NVIDIA’s Flagship Turing TU102 GPU For GeForce RTX 2080 Ti Detailed – 50% Faster Per Core Performance, 288 TMUs/96 ROPs on Full Die and New Overclocking Features
More details of the NVIDIA Turing TU102 GPU that powers the flagship GeForce RTX 2080 Ti and Quadro RTX 8000/6000 series graphics cards have leaked out over at Videocardz. In an exclusive article, the website discusses information that comes straight from the NVIDIA editor's day, an event which is heavily NDA'd and discusses the architectural and performance aspects of the new GeForce graphics cards coming to the market.
NVIDIA Turing TU102 GPU Powering RTX 2080 Ti Graphics Card Offers A Generational Leap Over Pascal - Full Block Diagram Unveiled
The first details that are talked about are the Turing TU102 GPU block diagram. We have already seen the bare chip and it's massive in terms of die size, the biggest ever **102 GPU that NVIDIA's ever produced. Measuring at 754mm2, the chip comes packed with 18.6 Billion transistors and a totally new architecture design featuring different cores that are operating in tandem to deliver the world's best graphics chips for gamers to date.
The Turing TU102 GPU has 72 Streaming Multiprocessors (SM) featuring 64 CUDA cores each. The full die features 4608 CUDA cores while the GeForce RTX 2080 Ti features 4352 cores. The chip has 576 Tensor cores, 72 RT cores, 36 Geometry Units, 288 Texture Units (TMUs) and 96 ROPs (Raster Operation Units). In addition to the core specs, the chip has 384-bit memory interface supporting a 7 GHz GDDR6 (14 GHz Effective) DRAM design and 2 NVLINK channels. The chip features 6 MB of L2 cache too.
Now since we showed that the GeForce RTX 2080 Ti is based on a cut down TU102 core, the specs are slightly different. We get 68 SMs with 4352 Cores, 544 Tensor cores, 68 RT cores, 34 Geometry Units, 288 TMUs and 96 ROPs. The actual clock speeds are maintained at 1350 MHz base and 1545 MHz boost (1635 MHz OC). The chip features 11 GB of GDDR6 (next-gen) memory featured across a 352-bit bus and clocked at 14 GB/s. This leads to a total bandwidth of 616 GB/s.
NVIDIA GeForce RTX/GTX "Turing" Family:
|Graphics Card Name||NVIDIA GeForce GTX 1650||NVIDIA GeForce GTX 1650 D6||NVIDIA GeForce GTX 1650||NVIDIA GeForce GTX 1660||NVIDIA GeForce GTX 1660 SUPER||NVIDIA GeForce GTX 1660 Ti||NVIDIA GeForce RTX 2060||NVIDIA GeForce RTX 2070||NVIDIA GeForce RTX 2080||NVIDIA GeForce RTX 2080 Ti|
|GPU Architecture||Turing GPU (TU117)||Turing GPU (TU117)||Turing GPU (TU116)||Turing GPU (TU116)||Turing GPU (TU116)||Turing GPU (TU116)||Turing GPU (TU106)||Turing GPU (TU106)||Turing GPU (TU104)||Turing GPU (TU102)|
|Process||12nm FNN||12nm FNN||12nm FNN||12nm FNN||12nm FNN||12nm FNN||12nm FNN||12nm FNN||12nm FNN||12nm FNN|
|Transistors||4.7 Billion||4.7 Billion||6.6 Billion||6.6 Billion||6.6 Billion||6.6 Billion||10.6 Billion||10.6 Billion||13.6 Billion||18.6 Billion|
|CUDA Cores||896 Cores||896 Cores||1280 Cores||1408 Cores||1408 Cores||1536 Cores||1920 Cores||2304 Cores||2944 Cores||4352 Cores|
|GigaRays||N/A||N/A||N/A||N/A||N/A||N/A||5 Giga Rays/s||6 Giga Rays/s||8 Giga Rays/s||10 Giga Rays/s|
|Cache||1.5 MB L2 Cache||1.5 MB L2 Cache||1.5 MB L2 Cache||1.5 MB L2 Cache||1.5 MB L2 Cache||1.5 MB L2 Cache||4 MB L2 Cache||4 MB L2 Cache||4 MB L2 Cache||6 MB L2 Cache|
|Base Clock||1485 MHz||1410 MHz||1530 MHz||1530 MHz||1530 MHz||1500 MHz||1365 MHz||1410 MHz||1515 MHz||1350 MHz|
|Boost Clock||1665 MHz||1590 MHz||1725 MHz||1785 MHz||1785 MHz||1770 MHz||1680 MHz||1620 MHz|
1710 MHz OC
1800 MHz OC
1635 MHz OC
|Compute||3.0 TFLOPs||3.0 TFLOPs||4.4 TFLOPs||5.0 TFLOPs||5.0 TFLOPs||5.5 TFLOPs||6.5 TFLOPs||7.5 TFLOPs||10.1 TFLOPs||13.4 TFLOPs|
