That should be the moment for AMD-when Navi finally arrived, giving the Red Team the ability to compete in the high-end of the graphics hardware market again. However, a few days before the launch of the Radeon RX 5700 and RX 5700 XT, Nvidia set off a wave: Three "tweener" cards were designed to provide superior performance at a similarly dangerous price, freeing AMD from the shackles of AMD. In fact, the company was so confident in its new product line that it even lifted the embargo forward -As a result, many believe that products that respond to AMD Navi will actually be reviewed before their potential competition.
When Nvidia teased in May, the Super series included three cards: RTX 2060 Super, RTX 2070 Super, and RTX 2080 Super. The 2080 Super is still waiting, but we have tested the first two members of the family and you can see from the data that they are impressive.
Compared with the original version, the Super series cards have improved in almost all aspects. The new GPU provides more CUDA cores, a higher base clock, and even improves hardware-accelerated ray tracing capabilities. However, these higher clocked components do consume more power. The 2060 Super is much more, and the 2070 Super is much more. Compared to the previous 6GB, the RTX 2060 Super also has 8GB of VRAM-considering that ray tracing in particular may increase memory capacity, it is a good addition. All in all, this is an impressive package, which may indicate that Nvidia has established a certain headroom with its first-generation Turing card.
The final price has been determined: the price of RTX 2060 Super is 379 / $ 399 / € 419, the price of vanilla card has increased, the price of RTX 2070 Super is 479 / $ 499 / € 529, which is currently the price requested by Nvidia. The original 2070 Founders Edition card. When we consider AMD to price its new Navi graphics card-$ 379 for Radeon 5700 and $ 449 for Radeon 5700 XT-it's clear that both companies have good products, but AMD couldn't put the original RTX 2060 and RTX in front Provides 2070 cards as planned. Based on AMD's own benchmark tests for older RTX cards, we should expect the RTX 2060 Super and its 2070S to rival or defeat Navi with rasterization performance alone, and then consider ray tracing, variable rate shadowing, and DLSS. As always, please wait for reviews and benchmarks before making any positive conclusions about the upcoming Navi vs Turing dust settled.
Going back to the current question, it is strange that the RTX 2060 Super uses the TU106 processor on the original RTX 2060 and RTX 2070. The RTX2070 Super actually uses the TU104 core that forms the basis of the RTX 2080, explaining its higher performance. The 6-pin and 8-pin power inputs on the TDP- and Founders Edition cards have been sent for review. At the same time, the RTX 2060 Super Founders Edition requires only an eight-pin input and is located on the back of the card. In terms of I / O, the 2070 Super FE includes three DisplayPorts, USB-C and HDMI, while the 2060 Super FE swaps one of them for DVI.
It's worth looking at the spec sheet below to see what each card offers at a hardware level compared to its predecessor. For example, the streamlined 192-bit memory bus in the original RTX 2060 has been restored to the full 256-bit strength of the 2060 Super, unlocking the corresponding increase in bandwidth to 448 GB / s, which is equivalent to the RTX 2070 and RTX 2070 Super.
At the same time, the RTX 2070 Super has made the biggest progress elsewhere, upgrading from 2304 CUDA cores to 2560 CUDA cores, and the boost clock has steadily increased from 1620MHz to 1770MHz-we often see that the card reaches 1935MHz in tests. In terms of ray tracing performance, the RTX 2060 Super has the same 6GR / s rating as the original RTX 2070, while the 7GR / s of the 2070 Super makes it just one step behind the RTX 2080.
| RTX 2060 | 2060 Super | RTX 2070 | 2070 Super | RTX 2080 | |
|---|---|---|---|---|---|
| CUDA Core | 1920 | 2176 | 2304 | 2560 | 2944 |
| Giga ray / second | 5 | 6 | 6 | 7 | 8 |
| Virtual RAM | 6GB GDDR6 | 8GB GDDR6 | 8GB GDDR6 | 8GB GDDR6 | 8GB GDDR6 |
| Memory bus | 192 bits | 256-bit | 256-bit | 256-bit | 256-bit |
| bandwidth | 336GB / s | 448GB / s | 448GB / s | 448GB / s | 448GB / s |
| Boost clock | 1680 MHz | 1650 MHz | 1620 MHz | 1770MHz | 1710 MHz |
| processor | TU106 | TU106 | TU106 | TU104 | TU104 |
| Technology Development Plan | 160 watts | 175 watts | 175 watts | 215 watts | 215 watts |
Now let's take a closer look at the performance these two cards can provide in a series of the latest games. In addition to the original RTX 2060 and RTX 2070, we will also compare the new Super card with the best Pascal card. Please note that AMD's upcoming Navi products RX 5700 and RX 5700 XT are not yet available, so they are not currently in our testing. However, once the 5700 Series cards reach consumers, we will correct this. stay tuned. Important note: We fully tested the Founders Edition RTX card here, while the original RTX version had overclocking capabilities in a small factory, while the Super FE card did not. This gives the older graphics cards a slight boost, but doesn't seem to have much impact on the overall outlook.
Nvidia GeForce RTX 2060 Super / RTX 2070 Super analysis
