New Account ?
Register Here
Unable to Sign In?
Visit Support
›
Micron graphics leadership
Graphics memory comparison
Graphics memory | GDDR7 | GDDR6X | GDDR6 |
---|---|---|---|
Workload | Graphics AI Inference Accellerator | Graphics AI Inference Accellerator | Graphics AI Inference Accellerator |
Platform | (TBD) | GeForce® RTX™ 4090 GeForce® RTX™ 4080 Super | Titan RTX RX 7900 RTX |
Number of placements | 12 | 12 | 12 |
GB/s/pin | 28-32 | 19-24 | 14-20 |
GB/s/placement | 112-128 | 76-96 | 56-80 |
GB/s/system | 1344-1536 | 912-1152 | 672-960 |
Configuration (example) | 384 IO | 384 IO | 384 IO |
Frame buffer of typical system | 24GB | 24GB | 24GB |
Average device power (pJ/bit) | 5 | 7.25 | 7.5 |
Typical IO channel | PCB (P2P SM) | PCB (P2P SM) | PCB (P2P SM) |