Nvidia Tightens Up Midrange Performance With The $300 GeForce 7950GT
Tech Specs
The GeForce 7950GT should exhibit performance gains over the GeForce 7900GT just from faster clock frequencies - it has all of the same core components as the 7900GT, but is running them 100 MHz faster. The vertex shaders are operating at 570 MHz, and the rest of the core is at 550 MHz. At 700 MHz (1.4 GHz DDR), the memory is 40 MHz faster than the GeForce 7900GT. Below is a chart that lays out all of the 7000 series cards and their Nvidia reference design specifications.
NVIDIA GeForce Reference Specifications | |||||
---|---|---|---|---|---|
Row 0 - Cell 0 | 7950GX2 | 7900GTX | 7950GT | 7900GT | 7900GS |
Process Technology (nm) | 90 | 90 | 90 | 90 | 90 |
Processor Core | G71 | G71 | G71 | G71 | G71 |
Number of Processors | 2 | 1 | 1 | 1 | 1 |
Number of Transistors per Processor (Millions) | 278 | 278 | 278 | 278 | 278 |
Vertex Frequency (MHz) | 500 | 700 | 570 | 470 | 470 |
Core Frequency (MHz) | 500 | 650 | 550 | 450 | 450 |
Memory Clock (MHz) | 600 | 800 | 700 | 660 | 660 |
DDR Rate (MHz) | 1200 | 1600 | 1400 | 1320 | 1320 |
Vertex Shaders (#) | 16 | 8 | 8 | 8 | 7 |
Pixel Shaders (#) | 48 | 24 | 24 | 24 | 20 |
ROPs (#) | 32 | 16 | 16 | 16 | 16 |
Memory Interface (bit) | 256 | 256 | 256 | 256 | 256 |
Frame Buffer Size per Processor (MB) | 512 | 512 | 256 | 256 | 256 |
Memory Bandwidth (GB/sec) per processor | 38.4 | 51.2 | 44.8 | 42.24 | 42.24 |
Vertices/Second (Millions) | 2000 | 1400 | 1140 | 940 | 822.5 |
Pixel Fill Rate (# ROPs x clk) in Billions/sec | 16 | 10.4 | 8.8 | 7.2 | 7.2 |
Texture Fill Rate (# pixel pipes x clk) in Billions/sec | 24 | 15.6 | 13.2 | 10.8 | 9 |
RAMDACs (MHz) | 400 | 400 | 400 | 400 | 400 |
Bus Technology | PCI Express | PCI Express | PCI Express | PCI Express | PCI Express |
Row 20 - Cell 0 | |||||
Row 21 - Cell 0 | 7800GTX 512 | 7800GTX | 7800GT | 7800GS AGP | 7600GT |
Process Technology (nm) | 110 | 110 | 110 | 110 | 90 |
Processor Core | G70 | G70 | G70 | G70 | G73 |
Number of Processors | 1 | 1 | 1 | 1 | 1 |
Number of Transistors per Processor (Millions) | 302 | 302 | 302 | 302 | 178 |
Vertex Frequency (MHz) | 550 | 470 | 440 | 375 | 560 |
Core Frequency (MHz) | 550 | 430 | 400 | 375 | 560 |
Memory Clock (MHz) | 850 | 600 | 500 | 600 | 700 |
DDR Rate (MHz) | 1700 | 1200 | 1000 | 1200 | 1400 |
Vertex Shaders (#) | 8 | 8 | 7 | 6 | 5 |
Pixel Shaders (#) | 24 | 24 | 20 | 16 | 12 |
ROPs (#) | 16 | 16 | 16 | 8 | 8 |
Memory Interface (bit) | 256 | 256 | 256 | 256 | 128 |
Frame Buffer Size per Processor (MB) | 512 | 256 | 256 | 256 | 256 |
Memory Bandwidth (GB/sec) per processor | 54.4 | 38.4 | 32 | 38.4 | 22.4 |
Vertices/Second (Millions) | 1100 | 940 | 770 | 562.5 | 700 |
Pixel Fill Rate (# ROPs x clk) in Billions/sec | 8.8 | 6.88 | 6.4 | 3 | 4.48 |
Texture Fill Rate (# pixel pipes x clk) in Billions/sec | 13.2 | 10.32 | 8 | 6 | 6.72 |
RAMDACs (MHz) | 400 | 400 | 400 | 400 | 400 |
Bus Technology | PCI Express | PCI Express | PCI Express | AGP 8X | PCI Express |
Stay On the Cutting Edge: Get the Tom's Hardware Newsletter
Get Tom's Hardware's best news and in-depth reviews, straight to your inbox.
Leaker claims Nvidia RTX 5090 mobile could offer 24GB of memory — Nvidia expected to employ 3GB GDDR7 modules to increase VRAM capacities on the same 256-bit interface
Nvidia RTX 5060 laptop GPU beats 4060 laptop by over 30% in leaked benchmark — performance gap stretches almost to 70% with the RTX 3060 mobile