⋮    ⋮  

NVIDIA GeForce RTX 3080 Ti Spotted In HP OEM Driver – GA102 Based Affair With 20GB Of VRAM

Submit

NVIDIA's upcoming RTX 3080 Ti flagship has been spotted in an HP driver courtesy of StefanG3D (via Videocardz). According to the source, RTX  3080 Ti is the final name for the GA102 based 20GB graphics card that NVIDIA is preparing to take out the big Navi lineup.

NVIDIA's upcoming RTX 3080 Ti 20GB graphics card finally starts popping up in drivers

User T4CFantasy also confirmed to Videocardz that the spotted GPU ID indeed belongs to the RTX 3080 Ti:

According to T4CFantasy to whom we reached out, those names are 99% correct. He is sure that 2205 ID belongs to the RTX 3080 Ti because he had already seen traces of this new SKU through the software that he wouldn’t name.

The exact GPU ID can be seen below:
The NVIDIA RTX 3080 Ti  was originally supposed to be based on the PG132 SKU20 (although NVIDIA decided to rename it) and has been in the pipeline for quite some time. While the exact CUDA core count is not yet confirmed, it will have twice the amount of memory when compared to the vanilla RTX 3080. To be clear, even the nomenclature is not confirmed as the SUPER suffix could appear anywhere in the name should Jensen deem it so.
We highly recommend users (that want to buy an NVIDIA card) to go for the RTX 3090 (if they can afford it) or wait for the RTX 3080 Ti/SUPER 20 GB.

With this amount of shading power, you will want to have a complimentary memory buffer. According to Videocardz, the card will be landing in January 2021. Considering the fact that killer apps like Microsoft Flight Simulator 2020 already chomp away all the vRAM the 2080 Ti had to offer, 10 GB is going to feel tiny after a year or two. Next-gen consoles are already driving games with asset-streaming of large worlds, so vRAM is going to start becoming a lot more crucial than before. With AMD rocking a flagship card with 16GB of vRAM, NVIDIA had to reply with their own high vRAM card.

According to Kopite, the RTX 3080 Ti will be based on the GA102 and will have specifications between an RTX 3080 and 3090. The exact chip nomenclature is GA102-250-A1 and will feature a 384 bits bus with GDDR6X memory. The bus-size means NVIDIA will either be using a 12 GB buffer or a 24 GB one. Considering Microsoft Flight Simulator 2020 is already bottlenecked by the 11 GB ram on the RTX 2080 Ti, it would be disappointing to see NVIDIA ship another powerful GPU with a small memory buffer. With RTX IO and asset streaming (Unreal Engine demo for next-generation consoles) becoming a thing in the next year or so, every bit of buffer will help in this paradigm shift.

It is also unclear at this point how the revelation of this new GPU fits into the rumors about NVIDIA planning a move back to TSMC. It does however lend credence to the belief that the RTX 3000 series, at least for now, is staying on the Samsung 8nm process. The company initially faced less than ideal yields and supply constraints at launch but those are expected to significantly improve as we enter into the new year.

This is going to be an insanely powerful card with 34 TFLOPs of power. That said, the current API, driver, and application infrastructure cannot fully take advantage of all this raw power. This is the actual reason why NVIDIA's insanely powerful cards don't scale linearly with TLFOPs. The company essentially made cards that are ahead of their time when compared to the surrounding software ecosystem.

The hard evidence for this lies in the fact that the RTX 3000 series has been experiencing non-linear positive scaling when going down the stack. An RTX 3070 which has slightly more cores than the RTX 2080 Ti beats the former flagship - putting to rest any rumors or allegations about the CUDA cores in the Ampere series not being as strong as the Turing series or misleading blames at architecture design with lacking INT performance.

AMD's big Navi series drops later today and it remains to be seen how the GPU market shapes up as we enter into the holiday season. NVIDIA's pricing is great but it needs to work with developers to fix the API and driver stacks to properly take advantage of the raw performance offered by the Ampere series (fine wine of the highest order) and in the meantime, AMD is going to churn out budget-friendly performance cards with what appears to be ample supply. They have also taken a lot of steps to make sure that the bot-scalping that happened with NVIDIA does not happen at their Radeon RX 6000 series launch.

NVIDIA GeForce RTX 30 Series 'Ampere' Graphics Card Specifications:

Graphics Card NameNVIDIA GeForce RTX 3050NVIDIA GeForce RTX 3050 TiNVIDIA GeForce RTX 3060NVIDIA GeForce RTX 3060 TiNVIDIA GeForce RTX 3070NVIDIA GeForce RTX 3070 TiNVIDIA GeForce RTX 3080NVIDIA GeForce RTX 3080 TiNVIDIA GeForce RTX 3090
GPU NameAmpere GA107Ampere GA107Ampere GA106-300Ampere GA104-200Ampere GA104-300Ampere GA104-400Ampere GA102-200Ampere GA102-225Ampere GA102-300
Process NodeSamsung 8nmSamsung 8nmSamsung 8nmSamsung 8nmSamsung 8nmSamsung 8nmSamsung 8nmSamsung 8nmSamsung 8nm
Die SizeTBATBATBA395.2mm2395.2mm2395.2mm2628.4mm2628.4mm2628.4mm2
TransistorsTBATBATBA17.4 Billion17.4 Billion17.4 Billion28 Billion28 Billion28 Billion
CUDA Cores2048?2560?358448645888614487041024010496
TMUs / ROPs64 / 4080 / 48112 / 64152 / 80184 / 96192/ 96272 / 96320 / 112328 / 112
Tensor / RT Cores64 / 1680 / 20112 / 28152 / 38184 / 46192/ 48272 / 68320 / 80328 / 82
Base ClockTBATBA1320 MHz1410 MHz1500 MHz1575 MHz1440 MHz1365 MHz1400 MHz
Boost ClockTBATBA1780 MHz1665 MHz1730 MHz1770 MHz1710 MHz1665 MHz1700 MHz
FP32 ComputeTBATBA12.7 TFLOPs16.2 TFLOPs20 TFLOPs22 TFLOPs30 TFLOPs34 TFLOPs36 TFLOPs
RT TFLOPsTBATBA25.4 TFLOPs32.4 TFLOPs40 TFLOPs42 TFLOPs58 TFLOPs67 TFLOPs69 TFLOPs
Tensor-TOPsTBATBA101 TOPs129.6 TOPs163 TOPs174 TOPs238 TOPs273 TOPs285 TOPs
Memory Capacity4 GB GDDR6?4 GB GDDR6?12 GB GDDR68 GB GDDR68 GB GDDR68 GB GDDR6X10 GB GDDR6X12 GB GDDR6X24 GB GDDR6X
Memory Bus128-bit128-bit192-bit256-bit256-bit256-bit320-bit384-bit384-bit
Memory SpeedTBATBA15 Gbps14 Gbps14 Gbps19 Gbps19 Gbps19 Gbps19.5 Gbps
BandwidthTBATBA360 Gbps448 Gbps448 Gbps608 Gbps760 Gbps912 Gbps936 Gbps
TGP~75W~100W170W200W220W290W320W350W350W
Price (MSRP / FE)$149?$199?$329$399 US$499 US$599 US$699 US$1199$1499 US
Launch (Availability)2021?2021?February 2021December 202029th October 202010th June 202117th September 20203rd June 202124th September 2020
Submit