NVIDIA's new T-Rex; hopefully not a flaming dinosaur

Subject: General Tech | December 3, 2018 - 02:36 PM |
Tagged: nvidia, rtx, rtx titan, turing

The new Titan has arrived for the RTX generation and the specs are as impressive as the $2500 price tag.

  • 576 multi-precision Turing Tensor Cores, providing up to 130 teraflops of deep learning performance.
  • 72 Turing RT Cores, delivering up to 11 GigaRays per second of real-time ray-tracing performance.
  • 24GB of high-speed GDDR6 memory with 672GB/s of bandwidth — 2x the memory of previous-generation TITAN GPUs — to fit larger models and datasets.
  • 100GB/s NVIDIA NVLink can pair two TITAN RTX GPUs to scale memory and compute.
  • Incredible performance and memory bandwidth for real-time 8K video editing.
  • VirtualLink port provides the performance and connectivity required by next-gen VR headsets.

From what The Inquirer saw over a weekend of YouTubing, the card sports a gold-coloured shroud, and requires two eight-pin PCIe power connectors.  As of yet we don't have any benchmarks to show how it performs but from the sounds of the PR this will be of more use to content creators than gamers.  However, that is unlikely to stop some from trying it out; stay tuned for more.

View Full Size

"The Titan RTX, dubbed fondly by Nvidia as 'T-Rex', is based on the same Turing architecture as the firm's RTX 2070, 2080 and bork-prone 2080 Ti GPUs, equipping it with 130 teraflops of deep learning performance and 11 GigaRays of ray-tracing performance. "

Here is some more Tech News from around the web:

Tech Talk





Source: The Inquirer

Video News

December 3, 2018 | 06:23 PM - Posted by RashyNuke

I am sure it will be a pretty penny. I love AMD and NVIDIA. That said I hope AMD has a Major break through to drive NVIDIA for awhile. I want the good stuff for a about 1/4 of NNIDIA price. :-)

December 3, 2018 | 06:59 PM - Posted by ProsumerNotProfessionalAtThatPrice (not verified)

Will Nvidia enable NVLink's full coherency on Titan RTX and properly allow more than one Titian/RTX GPU to share the same VRAM memory address space and any L2 cache coherency traffic between multi-GPU Titan RTX configurations.

Nvidia's Quadro SKUs get all the NVLink bells and whistles turned on and Multi-RTX/Quadros looking like one larger logical GPU to any graphics software. That "Real Time" Star Wars demo was a DGX Station and Nvidia's professional GPU SKUs that cost way more than even this Titan RTX.

Nvidia's driver licensing may not even allow for Titan RTX to be used on large production server configurations anyways and NVLink's full feature set may or may not be enabled on this SKU, as Nvidia reserves its best driver/NVLink IP(Fully Enabled) for its more expensive Quadro/Tesla SKUs.

December 4, 2018 | 02:23 PM - Posted by Vince (not verified)

Priced at $2,500 each. Ouch.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

This question is for testing whether you are a human visitor and to prevent automated spam submissions.