Samsung has been showing off its next-generation GDDR7 memory at NVIDIA's GTC 2024 event, showing off 28Gbps and 32Gbps at the show.
The new picture of Samsung's next-gen GDDR7 memory comes from HardwareLuxx, which spotted the GDDR7 memory at GTC 2024. NVIDIA unveiled its next-gen Blackwell GPUs at GTC 2024, but they were using ultra-fast HBM3E memory, while GDDR7 will be baked into the next-gen Blackwell-based GeForce RTX 50 series that will debut later this year.
Samsung says that its new GDDR7 memory has 20% better power efficiency at just 1.1V -- compared to the industry standard 1.2V -- much lower than the power-hungry 1.35V that GDDR6X memory modules consume.
Not only that, but Samsung says that the optimized circuit design leads to a 70% decrease in thermal resistance compared to GDDR6. This means we'll get super-fast VRAM inside of the RTX 50 series GPUs, but cards won't cook because GDDR7 memory temps are so high -- like the RTX 3090 could cook eggs on the back of it with the double-sided GDDR6X memory modules running at such high temperatures.
- Read more: Samsung to show off the world's fastest GDDR7 memory at 37Gbps for next-gen GPUs at ISSCC
- Read more: Samsung next-gen GDDR7 memory: 50% less stand-by power than GDDR6
- Read more: Samsung announces next-gen GDDR7 memory, with insane 36Gbps bandwidth
- Read more: Samsung GDDR7: 36Gbps bandwidth confirmed, uses PAM3 signaling
- Read more: Samsung teases next-gen GDDR7 memory: insane 32Gbps (GDDR6X = 21Gbps)
The fastest deployment of GDDR6 memory right now is inside of NVIDIA's new GeForce RTX 4080 SUPER graphics card, which has 16GB of GDDR6X memory clocked at 23Gbps bandwidth on a 256-bit memory bus. Remember, the flagship GeForce RTX 4090 features 24GB of GDDR6X, but it's only clocked at 21Gbps. GDDR7 memory is going to be really exciting for next-gen GPUs, to say the least.
Samsung GDDR7 memory features:
- Capacity: 16Gb (2GB)
- Speed: 32Gbps (PAM3)
- Power: 20% improved power efficiency
- Other features: Improved power efficiency with low voltage of 1.1V (surpassing industry standard of 1.2V), reduced heat in high-speed operation with new package materials, optimized circuit design (70% decrease in thermal resistance compared to GDDR6)