For this reason, Nvidia did not want to pass up the opportunity and will fill the gaming market with four models for desktop and two more for laptops, protecting itself like it has never seen against its two rivals.
RTX 3090 TI
It is without a doubt the most powerful gaming graphics card in history and the ultimate GPU on the planet. More cores, faster memory and higher overall performance as well as excessive power consumption, NVIDIA is of course going big.
As expected, it will be a GPU of 628 mm2 made by samsung a 8 nm and will have 28 billion transistors being the fully unlocked chip and the second largest GPU in size after Turing TU102.
What we will have in return are 84 SM with a total of 10,752 cores which will go at a frequency of 1.560MHz with a boost of 1.860MHz. It will of course be a weak GPU PCIe 4.0 x16 (no PCIe 5.0 as we said) which will integrate to its assets a 384-bit bus which will link up with 24 GB of GDDR6X at 1325 MHz or 21.2 Gbit / s efficient to give an impressive bandwidth of 1018 Gb / s, that is to say 1 TB / s.
This bandwidth should feed the 336 TMU, 112 ROP, 336 Tensor Colors and 84 RT Cores that will be directly linked to the 6 MB of L2 that Integra. Regarding overall performance, NVIDIA provides a round number: 40 TFLOPSwhich would mean being around 10% above the current RTX 3090.
The most impressive thing without a doubt is that your consumption or TGP will go to nothing less than 450 watts, 100 more to push the Ampere architecture to its limits. Therefore, you need the 12-pin connector developed by the company or, if it ends up reaching assemblers, three 8-pin connectors. Its price will be $ 1,499 and that will come in this month of january, especially the 27.
RTX 3080 12 GB
She will take care of getting up and looking over her shoulder at what Intel presents, since it is called to be the queen of high-end, renewed that yes. There are also no surprises here, because with 70 SM we have a total of 8960 shaders, where the biggest update is in its GDDR6X memory, which now reaches 12 GB at 19 Gbit / s, so with its 384-bit bus we will have 912 GB / s representing the biggest performance increase compared to its predecessor in this section (+ 20%).
Like its big sister, NVIDIA has not revealed its MSRP price, although it looks like it will arrive in the first quarter of this year.
RTX 3070 Ti 16 GB
Regarding this NVIDIA GPU, it was very brief, because despite the fact that it has the same chip and the same features as its original 8 GB version, the card is updated with twice the VRAM capacity and also passes to modules. GDDR6X at 21 Gbps, which with its 256-bit bus will give us no less than 672 GB / s. This should increase your performance over a period of 5% about.
Unfortunately, there are no official prices, but there is a release date, especially the June 10.
8 GB RTX 3050
There are also no surprises here compared to what leaked at the time, since this RTX 3050 will be the company’s lowest-end GPU and will have a GA106-150 chip at its core with 2560 Shaders, 20 RT Cores, 80 Tensor cores and 8 GB of GDDR6 at 14 Gbps connected to a 128-bit bus.
Your TGP is encrypted at 130 watts and will cost $ 249 as MSRP, entering market on 27 of this month of January.
Most powerful for the gaming laptop market
If those presentations weren’t enough, NVIDIA also showed the benefits of its two new gaming GPUs for laptops, covering the lineup with what Intel would have prepared for the market and also leaving behind AMD and its new GPUs on show for an hour. previously.
Obviously, these are two high-end laptop models, so the prices will be prohibitive for most mortals. What do we have on our hands with these GPUs?
RTX 3080 Ti Mobile
We start with a brand new chip called GA103 which is still manufactured by Samsung in 8 nm which integrates nothing less than 58 SM with 7.424 Shaders at a base frequency of 1.125 MHz with a Boost that will reach 1.590MHz.
GDDR6X will not be included here due to its high consumption, so we remain in the entire range of laptops as before with one caveat, the 16 GB of GDDR6 it incorporates will have a speed of 16 Gbit / s, being the only GPU that will benefit from this improvement.
This will give us no less than 512 GB / s of bandwidth with a consumption of 175 watts, on par with the best GPU for gaming laptops ever. There will of course be a Max-Q and Max-P version which will be chosen according to the needs of the manufacturer. It will be launched during this first quarter of 2021.
RTX 3070 Ti Mobile
This GPU is an update of lesser caliber than its already presented sister. We went from 40 SM to 46 SM in this new model, which means that we will have 5.888 Shaders at 1 035 MHz in base clock with a boost of 1.485 MHz.
The surprising thing is that the rest of the specs remain intact, including consumption, which is surprising when you take into account the increased performance of Shaders. It is true that it loses frequency in Boost compared to the RTX 3070, but these are still surprising 80-125 watts. By scrapped it will arrive with the Max-Q and Max-P versions in this first quarter of the year.
Table of Contents
