The rumors are correct: The GeForce RTX 3060 Ti is a real graphics card, and we tested it within the last week. Nvidia promises the RTX 2080 Super performance in a smaller, cheaper and more efficient Ampere design, and – spoiler alert – that’s exactly what our testing showed.

For £ 370 in the UK, $ 400 in the US and around € 399 in Europe the 3060 Ti is not. so far In the mainstream region, but much faster than the RTX 2060 Super, it replaces at the same price point, AMD’s competitive Big Navi architecture hasn’t yet appeared in anything less than the $ 580 / £ 530 price point of the RX 6800. If the green team is actually capable of Producing these cards in volume – and selling them to actual customers rather than robot farms – can be a winner.

In terms of specifications and infrastructure, the RTX 3060 Ti uses the same GA104 GPU as the RTX 3070, but with fewer CUDA cores – 4864 vs 5888. The card also operates at slightly slower hour speeds (boosting 1665MHz vs 1725MHz) to fit in the TDP drive Less at 20W (200W vs 220W). However, the memory subsystems have not changed, as both cards have the same 8GB GDDR6 running at 448GB / s. It’s good to see that 8GB VRAM has become the new standard, with all next-generation cards from both teams providing at least that much so far.

This content is hosted on an external platform, which will only display it if you accept targeted cookies. Please enable cookies to view.

RTX 3080 RTX 3070 RTX 3060T RTX 2060 Super RTX 2060 GTX 1060 card
GPU GA102 GA104 GA104 TU106 TU106 GP106
Coda colors 8704 5888 4864 2176 1920 1280
VRAM 10 GB GDDR6X 8 GB GDDR6 8 GB GDDR6 8 GB GDDR6 6 GB GDDR6 6 GB GDDR5
Memory bus 320 bits 256 bits 256 bits 256 bits 192 bits 192 bits
Band width 760 GB / s 448 GB / s 448 GB / s 448 GB / s 336 GB / s 192 GB / s
The basic hour 1440MHz 1500MHz 1410MHz 1470MHz 1365MHz 1506MHz
Increase the hour 1710MHz 1725MHz 1665MHz 1650MHz 1680MHz 1709MHz
TDP 320 watts 220W 200 watts 175W 160 watts 120 watts
Die size 628 mm2 392mm2 392mm2 445 mm2 445 mm2 200 mm2
Transistors 28B 17.4 B. 17.4 B. 10.8 b 10.8 b 4.4 B.

The comparison with the older generation cards is more interesting. The 3060 Ti more than doubles the number of graphics cores in the 2060 Super, in a smaller die that consumes a bit more power. Looking back, in the GTX 1060, the multiplier is closer to 4x – with a corresponding increase in the number of the transistor, thanks to the shift from 16 nm with Pascal to 12 nm with Turing and now 8 nm with amps.

In addition to the improvements in computing performance, all the usual features of Nvidia’s Ampere architecture are present and correct, including next-generation ray tracing and tensor core, so we can expect more noticeable increases in performance in RT and AI accelerated workloads. We’ve covered Ampere in more detail in previous RTX 3070, 3080, and 3090 reviews so let’s head over to the physical design of the card.

As usual, we’re testing the Founders version of the card, which comes with the same great industrial design and the same compact dimensions as the RTX 3070. We have two coaxial fans in a “flow-through” configuration, with a scientific-shaped motherboard design and a miniature 12-pin power connector, allowing By using the last third of the card completely for cooling. Even the inputs and outputs are arranged to maximize airflow, with a single row of display outputs (three DisplayPort 1.4, one HDMI 2.1) sitting beneath a 16 x 5 grid of air vents.

As with other RTX 30-series cards, the 3060 Ti is a PCIe 4.0 hardware but it runs on PCIe 3.0 motherboards without any performance loss in the vast majority of gaming loads. Another invisible feature is AV1 decoding support, which should allow sites like Twitch and Netflix to drastically increase resolution, frame rate and bitrate at a specific bandwidth – or reduce bandwidth by up to 50 percent while keeping those metrics intact. . AV1 support isn’t necessary right now, and it’s available on both new AMD and Nvidia cards, but it might be a pretty cherry on top for anyone choosing to upgrade once it hits the mainstream.

