News Posts matching "TU116"

Return to Keyword Browsing

NVIDIA TU116 GPU Pictured Up Close: Noticeably Smaller than TU106

Here is the first picture of NVIDIA's 12 nm "TU116" silicon, which powers the upcoming GeForce GTX 1660 Ti graphics card. While the size of the package itself is identical to that of the "TU106" on which the RTX 2060 and RTX 2070 are based; the die of the TU116 is visibly smaller. This is because the chip physically lacks RT cores, and only has two-thirds the number of CUDA cores as the TU106, with 1,536 against the latter's 2,304. The die area, too, is about 2/3rds that of the TU106. The ASIC version of TU116 powering the GTX 1660 Ti is "TU116-400-A1."

VideoCardz scored not just pictures of the ASIC, but also the PCB of an MSI GTX 1660 Ti Ventus graphics card, which reveals something very interesting. The PCB has traces for eight memory chips, across a 256-bit wide memory bus, although only six of them are populated with memory chips, making up 6 GB over a 192-bit bus. The GPU's package substrate, too, is of the same size. It's likely that NVIDIA is using a common substrate, with an identical pin-map between the TU106 and TU116, so AIC partners could reduce PCB development costs.

Palit and EVGA GeForce GTX 1660 Ti Cards Pictured

As we inch closer to the supposed 15th February launch of the GeForce GTX 1660 Ti, pictures of more AIC partner branded custom-design cards. The first two of these are from Palit and EVGA. Palit is bringing two very compact cards to the table under its StormX banner. These cards appear to be under 18 cm in length, and use an aluminium fin-stack cooler that's ventilated by a single 100 mm fan. There are two grades based on factory-overclock. The base model ticks at 1770 MHz boost, while the OC variant offers 1815 MHz boost.

EVGA's GTX 1660 Ti lineup includes two cards under its XC brand, with both cards being under 20 cm in length, but are 3 slots thick. Both cards appear to be using the same 3-slot single-fan cooling solution as the company's RTX 2060 XC. Once again, we see two variants based on clock-speeds, with the "Black" variant sticking to 1770 MHz boost, and the XC version slightly dialing up that frequency. Based on the 12 nm "TU116" silicon, the GTX 1660 Ti is rumored to feature 1,536 CUDA cores based on the "Turing" architecture, but lacking in RTX technology. The SKU succeeds the GTX 1060 6 GB.

More GeForce GTX 1660 Ti Specs Emerge

A Russian retailer has leaked more specifications of NVIDIA's upcoming GeForce GTX 1660 Ti graphics card. Based on the 12 nm "TU116" silicon, this card will be configured with 1,536 "Turing" CUDA cores, but have no RT cores, and hence no RTX features. The chip could end up with 96 TMUs and 48 ROPs. The GPU is clocked at 1500 MHz nominal, and the boost frequency is set at 1770 MHz, however, the latter could be a factory-overclock set by AIC partner Palit for their GTX 1660 Ti StormX graphics card.

The memory subsystem of the GTX 1660 Ti is interesting. While it's still 6 GB of GDDR6 memory across a 192-bit wide memory bus, the memory clock itself is lower than that of the RTX 2060. The memory ticks at 12 Gbps, which results in 288 GB/s of memory bandwidth, compared to the RTX 2060, which thanks to its 14 Gbps memory achieves 336 GB/s. The card draws power from a single 8-pin PCIe power connectors. Outputs include HDMI, DisplayPort, and DVI, we don't expect any cards to ship with VirtualLink.

MSI GeForce GTX 1660 Ti SKUs Listed on the Eurasian Economic Commission, Adds Fuel to 1660 Ti Fire

It seems only yesterday that we were discussing a Turing microarchitecture-based TU116 die that would power the yet-to-be-confirmed GeForce GTX 1660 Ti. With no RTX technology support, this was speculated to be NVIDIA's attempt to appease the mainstream gaming market that understands the GPU does not have enough horsepower to satisfactorily drive real-time ray tracing in games while still maintaining an optimal balance of visual fidelity and performance alike. Reports indicated an announcement next month, followed by retail availability in March, and today we got word of more concrete evidence pointing towards all these coming to fruition.

It appears that trade listings in various organizations are going to be a big source of leaks in the present and future, with MSI GeForce GTX 1660 Ti SKUs, including the Gaming Z, Armor, Ventus, and Gaming X, all listed on the Eurasian Economic Commission (EEC). The listing covers the associated trademarks, all awarded to MSI, and is one of the last steps towards setting up a retail channel for new and upcoming products. Does the notion of a Turing GTX GPU without real-time ray tracing interest you? Let us know in the comments section below.

NVIDIA GeForce GTX 1660 Ti Put Through AoTS, About 16% Faster Than GTX 1060

Thai PC enthusiast TUM Apisak posted a screenshot of an alleged GeForce GTX 1660 Ti Ashes of the Singularity (AoTS) benchmark. The GTX 1660 Ti, if you'll recall, is an upcoming graphics card based on the TU116 silicon, which is a derivative of the "Turing" architecture but with a lack of real-time raytracing capabilities. Tested on a machine powered by an Intel Core i9-9900K processor, the AoTS benchmark was set to run at 1080p and DirectX 11. At this resolution, the GTX 1660 Ti returned a score of 7,400 points, which roughly compares with the previous-generation GTX 1070, and is about 16-17 percent faster than the GTX 1060 6 GB. NVIDIA is expected to launch the GTX 1660 Ti some time in Spring-Summer, 2019, as a sub-$300 successor to the GTX 1060 series.

NVIDIA Readies GeForce GTX 1660 Ti Based on TU116, Sans RTX

It looks like RTX technology won't make it to sub-$250 market segments as the GPUs aren't fast enough to handle real-time raytracing, and it makes little economic sense for NVIDIA to add billions of additional transistors for RT cores. The company is hence carving out a sub-class of "Turing" GPUs under the TU11x ASIC series, which will power new GeForce GTX family SKUs, such as the GeForce GTX 1660 Ti, and other GTX 1000-series SKUs. These chips offer "Turing Shaders," which are basically CUDA cores that have the IPC and clock-speeds rivaling existing "Turing" GPUs, but no RTX capabilities. To sweeten the deal, NVIDIA will equip these cards with GDDR6 memory. These GPUs could still have tensor cores which are needed to accelerate DLSS, a feature highly relevant to this market segment.

The GeForce GTX 1660 Ti will no doubt be slower than the RTX 2060, and be based on a new ASIC codenamed TU116. According to a VideoCardz report, this 12 nm chip packs 1,536 CUDA cores based on the "Turing" architecture, and the same exact memory setup as the RTX 2060, with 6 GB of GDDR6 memory across a 192-bit wide memory interface. The lack of RT cores and a lower CUDA core count could make the TU116 a significantly smaller chip than the TU106, and something NVIDIA can afford to sell at sub-$300 price-points such as $250. The GTX 1060 6 GB is holding the fort for NVIDIA in this segment, besides other GTX 10-series SKUs such as the GTX 1070 occasionally dropping below the $300 mark at retailers' mercy. AMD recently improved its sub-$300 portfolio with the introduction of Radeon RX 590, which convincingly outperforms the GTX 1060 6 GB.
Return to Keyword Browsing