Submit Benchmarks!

Submit SSD Benchmark
Submit GPU Benchmark

Compare any two graphics cards:

Geforce GTX 670 vs Radeon HD 7970


The Geforce GTX 670 has a GPU clock speed of 915 MHz, and the 2048 MB of GDDR5 memory runs at 1500 MHz through a 256-bit bus. It also is comprised of 1344 SPUs, 112 Texture Address Units, and 32 ROPs.

Compare all that to the Radeon HD 7970, which features GPU clock speed of 925 MHz, and 3072 MB of GDDR5 memory running at 1375 MHz through a 384-bit bus. It also features 2048 SPUs, 128 TAUs, and 32 ROPs.

Display Graphs

Hide Graphs


These are real-world performance benchmarks that were submitted by Hardware Compare users. The scores seen here are the average of all benchmarks submitted for each respective test and hardware.

3DMark Fire Strike Graphics Score

Radeon HD 7970 8225 points
Geforce GTX 670 7424 points
Difference: 801 (11%)

Ethereum Mining Hash Rate

Radeon HD 7970 21 Mh/s
Geforce GTX 670 13 Mh/s
Difference: 8 (62%)

Power Usage and Theoretical Benchmarks

Power Consumption (Max TDP)

Geforce GTX 670 170 Watts
Radeon HD 7970 250 Watts
Difference: 80 Watts (47%)

Memory Bandwidth

The Radeon HD 7970, in theory, should be much faster than the Geforce GTX 670 overall. (explain)

Radeon HD 7970 264000 MB/sec
Geforce GTX 670 192000 MB/sec
Difference: 72000 (38%)

Texel Rate

The Radeon HD 7970 should be a bit (approximately 16%) more effective at texture filtering than the Geforce GTX 670. (explain)

Radeon HD 7970 118400 Mtexels/sec
Geforce GTX 670 102480 Mtexels/sec
Difference: 15920 (16%)

Pixel Rate

If using a high screen resolution is important to you, then the Radeon HD 7970 is a better choice, though only just barely. (explain)

Radeon HD 7970 29600 Mpixels/sec
Geforce GTX 670 29280 Mpixels/sec
Difference: 320 (1%)

Please note that the above 'benchmarks' are all just theoretical - the results were calculated based on the card's specifications, and real-world performance may (and probably will) vary at least a bit.

Price Comparison

Geforce GTX 670

Radeon HD 7970


Display Specifications

Hide Specifications

Model Geforce GTX 670 Radeon HD 7970
Manufacturer nVidia AMD
Year May 2012 January 2012
Code Name GK104 Tahiti XT
Memory 2048 MB 3072 MB
Core Speed 915 MHz 925 MHz
Memory Speed 6000 MHz 5500 MHz
Power (Max TDP) 170 watts 250 watts
Bandwidth 192000 MB/sec 264000 MB/sec
Texel Rate 102480 Mtexels/sec 118400 Mtexels/sec
Pixel Rate 29280 Mpixels/sec 29600 Mpixels/sec
Unified Shaders 1344 2048
Texture Mapping Units 112 128
Render Output Units 32 32
Bus Type GDDR5 GDDR5
Bus Width 256-bit 384-bit
Fab Process 28 nm 28 nm
Transistors 3540 million 4313 million
Bus PCIe 3.0 x16 PCIe 3.0 x16
DirectX Version DirectX 11.0 DirectX 11.1
OpenGL Version OpenGL 4.2 OpenGL 4.2

Memory Bandwidth: Memory bandwidth is the maximum amount of information (counted in megabytes per second) that can be transported across the external memory interface within a second. It is worked out by multiplying the interface width by the speed of its memory. In the case of DDR memory, it must be multiplied by 2 again. If DDR5, multiply by 4 instead. The better the card's memory bandwidth, the better the card will be in general. It especially helps with AA, High Dynamic Range and high resolutions.

Texel Rate: Texel rate is the maximum number of texture map elements (texels) that can be applied in one second. This figure is worked out by multiplying the total number of texture units by the core speed of the chip. The better this number, the better the video card will be at texture filtering (anisotropic filtering - AF). It is measured in millions of texels processed in a second.

Pixel Rate: Pixel rate is the maximum number of pixels the graphics card could possibly write to the local memory in a second - measured in millions of pixels per second. The figure is worked out by multiplying the number of Render Output Units by the the core speed of the card. ROPs (Raster Operations Pipelines - aka Render Output Units) are responsible for drawing the pixels (image) on the screen. The actual pixel output rate is also dependant on quite a few other factors, most notably the memory bandwidth of the card - the lower the memory bandwidth is, the lower the ability to reach the max fill rate.

Geforce GTX 670

Radeon HD 7970


10 Responses to “Geforce GTX 670 vs Radeon HD 7970”
Anonymous says:

[...] cierto aqui teneis una comparacion de la 7970 vs una gtx 670 Geforce GTX 670 vs Radeon HD 7970 – Performance Comparison Benchmarks @ Hardware Compare Es obvio que la 7970 es bastante mas en [...]

limbo | Pearltrees says:

[...] Geforce GTX 670 vs Radeon HD 7970 – Performance Comparison Benchmarks @ Hardware Compare Napisałeś książkę – sam się wydaj - Technologia, nowe technologie - The Corrupted Blood debuff being spread amongst characters in Ironforge, one of World of Warcraft ′s in-game cities. The Corrupted Blood incident is a video game glitch and virtual plague that occurred on September 13, 2005 in the MMORPG World of Warcraft . The epidemic began with the introduction of the new dungeon Zul'Gurub and its end boss Hakkar who, when confronted and attacked, would cast on players a hit point draining and highly-contagious debuff spell called "Corrupted Blood." The spell, intended to last only seconds and function only within the new area of Zul'Gurub, soon spread across the virtual world when players discovered that the use of teleportation spells could take the affliction out of its intended confines. [...]

