Nvidia

Nvidia RTX 5000 Series

by

The Nvidia RTX 5000 series has been announced.  We are disappointed by the lack of VRAM upgrades in several of the models.  However, we are not surprised as Nvidia has a good hold on the market and can get away with a lot.  We really want to see some more affordable models with 24gb of RAM.  Instead, we get only 16gb of RAM unless you go up to the $1,999 32GB of RAM flagship RTX 5090 video card.

The following table summarizes our current knowledge:

Model Cuda Ram Memory Speed Bus Power
RTX 5090 21,760 32GB 28Gbps 512-bit 600W
RTX 5070 Ti 8,960 16GB 28Gbps 256-bit 285W
RTX 5070 6,400 12GB 28Gbps 192-bit 250W
RTX 5060 Ti TBC 16GB 28Gbps ? 200W
RTX 5060 TBC 8GB 28Gbps ? ?
RTX 5080 10,752 16GB 30Gbps 256-bit 400W

 

Discuss on the Gear Rumors Forums

Nvidia Digits $3000 AI Supercomputer

by

With VRAM being king in the AI world, people started buying Apple computers as their RAM is integrated and used both by the CPU and GPU.  How else would you get 128gb of RAM so “cheaply” where a good portion of that could be used by the GPU?  This seemed like a more cost effective way of getting a large pool of VRAM, and as a plus you could do so in a smaller size with less power and technical hurdles (like when trying to build a 4 GPU rig).

Nvidia surprised us by announcing an ARM based computer for $3000 that comes with 128gb of unified memory at a price about $1700 less than you would pay from Apple.

This computer is said to be able to handle models up to 200 billion parameters in size, which is simply great.

We hope this product won’t be in so high demand that it will be hard to get a hold of.

Discuss on the Gear Rumors Forum

Nvidia RTX 5000 Blackwell Series

by

Rumors seem to be indicating that we might see the Nvidia RTX 5000 Blackwell cards in 2024.  We certainly hope this is true.  Not much is known about these cards yet although we are seeing several sources say there could be a 40% generational improvement which would be quite significant.

Discuss on the Gear Rumors Forums