Rubin Ultra GPU vs P104-100

Primary details

GPU architecture, market segment, value for money and other general parameters compared.

Place in the ranking	548	not rated
Place by popularity	not in top-100	not in top-100
Architecture	Pascal (2016−2021)	Rubin Ultra (2027)
GPU code name	GP104	GR110
Market segment	Workstation	Workstation
Release date	12 December 2017 (8 years ago)	2027

Detailed specifications

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

Pipelines / CUDA cores	1920	57344
Core clock speed	1607 MHz	700 MHz
Boost clock speed	1733 MHz	2267 MHz
Number of transistors	7,200 million	670,000 million
Manufacturing process technology	16 nm	3 nm
Power consumption (TDP)	no data	3600 Watt
Texture fill rate	208.0	4,062.5
Floating-point processing power	6.655 TFLOPS	260 TFLOPS
ROPs	64	24
TMUs	120	1792
Tensor Cores	no data	1792
L1 Cache	720 KB	112 MB
L2 Cache	2 MB	128 MB

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 1.0 x4	PCIe 6.0 x16
Length	267 mm	no data
Width	2-slot	SXM Module
Supplementary power connectors	1x 8-pin	no data

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	GDDR5X	HBM4e
Maximum RAM amount	4 GB	1024 GB
Memory bus width	256 Bit	32768 Bit
Memory clock speed	1251 MHz	1900 MHz
Memory bandwidth	320.3 GB/s	31.1 TB/s
Resizable BAR	-	+

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

No outputs

API and SDK support

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

DirectX	12 (12_1)	N/A
Shader Model	6.8	N/A
OpenGL	4.6	N/A
OpenCL	3.0	3.0
Vulkan	1.4	N/A
CUDA	6.1	10.7
DLSS	-	+

Pros & cons summary

---

Maximum RAM amount	4 GB	1024 GB
Chip lithography	16 nm	3 nm