RTX 30 vs RTX 20: Equivalency Guide

The presentation of the RTX 30 was without a doubt one of the most important technological events of the year. NVIDIA has set the bar highand as we could see in our analysis of the RTX 3080, Jen-Hsun Huang ran the company managed to meet expectations that was created.

It is true that the launch of the RTX 3080 It was a bit bumpyespecially due to the storage problems caused by speculation and the bulk purchase of the graphics card. We already had the chance to talk about it in this article, and we saw that NVIDIA had not only promised to stop buying and reselling its new high-end graphics card, but also assured them that it would Fill up.

The RTX 3090 was recently launched, a top of the line graphics card that has become NVIDIA’s new flagship, and as many of our readers know, it is the first general consumer graphics solution capable of moving games 8K resolution. Yes it can, but as you might have imagined it has to resort to it DLSS 2.0 maintain a good level of fluency.

We are in a situation where 4K resolution games are not yet fully standardized. So it’s too early to talk about 8K. There is no doubt about that, but we must not distract us the potential of the RTX 3090and we must not forget the value that DLSS 2.0 technology offers that can be achieved superior image qualityeven to which we would work with native resolution.

RTX 30 equivalency guide: how do you position yourself versus the RTX 20? 31

I know some of our readers Consider upgrading your graphics card Take advantage of the launch of the new NVIDIA RTX 30 while others wait to see what AMD presents on October 28th. On this day we will know the keys to the RDNA 2 architecture and theoretically discover the equipped Radeon RX 6000 with dedicated hardware to accelerate the ray tracing function.

Personally, I would wait and see what AMD presents before I start updating my graphics card, even if you are sure that you want a graphics solution from NVIDIA. The reason is very simple, the arrival of the Radeon RX 6000 implies competitionThis is always good for the user as it can affect the price of the RTX 30 and allow us to find models at a lower cost.

However, I understand that many of you have doubts and that you don’t quite fit in with how the new NVIDIA graphics cards really position themselves versus the RTX 20 series, and that’s why I wanted to share with you this article where we are going Rate the raw performance of the RTX 30 versus the RTX 20and therefore where we will find out how to position the new graphics cards based on amps.

GeForce RTX 30 vs GeForce RTX 20: greater distance at high resolutions

RTX 30 Equivalency Guide: How Are You Positioning Against the RTX 20? 33

Currently, NVIDIA has only confirmed three GeForce RTX 30 series graphics cards, the 3090, which positions it as the front runner in the product range. the 3080, which is in the high range; and the 3070, which fits into the high-end “budget” range. However, thanks to numerous leaks, we were able to determine that the green giant is on other interesting models like the RTX 3070 Ti and RTX 3060-3060 Tiand we cannot rule out the arrival of an economic model that “democratizes” ray tracing, such as RTX 3050.

The attached picture, received at an event of GALAX assembler and published by VideoCardz, That way, we can find out how they position the 30-series RTXs that will hit the market in the next few months compared to the RTX 20-series. Yes, everything not only fits the information we have seen, but also fits together perfectly the official results that appeared in the last analyzes.

At the top there are no surprises, the RTX 3090 is currently the most powerful graphics card on the market and it has no direct equivalence. In second place is the RTX 3080, which also has no direct equivalent, as it far surpasses both the RTX 2080 and the RTX 2080 Ti. The third position is for an alleged RTX 3070 Ti (it could also come with the Super Tag), the it would be a little more powerful than the RTX 2080 Ti, but it wouldn’t clearly outperform the TITAN RTX.

As time goes on, things get more and more interesting. The GALAX picture suggests this The RTX 3070 would not be more powerful than the RTX 2080 Ti. it would rank a little below, although it would easily outperform the RTX 2080 Super. The RTX 3060 sits between the RTX 2080 and the RTX 2080 Super, which means the RTX 3060 Ti should be roughly on par with an RTX 2080 Super.

I don’t have performance data for the RTX 3070 and below so I can’t make an illustrative table of specific data, but I can share it with you a list from highest to lowest performance This will serve as a reference to make it clear how they position the new RTX 30 series compared to the RTX 20 series.

1.-GeForce RTX 3090

RTX 30 Equivalency Guide: How Are You Positioning Against the RTX 20? 35

  • GA102 graphics core at 8 nm
  • 10,496 shaders at 1,395 MHz – 1,695 MHz.
  • 328 texturing units.
  • 112 grid units.
  • 328 tensor cores.
  • 82 RT cores.
  • 35.8 TFLOPs of the service in FP32.
  • 24 GB GDDR6X at 19.5 GHz.
  • 384-bit bus.
  • 350 watt TDP.
  • Requires a 12-pin connector and a 750 watt power supply.

