After an extended quiet period in the realm of high-end GPUs, things are going to get exciting again. Last month AMD announced their new mid-to-high selection of video cards, the Radeon RX 5700 series, predicated on their new Navi GPU architecture and associated 7nm Navi 10 GPU. AMD has recently revealed essentially everything about these cards beforehand, and that launch is fittingly approaching on Sunday, July 7th (7/7).
But first, NVIDIA wish to have a word.
Since Computex, NVIDIA has been teasing a “Super” announcement of their own. Which morning, that announcement is finally arriving at fruition. NVIDIA is launching a mid-generation kicker because of their mid-to-high-end video card lineup in the sort of their GeForce RTX 20 series Super cards. Predicated on the same category of Turing GPUs as the initial GeForce RTX 20 series cards, these new Super cards – all suffixed Super, appropriately enough – include new configurations and new clockspeeds. They are, essentially, NVIDIA’s 2019 card family for the $399+ video card market.
If they are released on July 9th, the GeForce RTX 20 series Super cards will be sharing store shelves with all of those other GeForce RTX 20 series cards. Some cards just like the RTX 2080 and RTX 2070 are set to disappear completely, while other cards just like the RTX 2080 Ti and RTX 2060 will stay out there as-is. Used, it’s probably better to think of the brand new cards as NVIDIA executing as the lessen price or a spec bump – according to if you start to see the glass as half-empty or half-full – all without meaningfully changing their price tiers.
In conditions of performance, the RTX 2060 and RTX 2070 Super cards aren’t likely to bring anything not used to the table. Actually if we’re being blunt, the RTX 2070 Super is actually a slightly slower RTX 2080, and the RTX 2060 Super may aswell be the RTX 2070. So instead, what has changed may be the price these performance levels can be found at, and in the end the performance-per-dollar ratios in elements of NVIDIA’s lineup. The performance of NVIDIA’s former $699 and $499 cards will now be accessible for $499 and $399, respectively. This leaves the vanilla RTX 2060 to carry the line at $349, and the upcoming RTX 2080 Super to fill the $699 spot. This means if you’re in the $400-$700 market for video cards, your alternatives are going to get noticeably faster.
The timing on all this is pretty interesting, of course. NVIDIA is launching these cards just with time to complement AMD’s own Radeon RX 5700 launch. Even though NVIDIA’s official line is that launch is driven by economies of scale and slow card sales because of poor pricing – both valid points that I’ll grant NVIDIA – to call a spade a spade that is obviously NVIDIA seeking to counter AMD’s launch. AMD showed their cards early (literally and figuratively), so NVIDIA and all of those other world could actually see what sort of performance AMD is looking to get and at what prices those cards will roll out.
Ultimately by launching faster cards at $400 and $500 now, NVIDIA is seeking to protect their massive market share while moving the goalposts on AMD (ed: or possibly just removing the football entirely). After today, another move is AMD’s, but as we’ll see inside our benchmarks of the brand new RTX 20 Super cards, they’ll have to react if they would like to maintain their momentum. For the time being, I believe it’s fair to state that everyone is merely pleased to see pricing drop on NVIDIA’s capable but overpriced Turing video cards.
Launching July 9th: GeForce RTX 2070 Super & GeForce RTX 2060 Super
Engaging in the meat of today’s review, let’s have a glance at the specs for NVIDIA’s new cards, the GeForce RTX 2070 Super and GeForce RTX 2060 Super. As I discussed earlier, these are more often than not NVIDIA’s 2080 and 2070 vanilla cards slightly tweaked and given lower model numbers. This enables NVIDIA to wind up their performance without actually cutting prices at their major price points. Neither card is accurately identical to the 2018 card its borrowing so heavily from – there are small specification distinctions for binning and product differentiation reasons – but even regarding the RTX 2070, we’re discussing simply a 4% performance deficit versus the RTX 2080.