Nvidia 670 vs ATI 7970 says:

[...] should benefit in editing too. i know 670 might prove to less stronger theoretically than 7970 ( Geforce GTX 670 vs Radeon HD 7970 – Performance Comparison Benchmarks @ Hardware Compare ) but may freaking surprisingly and incredibly prove to be competing with 7970 but recently 660ti [...]

Aankoop nieuwe grafische kaart - 9lives - Games Forum says:

[...] Geforce GTX 670 vs Radeon HD 7970 – Performance Comparison Benchmarks @ Hardware Compare De 7970 is volgens deze benchmarks een stuk sneller (niet immens veel). Maar die heeft waarschijnlijk nog wel een 10% OC mogelijkheden en dan zit je al een aardig stuk boven die 670. [...]

Pc gaming - says:

[...] GDDR5 5700/5800MHz 3GB 2xDVI HDMI DP Se servono comparazioni di schede video basta usare il sito Geforce GTX 670 vs Radeon HD 7970 – Performance Comparison Benchmarks @ Hardware Compare Zanmato Rispondi [...]

Gamer PC zusammenstellen - Forum says:

[...] reichen. Ich hab hier einmal einen kleinen Vergleich, der jedoch nur auf den reinen Daten basiert Vergleich 7970 mit 670 Netzteil: Auch eine gute Wahl und dürfte aktuell auch ausreichend sein. Wenn du allerdings [...]

Gaming Computer Build - PCMech Forums says:

[...] It all depends on what gaming style you enjoy. If you like FPS games (Crysis, BF3 and Halo types) then Nvidia for sure and I would personally get the 670 over the 580 Look Here If you like Sports, racing simulators, MMORPG or Strategy games then ATI/AMD is the way to go and $20 more for the card would be worth it. Look Here [...]

Anonymous says:

NVIDIA GeForce GTX 670 2GB GDDR5 is better than AMD Radeon HD 7970 3GB GDDR5, because it has a lower power consumption and it can be overclocked to the same performance as NVIDIA GeForce GTX 680 2GB GDDR5.

barraCUDA says:

I personally went with the GTX670 for this "round", for a few reasons... However, I have had a lot of experience with the 7970 as well, and here are some notes.
First, the two systems:
My rig's important specs are 3x EVGA GTX670 FTW 2GB models in 3-Way SLI running at 1384Mhz GPU/7450Mhz-Effective VRAM, a GT650Ti 2GB for PhysX/Folding, a Rampage 4 Extreme, i7 3930K @ 4.8Ghz, 16GB DDR3-2133 9-11-10-28 (4x4GB OC'd @ DDR3-2400 9-12-11-28), Samsung 830 256GB SSD (OS/Boot), Plextor M5P-Extreme 256GB SSD (Games), 2x WD VR 1TB RAID0, 1x WD RE 4TB, (had to move my RAID setup to a dedicated Home Media Server), X-Fi Titanium HD, all custom water cooled (CPU/GPU's/MB/RAM).
My friend's rig consists of 2x MSI R7970 Lightning GHz-Edition 3GB in CF-X (1250Mhz Core/6600Mhz-Effective VRAM), a 650Ti 1GB for PhysX, Gigabyte GA-Z77X-UP7 Motherboard, i7 3770K @ 4.9Ghz, 16GB Samsung 30nm RAM DDR3-1600 11-11-11-30 1.35V (4x4GB OC'd @ DDR3-2600 11-13-11-31 1.585v), Samsung 840Pro 128GB (OS/Boot/Apps), 180GB Intel 330 SSD, LSI 9680-16i RAID Controller Card w 4GB Cache and BBU, 4x Seagate 7200.13 2TB HDD's RAID0, 6x Hitachi Ultrastar 500GB RAID0, 4x Seagate 3TB 7200rpm HDD RAID5+HS, and the CPU and GPU's are custom water-cooled (MB Block is on its way).

We have friendly benchmarking competitions all the time, but we also game a lot.
Occasionally, I will disable one of my 670FTW's so that we have "fair" 2vs2-card benchmarks (one reason that we both have dedicated PhysX cards, aside from BL2).

Despite his advantage in VRAM, we both end up with scores that are within 5% of one another, max, typically around 2%. This includes Heaven3.0, 3dMark11, etc. I do have a significant advantage in SuperPi/HyperPi/CineBench because of having 50% more CPU Cores/Threads, even if they are running 100Mhz slower and at a 5-10% IPC disadvantage.

When it comes to gaming, though, it's almost impossible to differentiate the two. Both of us can keep Far Cry 3 at >100fps at ALL times running everything on ULTRA at either 1080p or 1440p; Crysis and Crysis2 are >75fps with everything completely maxxed and with the DX11 Texture pack at the same resolutions, and even Metro2033 at 1440p with everything completely maxxed out will not drop below 50fps (Avg 78/88, Min 51/52, Max 131/178; mine scores a bit higher, but the minimum framerate is very similar... likely due to his VRAM advantage).

At the end of the day, we each paid EXACTLY the same amount of money for our cards, a whole $1.28 difference between the two GPU setups.

(FWIW, I have used 3-way EVGA GTX680 4GB SLI, and it was all of 1.2-2.6% faster than my current setup, yet cost $500 more!)

John Cock says:

Thy Asscheeks Are Thy Friend

Your email address will not be published. Required fields are marked *


WordPress Anti-Spam by WP-SpamShield