If you’ve been waiting for a detailed test of the NVIDIA GeForce RTX 3080 Ti graphics card, which was officially debuted just a moment ago, you’ve come to the right place. The latest version of the Ampere architecture is a specific piece of silicon slightly thinner than the NVIDIA GeForce RTX 3090, the fastest graphics chip available on store shelves. However, the gaming giant made its debut in tragic market conditions, when most of the Ampere orders did not reach buyers, while in the meantime the prices rose several times … Will the NVIDIA GeForce RTX 3080 Ti improve the situation? Little? Well… if it’s available at the suggested price, enthusiasts will at least benefit, but we’re still talking about high-end for a lot of money.
Author: Sebastian Oktaba
GeForce RTX 3000 It is the second generation of NVIDIA graphics cards with hardware modules for ray tracing and deep learning acceleration, using the new Ampere architecture, an upgraded version of Turing that supports the GeForce RTX 2000 family in the. Increased cache capacity, doubling the number of shaders in SM (Multiprocessor Stream) blocks while increasing the performance of RT and Tensor cores. Another significant change was the move to lower lithography from Samsung, which made it possible to improve the power efficiency ratio, reduce core size and raise the clock. The best models received GDDR6X memory with high-frequency and wide buses, the bandwidth of which is equal to HBM2 modules. NVIDIA Ampere also offers a series of interesting solutions and technologies, including a Reflex connector, RTX IO, broadcasting, and HDM 2.1 that allow 8K HDR image rendering at a refresh rate of 60Hz. More rigorous technical information can be found on the next page, dedicated to the most important changes made in the new architecture.
The GeForce RTX 3080 Ti has a core similar to the GeForce RTX 3090, but its graphics memory has been cut out and a smaller cooling system is used. Efficiency in cryptocurrency mining is also lower.
The NVIDIA GeForce RTX 3080 Ti is built around the GA102 Ampere chip, which also powers the models. GeForce RTX 3080 I GeForce RTX 3090, although the described variation is much closer to the highest representative of the Ampere family. relatively GeForce RTX 3090 Only two SM groups (Streaming Multiprocessors) have been removed, resulting in relatively small differences in the execution units. Looking at the technical specifications, we can conclude that the GeForce RTX 3080 Ti should be inferior to its older siblings by a maximum of 2-3%, of course ignoring the problems of the actual core and memory clock. Core dimensions, number of transistors, 8 nm ultrathin lithography, interface used, and TDP of 350 W remained unchanged. By default, NVIDIA could also release a GeForce RTX 3090 Ti, containing all 84 SM blocks, although it would be an expensive solution with a token performance boost (currently reserved for the Quadro A6000). Now you’re likely to ask – why the Greens decided to introduce the GeForce RTX 3080 Ti, which undermines the top position GeForce RTX 3090? The GA102 Slim Core is not everything…
|RTX 3090||RTX 3080 Ti||RTX 3080||RTX 3070|
|lithography||8 nm||8 nm||8 nm||8 nm|
|pulp size||628 mm²||628 mm²||628 mm²||392 mm²|
|transistors||28 billion||28 billion||28 billion||17 billion|
|SP . units||10496||10240||8704||5888|
|Royal Oman Police Units||112||112||96||96|
|RT . units||82||80||68||46|
|base clock||1395 MHz||1365 MHz||1440MHz||1500 MHz|
|increase the clock||1650 MHz||1665 MHz||1710 MHz||1730 MHz|
|memory clock||19500 MHz||19000 MHz||19000 MHz||14000 MHz|
|Memory amount||24 GB||12 GB||10 GB||8 GB|
|Memory bus||384 bit||384 bit||320 bit||256 bit|
|eligibility||937 GB/sec||912 GB/sec||760 GB/sec||448 GB/sec|
|user interface||PCI-E 4.0 x16 . slot||PCI-E 4.0 x16 . slot||PCI-E 4.0 x16 . slot||PCI-E 4.0 x16 . slot|
|TGP coefficient||350 watts||350 watts||320 watts||220 watts|
|Initial price||1499 USD99||1199 USD||699 USD||499 USD|
|date of publication||September 2020||June 2021||September 2020||October 2020|
AMD Ryzen 9 5900X Performance Test vs. Intel Core i9-11900K vs. Intel Core i9-10900K with Fast DDR4 Memory I GeForce RTX 3090
The NVIDIA GeForce RTX 3080 Ti has 12GB of GDDR6X memory on the 384-bit bus, while NVIDIA GeForce RTX 3090 It’s got double the VRAM, with a slightly higher clock. Notable progress against GeForce RTX 3080 Yes, it did, although the competition still offers more memory, even on much cheaper models. The core frequency is set to 1365MHz, the advertised GPU Boost is at a minimum of 1665MHz, while the actual hardware is operating in the 1785-1815MHz range. Here, the differences between the top amps are minor again. However, the NVIDIA GeForce RTX 3080 Ti Founders Edition received a cooling system of the model GeForce RTX 3080 Enterprise EditionAlthough the TDP of 350W shares with GeForce RTX 3090Hence, temperatures and noise levels are likely to rise. However, important information limits the efficiency in mining cryptocurrency (Ethereum) through the factory-enabled LHR function (lite hash rate), and cut the hash in half. Time will tell if miners will find a way around this security, but the starting price will be the average investment. Exactly… The NVIDIA GeForce RTX 3080 Ti Founders Edition is priced at $1,199 which translates to PLN 5600 (MSRP), so it’s not really affordable.
“Player. Introvert. Problem solver. Creator. Thinker. Lifetime food evangelist. Alcohol advocate.”