NVIDIA's GeForce RTX 3090 Ti graphics card is going to feature some insane custom designs such as MSI's flagship SUPRIM X model. These graphics cards will be the most power consuming and powerful products ever produced for the ultimate enthusiasts out there and also those with huge wallets.
MSI GeForce RTX 3090 Ti SUPRIM X is a monster graphics card: Up to 1900 MHz Extreme Mode Boost Clock & 1000W PSU recommended
Update: It appears that the final specifications may be different from those listed below, as MSI has stated to us.
The NVIDIA GeForce RTX 3090 Ti is everything, extremely powerful, extremely expensive and extremely power-intensive. It's the ultimate graphics card for those who do not care about power consumption or price and just want the best of the best out there, a bit like a flex. The MSI SUPRIM X variant will be a fast custom model, so fast that it will be priced up to $ 4500 US if early lists are something to go by.
The specifications for this beast of a graphics card have been leaked @wxnod and they are simply set SUPREME! in all respects. So starting with the basic specifications, the NVIDIA GeForce RTX 3090 will have a total of 84 SM units on its flagship, resulting in a total of 10,752 CUDA cores (versus 82 SM / 10496 cores on the RTX 3090 Non-Ti). In addition to the CUDA cores, NVIDIA's GeForce RTX 3090 Ti also comes packed with next-generation RT (Ray-Tracing) cores, Tensor cores, and brand new SM or streaming multiprocessor devices. The reference model will run at a base clock speed of 1560 MHz and a boost clock speed of 1860 MHz with a TDP of 450 W.
However, MSI is taking things in stride by pushing the Gaming / Silent Profiles box to 1880 MHz, and there is even a third GPU profile known as 'Extreme Mode', which can be activated through the MSI Center for a 1900 MHz boost clock . The card will also have an increased TDP of 480W. This is an increase of 30 W compared to the reference and an increase of 60 W compared to the RTX 3090 Non-Ti model.
But wait, that's not all, while the standard RTX 3090 SUPRIM X had a recommended PSU requirement of 850 Watts, the RTX 3090 Ti SUPRIM X has a 1000W PSU recommendation. The card will still be powered from a triple 8-pin socket configuration, and the screen outputs will remain the same.
In terms of memory, the GeForce RTX 3090 Ti comes packed with 24GB of memory and also the next generation GDDR6X design. With Micron's latest and greatest graphics memory dies, the RTX 3090 Ti can deliver 21Gbps GDDR6X memory speeds. That along with a 384-bit bus interface will deliver a cumulative bandwidth of 1008 Gbps. Interestingly, while the card has almost the same design as the RTX 3090 SUPRIM X, it appears that there are a few extra components and thermal padding that add 10 grams of extra weight to the card compared to the Non-Ti SUPRIM X model .
The graphics card is expected to be fully unveiled later this month or in a few weeks, so stay tuned for more information such as pricing and availability.
NVIDIA GeForce RTX 30 'SUPER' Series graphics card specifications
|Graphics card name||NVIDIA GeForce RTX 3090 Ti||NVIDIA GeForce RTX 3090||NVIDIA GeForce RTX 3080 Ti||NVIDIA GeForce RTX 3080 12 GB||NVIDIA GeForce RTX 3080||NVIDIA GeForce RTX 3070 Ti 16GB||NVIDIA GeForce RTX 3070 Ti||NVIDIA GeForce RTX 3070||NVIDIA GeForce RTX 3060 Ti||NVIDIA GeForce RTX 3060||NVIDIA GeForce RTX 3050|
|GPU name||Ampere GA102-350?||Ampere GA102-300||Ampere GA102-225||Ampere GA102-220?||Ampere GA102-200||Ampere GA104-400||Ampere GA104-400||Ampere GA104-300||Ampere GA104-200||Ampere GA106-300||Ampere GA106-150|
|Process Node||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm||Samsung 8nm|
|Die Size||628.4 mm2||628.4 mm2||628.4 mm2||628.4 mm2||628.4 mm2||395.2 mm2||395.2 mm2||395.2 mm2||395.2 mm2||276 mm2||276 mm2|
|Transistors||28 billion||28 billion||28 billion||28 billion||28 billion||17.4 billion||17.4 billion||17.4 billion||17.4 billion||13.2 billion||13.2 billion|
|TMUs / ROPs||336/112||328/112||320/112||280/104||272/96||184/96||184/96||184/96||152/80||112/64||TBC|
|Tensor / RT cores||336/84||328/82||320/80||280/70||272/68||184/46||184/46||184/46||152/38||112/28||TBC|
|Basis ur||1560 MHz||1400 MHz||1365 MHz||TBA||1440 MHz||TBA||1575 MHz||1500 MHz||1410 MHz||1320 MHz||1550 MHz|
|Boost ur||1860 MHz||1700 MHz||1665 MHz||TBA||1710 MHz||TBA||1770 MHz||1730 MHz||1665 MHz||1780 MHz||1780 MHz|
|FP32 Compute||40 TFLOPs||36 TFLOPs||34 TFLOPs||TBA||30 TFLOPs||TBA||22 TFLOPs||20 TFLOPs||16 TFLOPs||13 TFLOPs||9.1 TFLOPs|
|RT TFLOPs||74 RFLOPs||69 TFLOPs||67 TFLOPs||TBA||58 TFLOPs||TBA||44 TFLOPs||40 TFLOPs||32 TFLOPs||25 TFLOPs||18.2 TFLOPs|
|Tensor TOPs||TBA||285 TOP||273 TOP||TBA||238 TOP||TBA||183 TOP||163 TOP||192 TOP||101 TOP||72.8 TOP|
|Memory capacity||24 GB GDDR6X||24 GB GDDR6X||12 GB GDDR6X||12 GB GDDR6X||10 GB GDDR6X||16 GB GDDR6X||8 GB GDDR6X||8 GB GDDR6||8 GB GDDR6||12 GB GDDR6||8 GB GDDR6|
|Memory speed||21 Gbps||19.5 Gbps||19 Gbps||19 Gbps||19 Gbps||21 Gbps||19 Gbps||14 Gbps||14 Gbps||16 Gbps||14 Gbps|
|Bandwidth||1008 GB / s||936 GB / s||912 Gbps||912 Gbps||760 GB / s||672 GB / s||608 GB / s||448 GB / s||448 GB / s||384 GB / s||224 GB / s|
|Price (MSRP / FE)||TBD||1499 USD||$ 1199||999 USD?||699 USD||599 USD?||599 USD||499 USD||399 USD||329 USD||249 USD|
|Launch (availability)||January 27, 2022||September 24, 2020||June 3, 2021||11 January 2022||September 17, 2020||Q1 2022?||June 10, 2021||October 29, 2020||December 2, 2020||February 25, 2021||January 27, 2022|