ATI Radeon HD 4890 vs Tesla K8

Place in the ranking	520	769
Place by popularity	not in top-100	not in top-100
Cost-effectiveness evaluation	no data	0.41
Power efficiency	7.06	1.50
Architecture	Kepler (2012−2018)	TeraScale (2005−2013)
GPU code name	GK104	RV790
Market segment	Workstation	Desktop
Release date	16 September 2014 (11 years ago)	2 April 2009 (17 years ago)
Launch price (MSRP)	no data	$249

Cost-effectiveness evaluation

The higher the ratio, the better. We use the manufacturer's recommended prices.

no data

Performance to price scatter graph

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	1536	800
Core clock speed	693 MHz	850 MHz
Boost clock speed	811 MHz	no data
Number of transistors	3,540 million	959 million
Manufacturing process technology	28 nm	55 nm
Power consumption (TDP)	100 Watt	190 Watt
Texture fill rate	103.8	34.00
Floating-point processing power	2.491 TFLOPS	1.36 TFLOPS
ROPs	32	16
TMUs	128	40
L1 Cache	128 KB	160 KB
L2 Cache	512 KB	256 KB

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 2.0 x16	PCIe 2.0 x16
Length	241 mm	241 mm
Width	1-slot	2-slot
Supplementary power connectors	1x 6-pin	2x 6-pin

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	8 GB	1 GB
Memory bus width	256 Bit	256 Bit
Memory clock speed	1250 MHz	975 MHz
Memory bandwidth	160.0 GB/s	124.8 GB/s

Connectivity and outputs

This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.

Display Connectors

No outputs

2x DVI, 1x S-Video

API and SDK support

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

DirectX	12 (11_0)	10.1 (10_1)
Shader Model	5.1	4.1
OpenGL	4.6	3.3
OpenCL	1.2	1.1
Vulkan	1.1.126	N/A
CUDA	3.0	-

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	9.17	3.69
Recency	16 September 2014	2 April 2009
Maximum RAM amount	8 GB	1 GB
Chip lithography	28 nm	55 nm
Power consumption (TDP)	100 Watt	190 Watt

Tesla K8 has a 149% higher aggregate performance score, an age advantage of 5 years, a 700% higher maximum VRAM amount, a 96% more advanced lithography process, and 90% lower power consumption.

The Tesla K8 is our recommended choice as it beats the Radeon HD 4890 in performance tests.

Be aware that Tesla K8 is a workstation graphics card while Radeon HD 4890 is a desktop one.