One of the biggest questions we asked when testing the 3060 Ti – besides gaming performance, which we’ll get to soon – surrounds energy efficiency. The RTX 3070 was much more efficient than its more powerful brothers, so is the 3060 Ti more efficient?

To answer this question, we use Nvidia’s Power Capture Analyzer or PCAT. This is an interface board located between the PCIe slot and the graphics card, as well as between the additional 8-pin power inputs used by our 3060 Ti and our power supply. This way, we can measure the number of watts the card itself is pulling, rather than the entire system load that can naturally change over time, and plot this force pull precisely against frame rates to get an idea of ​​how much power is being used to create each frame.

Nvidia’s PCAT system uses the PCIe interface and power-supply cable interference panels to measure the actual juice a GPU is consuming. Note that we have been using the RTX 3080 PSU adapter here – the 3060 Ti only requires one 8-pin PCIe power input and uses a different dongle.
Joules per frame RX 6800 XT RX 6800 RTX 3080 FE RTX 3070 FE RTX 3060 Ti FE RTX 2070 FE RX 5700 XT
Death by strand 2.933 2.644 3.349 2.915 3.064 3.651 3.846
Percentage difference 110.9% 100% 126.6% 110.2% 115.9% 138.1% 145.5%
Joules per frame RX 6800 XT RX 6800 RTX 3080 FE RTX 3070 FE RTX 3060 Ti FE RTX 2070 FE RX 5700 XT
Gears 5 4.384 3.792 4.156 3.603 3.163 4.734 5.403
Percentage difference 138.6% 119.9% 131.4% 113.9% 100% 149.7% 170.82%

In Death Stranding, the RTX 3060 Ti is roughly equal to the RTX 3070, and it requires 3,064 joules per frame to display our test scene compared to 2,915 joules per frame in the 3070. Big Navi graphics cards from AMD perform better in Nvidia’s Ampere’s Death Stranding, so we watched Better efficiency rating for the RX 6800 and RX 6800 XT. So the 3060 Ti uses less power than the 3070, but the decrease in performance is greater than the decrease in energy use – at least in this test.

Gears 5 is even more optimistic, with the 3060 Ti setting a new record here of 3.163 joules per frame, with the 3070 requiring 14 percent more power per tire, while the RX 6800 requiring about 20 percent more. We usually expect cards from the same family to be ranked in the same order in every game, so it’s unusual to see the 3060 Ti advance against the 3070 in one test and track in another. However, re-tests showed our preliminary results, allowing the 3060 Ti to claim the title “Most Efficient Ampere GPU” in at least one game. It’s worth looking at the RTX 2070 FE as well, which requires 50 percent joules per frame – the ampere and its 8nm conversion have had really solid results in terms of energy efficiency.

After completing the brief power test, it was time to confirm the specifications of our test machine. You’ll see some new components here if you haven’t checked out our RTX 30 Series reviews yet, but for those of you familiar, there are no surprises here. we’ve got Core i9 10900KB It is locked at 5GHz for all cores and cooled by the Alpacool Eisbaer Aurora AiO at 240mm – keeping the overclocking system at around 75 ° C under full load.

10900K is supported by the Asus Maximus 12 Extreme Z490 motherboard and stick 8 GB from G Skill Trident Costume Royal 3600MHz CL16. Our games are run from roomy 2 TB Samsung 970 Evo Plus NVMe Drive Provided by Box. The excavator is fully powered by a Gamer Storm power supply gold rated 850W.

With the stage set, let’s start the show with some gaming benchmarks.

Nvidia GeForce RTX 3060 Ti analysis

  • Introduction, Hardware and Power Analysis [This Page]

  • Doom Eternal, Control, Borderlands 3, Shadow of the Tomb Raider – Game Benchmarks Part 1

  • Death Stranding, Far Cry 5, Hitman 2, Assassin’s Creed Odyssey – Game Criteria Part 2

  • Metro Exodus, Dirt Rally 2, Assassin’s Creed Unity – Game Criteria Part 3

  • Control, Metro Exodus, Battlefield 5 – RTX Game Benchmarks

  • Nvidia GeForce RTX 3060 Ti – Digital Foundry Verdict

See also  New ugly YouTube line art icons coming to Android