Tesla P40 vs GeForce GTX 970

Place in the ranking	219	177
Place by popularity	78	not in top-100
Cost-effectiveness evaluation	13.65	2.77
Power efficiency	11.68	8.80
Architecture	Maxwell 2.0 (2014−2019)	Pascal (2016−2021)
GPU code name	GM204	GP102
Market segment	Desktop	Workstation
Release date	19 September 2014 (10 years ago)	13 September 2016 (8 years ago)
Launch price (MSRP)	$329	$5,699

Cost-effectiveness evaluation

Performance to price ratio. The higher, the better.

GTX 970 has 393% better value for money than Tesla P40.

Detailed specifications

General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.

Pipelines / CUDA cores	1664	3840
Core clock speed	1050 MHz	1303 MHz
Boost clock speed	1178 MHz	1531 MHz
Number of transistors	5,200 million	11,800 million
Manufacturing process technology	28 nm	16 nm
Power consumption (TDP)	148 Watt	250 Watt
Maximum GPU temperature	98 °C	no data
Texture fill rate	122.5	367.4
Floating-point processing power	3.92 TFLOPS	11.76 TFLOPS
ROPs	56	96
TMUs	104	240

Form factor & compatibility

Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).

Bus support	PCI Express 3.0	no data
Interface	PCIe 3.0 x16	PCIe 3.0 x16
Length	267 mm	267 mm
Height	4.376" (11.1 cm)	no data
Width	2-slot	2-slot
Recommended system power (PSU)	500 Watt	no data
Supplementary power connectors	2x 6-pin	8-pin EPS
SLI options	+	-

VRAM capacity and type

Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.

Memory type	GDDR5	GDDR5
Maximum RAM amount	4 GB	24 GB
Memory bus width	256 Bit	384 Bit
Memory clock speed	7.0 GB/s	1808 MHz
Memory bandwidth	224 GB/s	347.1 GB/s
Shared memory	-	-

Connectivity and outputs

Types and number of video connectors present on the reviewed GPUs. As a rule, data in this section is precise only for desktop reference ones (so-called Founders Edition for NVIDIA chips). OEM manufacturers may change the number and type of output ports, while for notebook cards availability of certain video outputs ports depends on the laptop model rather than on the card itself.

Display Connectors	Dual Link DVI-I, HDMI 2.0, 3x DisplayPort 1.2	No outputs
Multi monitor support	4 displays	no data
HDMI	+	-
HDCP	+	-
Maximum VGA resolution	2048x1536	no data
G-SYNC support	+	-
Audio input for HDMI	Internal	no data

Supported technologies

Supported technological solutions. This information will prove useful if you need some particular technology for your purposes.

GameStream	+	-
GeForce ShadowPlay	+	-
GPU Boost	2.0	no data
GameWorks	+	-

API compatibility

List of supported 3D and general-purpose computing APIs, including their specific versions.

DirectX	12 (12_1)	12 (12_1)
Shader Model	6.4	6.7
OpenGL	4.4	4.6
OpenCL	1.2	3.0
Vulkan	1.1.126	1.3
CUDA	+	6.1

Synthetic benchmark performance

Non-gaming benchmark results comparison. The combined score is measured on a 0-100 point scale.

Combined synthetic benchmark score

This is our combined benchmark score. We are regularly improving our combining algorithms, but if you find some perceived inconsistencies, feel free to speak up in comments section, we usually fix problems quickly.

GTX 970 25.06

Tesla P40 31.90
+27.3%

Passmark

This is the most ubiquitous GPU benchmark. It gives the graphics card a thorough evaluation under various types of load, providing four separate benchmarks for Direct3D versions 9, 10, 11 and 12 (the last being done in 4K resolution if possible), and few more tests engaging DirectCompute capabilities.

GTX 970 9636

Tesla P40 12266
+27.3%

Gaming performance

Let's see how good the compared graphics cards are for gaming. Particular gaming benchmark results are measured in FPS.

Average FPS across all PC games

Here are the average frames per second in a large set of popular games across different resolutions:

Full HD	80 −25%	100−110 +25%
1440p	52 −25%	65−70 +25%
4K	40 −25%	50−55 +25%

Cost per frame, $

1080p	4.11 +1286%	56.99 −1286%
1440p	6.33 +1286%	87.68 −1286%
4K	8.23 +1286%	113.98 −1286%

GTX 970 has 1286% lower cost per frame in 1080p
GTX 970 has 1286% lower cost per frame in 1440p
GTX 970 has 1286% lower cost per frame in 4K

FPS performance in popular games

Full HD
Low Preset

Counter-Strike 2	45−50 −17%	55−60 +17%
Cyberpunk 2077	50−55 −17.6%	60−65 +17.6%
Elden Ring	80−85 −22%	100−105 +22%

Full HD
Medium Preset

Battlefield 5	75−80 −23.4%	95−100 +23.4%
Counter-Strike 2	45−50 −17%	55−60 +17%
Cyberpunk 2077	50−55 −17.6%	60−65 +17.6%
Forza Horizon 4	110−120 −27.3%	140−150 +27.3%
Metro Exodus	65−70 −21.2%	80−85 +21.2%
Red Dead Redemption 2	55−60 −27.3%	70−75 +27.3%
Valorant	100−110 −18.8%	120−130 +18.8%

