GPU Wiki

Compare NVIDIA GPUs

Select up to four GPU Wiki entries to compare matching specifications row by row. The comparison uses the site database and keeps generation-specific specs visibly separate.

Swap Clear
Row order:
Muted cells indicate unavailable, not applicable, or not directly comparable specifications.
Spec GeForce 256
Overview
Product name NVIDIA GeForce 256
Architecture Celsius / NV10 fixed-function DirectX 7 GPU with hardware transform and lighting
Generation / family GeForce 256
GPU codename NV10
Launch date 1999-10-11
Launch MSRP About $249 to $299 for early 32 MB boards, depending on partner and memory type
Market segment High-end desktop gaming and enthusiast 3D graphics
Board type / entry type Reference GPU
GPU
Process node TSMC 220 nm
Die size About 139 mm2
Transistor count About 17 million in GPU database sources; some launch-era coverage quoted nearly 23 million for the broader GeForce 256 chip description
GPU chip NV10
SM / multiprocessor count 0
CUDA Cores 0
Clocks
Base / Core clock 120 MHz
Boost clock Not applicable
Memory clock 166 MHz SDR on SDR boards; 150 MHz DDR clock, 300 MT/s effective, on DDR boards
Memory
Memory size 32 MB typical; some later board-partner variants used 64 MB
Memory type SDR SDRAM or SGRAM on early cards; DDR SDRAM or DDR SGRAM on later DDR cards
Memory bus width 128-bit
Memory bandwidth 2.656 GB/s on SDR boards; 4.8 GB/s on DDR boards
Bus / Interface
Bus interface AGP 2x / AGP 4x depending on board and platform
AGP / PCI / PCIe distinction AGP Legacy graphics interface.
Power
TDP / TBP / board power About 12 to 13 W class, depending on SDR or DDR board design
Power connectors None
Slot width Single slot
Cooling type Small active fan cooler on many boards; some partner designs vary
Display / Media
Display outputs 15-pin VGA on typical boards.
Some board-partner designs exposed or provisioned DVI, S-Video or TV-out.
Features
DirectX support DirectX 7.0 with hardware transform and lighting
OpenGL support OpenGL 1.2 class in period drivers
Vulkan support Not supported
CUDA support Not applicable. This predates CUDA.