Tesla T4 vs Tesla P4

Place in the ranking	241	200
Place by popularity	not in top-100	not in top-100
Power efficiency	21.19	27.34
Architecture	Pascal (2016−2021)	Turing (2018−2022)
GPU code name	GP104	TU104
Market segment	Workstation	Workstation
Release date	13 September 2016 (8 years ago)	13 September 2018 (6 years ago)

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	2560	2560
Core clock speed	886 MHz	585 MHz
Boost clock speed	1114 MHz	1590 MHz
Number of transistors	7,200 million	13,600 million
Manufacturing process technology	16 nm	12 nm
Power consumption (TDP)	75 Watt	70 Watt
Texture fill rate	178.2	254.4
Floating-point processing power	5.704 TFLOPS	8.141 TFLOPS
ROPs	64	64
TMUs	160	160
Tensor Cores	no data	320
Ray Tracing Cores	no data	40

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).

Interface	PCIe 3.0 x16	PCIe 3.0 x16
Length	168 mm	168 mm
Width	1-slot	1-slot
Supplementary power connectors	None	None

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	GDDR6
Maximum RAM amount	8 GB	16 GB
Memory bus width	256 Bit	256 Bit
Memory clock speed	1502 MHz	1250 MHz
Memory bandwidth	192.3 GB/s	320.0 GB/s

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

No outputs

API compatibility

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

DirectX	12 (12_1)	12 Ultimate (12_1)
Shader Model	6.4	6.5
OpenGL	4.6	4.6
OpenCL	1.2	1.2
Vulkan	1.2.131	1.2.131
CUDA	6.1	7.5

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.

Tesla P4 23.16

Tesla T4 27.89
+20.4%

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.

Tesla P4 8919

Tesla T4 10744
+20.5%

Gaming performance

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

Pros & cons summary

Performance score	23.16	27.89
Recency	13 September 2016	13 September 2018
Maximum RAM amount	8 GB	16 GB
Chip lithography	16 nm	12 nm
Power consumption (TDP)	75 Watt	70 Watt

Tesla T4 has a 20.4% higher aggregate performance score, an age advantage of 2 years, a 100% higher maximum VRAM amount, a 33.3% more advanced lithography process, and 7.1% lower power consumption.

The Tesla T4 is our recommended choice as it beats the Tesla P4 in performance tests.

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