We’ll start with the RTX 2070 Super. This card makes the largest leap among the brand new RTX 20 series Super card lineup, as its predecessor, the RTX 2070 vanilla, had been by using a fully-enabled TU106 GPU. Due to this fact, in order to wind up the performance, NVIDIA has made the jump to the TU104 GPU, which previously was only used for the RTX 2080. Juxtaposed against the RTX 2080 (and forthcoming RTX 2080 Super), this card is a far more “traditional” 70 card, for the reason that it’s now predicated on a cut-down version of the same GPU in the 80 card.
All told, NVIDIA has disabled 8 of TU104’s 48 SMs here, leaving a card with 40 SMs, or 2560 Turing CUDA cores. To create up because of this, NVIDIA has cranked up clockspeeds a bit; RTX 2070 Super has higher base and boost clocks compared to the RTX 2080, which helps it to narrow what gap there is to start out with. In the event that you look at pure SM throughput, the RTX 2070 should deliver around 90% of RTX 2080’s compute/texturing/geometry performance; but because its 64 ROPs are fully-enabled and clocked greater than RTX 2080, used it trails the RTX 2080 by simply 4% at 2560×1440.
Meanwhile on the memory side of matters, things remain unchanged. Both TU104 and TU106 sport 256-bit memory busses, and every one of the 2070 cards are fully populating this with 8GB of GDDR6 clocked at 14Gbps. This ensures that the RTX 2070 Super doesn’t have any longer memory bandwidth than its vanilla counterpart, but in addition, it means it has as much memory bandwidth as the RTX 2080.
The drawback here’s that with RTX 2080-like performance comes RTX 2080-like power consumption. The TDP of the card is 215 Watts, specifically exactly like the RTX 2080 and 40W greater than the RTX 2070 vanilla. In some recoverable format, NVIDIA’s energy efficiency actually takes a tiny hit here as a result of this – it’s significantly less than RTX 2080 performance for the same TDP – which is partly a rsulting consequence using higher clockspeeds. But with the cards’ performance being so close, used it’s a negligible shift.
The mark market for the brand new RTX 2070 Super then is basically exactly like the RTX 2080 it borrows so liberally from. The RTX 2070 isn’t quite enough to operate a vehicle 4K resolution gaming – that’s employment for the RTX 2080 Ti – nonetheless it is incredibly capable for 2560x1440p. Alternatively, nowadays the 2080/2070S does pretty much in ray traced games at 1080p, though it still varies some from game to game.
Meanwhile, to arrive below the RTX 2070 Super at the brand new price tier of $399 may be the RTX 2060 Super. That is a card that’s for all intents and purposes a RTX 2070 at less price; it uses an ever-so-slightly cut-down version of TU106 to provide performance that’s within 1% of the RTX 2070. All told, NVIDIA did something fairly similar here much like the RTX 2070 Super, disabling some SMs and cranking up the clockspeeds to pay. The outcome is a card with 34 of 36 SMs enabled, plus some extra clocks to greatly help minimize the gap. By the numbers it gives 96% of RTX 2070’s shading/texturing/geometry performance, but 102% of its ROP/rasterization performance. Or even to frame things in accordance with the RTX 2060, it’s about 14% faster because of its combo of SM and clockspeed advantage.
Equally (only if no more) important here however may be the memory configuration. Whereas the RTX 2060 vanilla posseses an awkward 6GB of GDDR6 – leaving it just a lttle bit short on capacity and bandwidth – the RTX 2060 Super gets a complete 8GB. This ensures that it can deliver as much VRAM capacity and bandwidth as the 2070 and 2080 cards – 33% more of every compared to the vanilla 2060 – which not merely helps it a whole lot in performance, but also in future-proofing. NVIDIA because of its part isn’t retiring the vanilla RTX 2060, but also for just $50 over that card, there’s an extremely compelling argument to disregard the hobbled 6GB card entirely.