Full HD
High Preset

Battlefield 5	75−80 −23.4%	95−100 +23.4%
Counter-Strike 2	45−50 −17%	55−60 +17%
Cyberpunk 2077	50−55 −17.6%	60−65 +17.6%
Dota 2	47 −17%	55−60 +17%
Elden Ring	80−85 −22%	100−105 +22%
Far Cry 5	75−80 −23.4%	95−100 +23.4%
Fortnite	120−130 −27%	160−170 +27%
Forza Horizon 4	110−120 −27.3%	140−150 +27.3%
Grand Theft Auto V	71 −26.8%	90−95 +26.8%
Metro Exodus	16 −12.5%	18−20 +12.5%
PLAYERUNKNOWN'S BATTLEGROUNDS	63 −27%	80−85 +27%
Red Dead Redemption 2	55−60 −27.3%	70−75 +27.3%
The Witcher 3: Wild Hunt	70 −21.4%	85−90 +21.4%
Valorant	100−110 −18.8%	120−130 +18.8%
World of Tanks	250−260 −16.3%	300−310 +16.3%

Full HD
Ultra Preset

Battlefield 5	60 −25%	75−80 +25%
Counter-Strike 2	45−50 −17%	55−60 +17%
Cyberpunk 2077	50−55 −17.6%	60−65 +17.6%
Dota 2	85−90 −16.3%	100−105 +16.3%
Far Cry 5	75−80 −23.4%	95−100 +23.4%
Forza Horizon 4	110−120 −27.3%	140−150 +27.3%
PLAYERUNKNOWN'S BATTLEGROUNDS	47 −17%	55−60 +17%
Valorant	100−110 −18.8%	120−130 +18.8%

1440p
High Preset

Dota 2	40−45 −19%	50−55 +19%
Elden Ring	40−45 −25%	55−60 +25%
Grand Theft Auto V	40−45 −19%	50−55 +19%
PLAYERUNKNOWN'S BATTLEGROUNDS	170−180 −25.7%	220−230 +25.7%
Red Dead Redemption 2	21−24 −17.4%	27−30 +17.4%
World of Tanks	160−170 −26.5%	210−220 +26.5%

1440p
Ultra Preset

Battlefield 5	50−55 −20%	60−65 +20%
Counter-Strike 2	21−24 −14.3%	24−27 +14.3%
Cyberpunk 2077	21−24 −14.3%	24−27 +14.3%
Far Cry 5	70−75 −23.3%	90−95 +23.3%
Forza Horizon 4	65−70 −26.9%	85−90 +26.9%
Metro Exodus	55−60 −22.8%	70−75 +22.8%
The Witcher 3: Wild Hunt	35−40 −21.6%	45−50 +21.6%
Valorant	65−70 −25%	85−90 +25%

4K
High Preset

Counter-Strike 2	21−24 −17.4%	27−30 +17.4%
Dota 2	46 −19.6%	55−60 +19.6%
Elden Ring	20−22 −20%	24−27 +20%
Grand Theft Auto V	46 −19.6%	55−60 +19.6%
Metro Exodus	13 −23.1%	16−18 +23.1%
PLAYERUNKNOWN'S BATTLEGROUNDS	61 −23%	75−80 +23%
Red Dead Redemption 2	16−18 −12.5%	18−20 +12.5%
The Witcher 3: Wild Hunt	46 −19.6%	55−60 +19.6%

4K
Ultra Preset

Battlefield 5	24−27 −15.4%	30−33 +15.4%
Counter-Strike 2	21−24 −17.4%	27−30 +17.4%
Cyberpunk 2077	8−9 −25%	10−11 +25%
Dota 2	40−45 −25%	55−60 +25%
Far Cry 5	30−35 −21.2%	40−45 +21.2%
Fortnite	25 −20%	30−33 +20%
Forza Horizon 4	35−40 −18.4%	45−50 +18.4%
Valorant	30−35 −21.2%	40−45 +21.2%

This is how GTX 970 and Tesla P40 compete in popular games:

Tesla P40 is 25% faster in 1080p
Tesla P40 is 25% faster in 1440p
Tesla P40 is 25% faster in 4K

Pros & cons summary

Performance score	25.06	31.90
Recency	19 September 2014	13 September 2016
Maximum RAM amount	4 GB	24 GB
Chip lithography	28 nm	16 nm
Power consumption (TDP)	148 Watt	250 Watt

GTX 970 has 68.9% lower power consumption.

Tesla P40, on the other hand, has a 27.3% higher aggregate performance score, an age advantage of 1 year, a 500% higher maximum VRAM amount, and a 75% more advanced lithography process.

The Tesla P40 is our recommended choice as it beats the GeForce GTX 970 in performance tests.

Be aware that GeForce GTX 970 is a desktop card while Tesla P40 is a workstation one.

Should you still have questions concerning choice between the reviewed GPUs, ask them in Comments section, and we shall answer.