used in testing of the GTX 680. Other NVIDIA comparison cards will be using the 296.10 drivers; AMD will be using Catalyst 12.3 drivers.
- Processors: Core i7 2600K @ 4.4 GHz 100 x 44
- CPU Cooling: Corsair Hydro Series H100
- Motherboard: Gigabyte Z68AP-D3
- Memory: Mushkin 998990 Blackline PC3-1000 @ 1866MHz 9-11-9-27 8 GB
- Video Card: NVIDIA GTX 680
- Power Supply: Mushkin 1000 watt Joule Modular power supply
- Hard Drive: 1 x Seagate 1TB SATA
- Optical Drive: Lite-On DVD-RW
- Case: Corsair Graphite Series 600T
- OS: Windows 7 Professional 64-bit
Comparison Video Cards:
- XFX HD 7970 Black Edition
- XFX HD 7950 Black Edition
- ASUS GTX 580 Direct CU II
- ASUS GTX 570 DIrect CU II
- ASUS GTX 590
- NVIDIA GTX 680 — Core 1305MHz, Memory 1628MHz
Really getting into the overclocking of the GTX 680 is an interesting task. The intent of NVIDIA's overclocking for Kepler cards (right out of the box) is unique to what we're probably all used to. Built into the GTX 680's circuitry is a power usage monitor. By default, the GTX 680 targets 100% power usage by increasing the core clock of the card if there is "opportunity" available. For example, a less-stressing game running at 100% GPU usage may not require the same amount of power (100% power target) as a very intense game running the GPU at 100% usage. So, the hardware circuitry recognizes that for this less-stressing game, say 85% of the power target, it increases its clock speeds to reach the desired power target up to the stock GPU Boost clock of 1058MHz. In some cases, clock speeds will actually go below the base clock — even when there is GPU activity (aside from a very low, 2D clock). If a process does not require 100% GPU usage, clock speeds will decrease. Only at very high GPU usage levels (probably 80% or higher) will the card run at its "Base Clock".
User overclocking comes in by using EVGA's Precision software. Just like any other overclocking software, it displays current clocks, temperatures, fan speeds, and a monitor of all the values. There are three main adjustments that can be made for overclocking purposes. These are the card's power target, GPU clock offset, and memory clock offset. Adjusting the offset values simply increases the Base and Boost clock. The power target slider is going to control how often those clock speeds are met. Since I am not concerned about my power usage for my overclocked testing, I will set this slider to a maximum value. By doing so, the card will never meet this power target — meaning the card will never try to decrease its clocks, resulting in steady clock speeds.
A good bit of negative feedback is present on the Internet about how the Kepler cards overclock. Even I was worried, but after realizing that the level to which the power target can be set can not possibly cause the card to decrease its clocks, those worries went away. As soon as I saw my set clock speed steady for every benchmark, I felt loads better. So, to put this to rest, your overclocking potential is not going to be limited AT ALL! So, looping Unigine 3.0 while I was overclocking, I was able to reach 1305MHz on the core and 1628MHz (6500MHz effective) on the memory. That's nearly a 30% increase on the core and just under 10% on the memory. I was hoping for 7GHz memory here, but it couldn't quite make it. Even while pushing the card past its limit and having the overclock fail, the system never crashed. Only the graphics card driver fails, but successfully recovers. It is certainly the most painless overclocking I've had for sure; the system never locked up!
Maximum Clock Speeds:
Testing for the maximum clock speed consists of looping Unigine 3.0 for 30 minutes each to see where the clock speeds fail when pushed. If the clock speed adjustment fails, then the clock speeds are adjusted and the test is rerun until each card passes the testing.
- Gaming Tests:
- Metro 2033
- Batman: Arkham City
- Battlefield 3
- Sid Meier's Civilization V
- Unigine Heaven Benchmark 2.5
- DiRT 3
- Mafia II
- 3DMark 11
- Power Consumption