|Memory||Up To 4 GB GDDR5||Up To 4 GB GDDR6||Up To 4 GB GDDR6||Up To 6 GB GDDR5||Up To 6 GB GDDR6||Up To 6 GB GDDR6||Up To 6 GB GDDR6||Up To 8 GB GDDR6||Up To 8 GB GDDR6||Up To 11 GB GDDR6|
|Memory Speed||8.00 Gbps||12.00 Gbps||12.00 Gbps||8.00 Gbps||14.00 Gbps||12.00 Gbps||14.00 Gbps||14.00 Gbps||14.00 Gbps||14.00 Gbps|
|Memory Bandwidth||128 GB/s||192 GB/s||192 GB/s||192 GB/s||336 GB/s||288 GB/s||336 GB/s||448 GB/s||448 GB/s||616 GB/s|
|Power Connectors||N/A||N/A||6 Pin||8 Pin||8 Pin||8 Pin||8 Pin||8 Pin||8+8 Pin||8+8 Pin|
|Starting Price||$149 US||$149 US||$159 US||$219 US||$229 US||$279 US||$349 US||$499 US||$699 US||$999 US|
|Price (Founders Edition)||$149 US||$149 US||$159 US||$219 US||$229 US||$279 US||$349 US||$599 US||$799 US||$1,199 US|
|Launch||April 2019||April 2020||November 2019||March 2019||October 2019||February 2019||January 2019||October 2018||September 2018||September 2018|
NVIDIA Turing GPU Packs 50% Better Performance Per Core Than Pascal GPUs
In terms of shading performance which is the direct result of the enhanced core design and GPU architecture revamp, the Turing GPU offers an average uplift of 50% better performance per core compared to Pascal GPUs. In VR games, the shading performance would be a good 2x ahead than what Pascal achieved while many modern gaming titles show a ~50% lead over Pascal with Turing's enhanced core design.
It should be pointed that these are just per core performance gains at the same clock speeds without adding the benefits of other technologies that Turing comes with. That would further increase the performance in a wide variety of gaming applications as we have already seen the gaming performance of a GeForce RTX 2080 to be 50% faster than the GTX 1080 on average and twice as fast with the new DLSS technology.
NVIDIA's New Overclocking Feature Let's The OC Utility Detect The Best Possible Clock Speeds and Voltages For You
Lastly, there's a new overclocking feature being talked about which the new Turing GeForce RTX graphics cards will be able to make use of. To be known as the Scanner (Final Name is still a work-in-progress), the feature will let the OC Utility detect the best clock speeds and voltages for you without the need to do anything. Just run a test through the overclocking utility and you are all set.
The feature is said to be implemented in many Overclocking utilities which will be updated by AIBs such as EVGA, MSI, ASUS, etc but would also be shipping inside new GeForce Experience software which sounds really interesting. This means we may be looking at an updated GeForce experience and software stack with the new cards launch.
NVIDIA GeForce RTX 2080/RTX 2080 Ti Performance Review Embargo Ends 14th September
One more thing to add, it is stated that the performance reviews of the GeForce RTX 2080 Ti and RTX 2080 have an embargo until 14th September so that is when we will be looking at the final reviewers. Also, NVIDIA hasn't shipped the press with working drivers so for the time being, we will only have these official performance figures with us. That gives a 6-day margin for users to get the GeForce RTX products based on their impressions of the reviews.
The GeForce RTX 20 Series Market Availability – Preorder and Shipping Today, On Shelves 20th September
The NVIDIA GeForce RTX 20 series launches today in reference variants first. This time, NVIDIA has already given the green light to their manufacturers to announce custom cards soon after the reference launch which are now available to pre-order on the official GeForce webpage. Or you can head over to this article and check out all the glorious non-reference models which you will be able to get very soon.
Check out the other cards in the links below:
- GeForce RTX 2080 Ti (999 US) Graphics Card
- GeForce RTX 2080 ($699 US) Graphics Card
- GeForce RTX 2070 ($499 US) Graphics Card