GPU Wiki

Riva 128 vs Riva 128ZX

Swap Clear comparison
Edit GPUs
Add a third or fourth GPU

Key Differences

  • No game was measured on every one of these cards on one shared test system, so there is no speed comparison to make.
  • Riva 128ZX has the most memory: 8 MB against 4 MB on standard RIVA 128 boards.
Spec Riva 128 Riva 128ZX
Overview
Product name NVIDIA RIVA 128 NVIDIA Riva 128ZX
Architecture NV3 / RIVA fixed function 2D, video, and 3D accelerator NV3 / RIVA
Generation / family RIVA Series RIVA Series
GPU codename NV3 NV3
Launch date 1997-08-25 1998-02-23
Launch MSRP Board pricing varied by partner Unknown
Market segment Consumer desktop 2D/3D graphics accelerator Desktop in RIVA Series
Board type / entry type Reference GPU Reference GPU
GPU
Process node SGS-Thomson 350 nm class ST 350 nm
Die size About 90 mm2 71 mm²
Transistor count About 3.5 to 4 million, depending on source wording 4 million
GPU chip NV3 NV3
SM / multiprocessor count 0 0
CUDA Cores 0 0
ROPs 1 1
TMUs 1 1
Clocks
Base / Core clock 100 MHz 100 MHz
Boost clock 100 MHz 100 MHz
Memory clock 100 MHz 100 MHz
Memory
Memory size 4 MB on standard RIVA 128 boards 8 MB
Memory type SGRAM SDR
Memory bus width 128-bit 128-bit
Memory bandwidth 1.6 GB/s 1.6 GB/s
Bus / Interface
Bus interface AGP 1x or PCI AGP 2x
AGP / PCI / PCIe distinction AGP Legacy graphics interface. AGP Legacy graphics interface.
Power
TDP / TBP / board power About 4 W class Unknown
Recommended PSU 200 W 200 W
Power connectors None None
Slot width Single slot Single-slot
Cooling type Passive heatsink on most boards Unknown
Display / Media
Display outputs 15-pin VGA on typical boards.
Some partner cards added video input or video output. 		1x VGA
Features
DirectX support Direct3D 5.0 class DirectX 5.0
OpenGL support OpenGL 1.0 class, with period driver limitations OpenGL 1.0
Vulkan support Not supported Unknown
CUDA support Not applicable Unknown
Muted cells indicate unavailable, not applicable, or not directly comparable specifications.

Frame rates compare runs of the same game at the same quality preset on one test system. DLAA counts as native resolution, because it applies anti-aliasing at full render resolution rather than upscaling; DLSS and FSR upscaling are never counted as native. Where a capture named ray tracing without a readable on or off value, it is read as on.