2.-GeForce RTX 3080

RTX 30 Equivalency Guide: How Are You Positioning Against the RTX 20? 37

  • GA102 graphics core at 8 nm
  • 8,704 shaders at 1,440 MHz – 1,710 MHz.
  • 272 texturing units.
  • 96 grid units.
  • 272 tensor cores.
  • 68 RT cores.
  • 29.7 TFLOPs of the service in FP32.
  • 10 GB GDDR6X at 19 GHz.
  • 320-bit bus.
  • 320 watt TDP.
  • Requires a 12-pin connector and a 700 watt power supply.

3.-TITAN RTX

  • TU102 GPU made in the 12nm process.
  • 4,608 shaders at 1.350 GHz-1.77 GHz normal and turbo mode.
  • 288 texturing units.
  • 96 grid units.
  • 576 tensor cores.
  • 72 RT cores.
  • 16.31 TFLOPs of the performance in FP32.
  • 384-bit bus.
  • 24 GB of effective 14 GHz GDDR6 memory.
  • 280 watt TDP.
  • Two 8-pin power connections and a 650 watt power supply.

4.-GeForce RTX 3070 Ti

  • Graphics core GA104 at 8 nm.
  • 6,144 shaders at 1,410 MHz – 1,695 MHz.
  • 192 texturing units.
  • 96 grid units.
  • 192 tensor cores.
  • 48 RT cores.
  • 20.83 TFLOPs of performance in FP32.
  • 8 GB GDDR6 at 16 GHz.
  • 256-bit bus.
  • 250 watt TDP.
  • Requires a 12-pin connector and a 650 watt power supply.

5.-GeForce RTX 2080 Ti

  • TU102 graphics core in the 12nm process.
  • 4,352 shaders at 1,350 MHz – 1,545 MHz.
  • 272 texturing units.
  • 88 grid units.
  • 544 tensor cores.
  • 68 RT cores.
  • 13.45 TFLOPs of the performance in FP32.
  • 352-bit bus.
  • 11 GB of effective 14 GHz GDDR6 memory.
  • 250 watt TDP.
  • You need two 8-pin power connectors and a 650-watt power supply.

6.-GeForce RTX 3070

RTX 30 Equivalency Guide: How Are You Positioning Against the RTX 20? 39

  • Graphics core GA104 at 8 nm.
  • 5,888 shaders at 1,500 MHz – 1,725 ​​MHz.
  • 184 texturing units.
  • 64 grid units.
  • 184 tensor cores.
  • 46 cores RT.
  • 19.9 TFLOPs of the performance in FP32.
  • 8 GB GDDR6 at 14 GHz.
  • 256-bit bus.
  • 220 watt TDP.
  • Requires a 12-pin connector and a 600 watt power supply.

7.-GeForce RTX 2080 Super

RTX 30 Equivalency Guide: How Are You Positioning Against the RTX 20? 41

  • TU104 graphics core at 12 nm.
  • 3,072 shaders at 1,650 MHz – 1,815 MHz.
  • 192 texturing units.
  • 64 grid units.
  • 384 tensor cores.
  • 48 RT cores.
  • 11.15 TFLOPs of the performance in FP32.
  • 256-bit bus.
  • 8 GB GDDR6 at 15.5 GHz.
  • 250 watt TDP.
  • It requires an 8-pin and a 6-pin connector and a 600 watt power supply.

8.-GeForce RTX 3060 Ti

  • Graphics core GA104 at 8 nm.
  • 4,864 shaders at 1,410 MHz – 1,665 MHz.
  • 152 texturing units.
  • 80 grid units.
  • 152 clamping cores.
  • 38 RT cores.
  • 16.2 TFLOPs of the service in FP32.
  • 8 GB GDDR6 at 14 GHz.
  • 256-bit bus.
  • 200 watt TDP.
  • Requires a 12-pin connector and a 550 watt power supply.

10.-GeForce RTX 2080

RTX 30 Equivalency Guide: How Are You Positioning Against the RTX 20? 43

  • TU104 GPU made in the 12nm process.
  • 2,944 shaders at 1,515 MHz – 1,847 MHz, normal and turbo mode.
  • 184 texturing units.
  • 64 grid units.
  • 368 tensor cores.
  • 46 cores RT.
  • 10 TFLOPs of performance in FP32.
  • 256-bit bus.
  • 8 GB of effective 14 GHz GDDR6 memory.
  • 215 watt TDP.
  • It requires an 8-pin and a 6-pin connector and a 600-watt power supply.

10.-GeForce RTX 3060

  • Graphics core GA104 at 8 nm.
  • 3,840 shaders at 1,500 MHz – 1,725 ​​MHz.
  • 120 texturing units.
  • 64 grid units.
  • 120 tensor cores.
  • 30 RT cores.
  • 13.2 TFLOPs of the service in FP32.
  • 6 GB GDDR6 at 14 GHz.
  • 192-bit bus.
  • 180 watt TDP.
  • Requires a 12-pin connector and a 500 watt supply.
Click to rate this entry!
(Votes: 1 Average: 4)
Share!

Leave a Comment