Turing Discussion Thread

[SteelCrysis]

https://www.techpowerup.com/254814/nvidi...tx-1650-ti

It seems NVIDIA's partners are gearing up for yet another launch, sometime after the GTX 1650 finally becomes available. ECC Listings have made it clear that partners are working on another TU117 variant, with improved performance, sitting between the GTX 1650 and the GTX 1660, which will should bring the fight to AMD's Radeon RX 580. Of course, with the GTX 1660 sitting pretty at a $219 price, this leaves anywhere between the GTX 1650's $149 and the GTX 1660's $229 for the GTX 1650 Ti to fill. With the GTX 1660 being an average of 13% faster than the RX 580, it makes sense for NVIDIA to look for another SKU to cover that large pricing gap between the 1650 and the 1660.

https://www.techpowerup.com/254827/nvidi...r5-usd-150

NVIDIA today rolled out the GeForce GTX 1650 graphics card at USD $149.99. Like its other GeForce GTX 16-series siblings, the GTX 1650 is derived from the "Turing" architecture, but without RTX real-time raytracing hardware, such as RT cores or tensor cores. The GTX 1650 is based on the 12 nm "TU117" silicon, which is the smallest implementation of "Turing." It is equipped with 896 CUDA cores, 56 TMUs, 32 ROPs, and a 128-bit wide GDDR5 memory interface, holding 4 GB of memory clocked at 8 Gbps (128 GB/s bandwidth). The GPU is clocked at 1485 MHz, and the GPU Boost at 1665 MHz.

[SteelCrysis]

https://www.tomshardware.com/news/gtx-16...39153.html

Yet the process was changed a bit for the GTX 1650. Reviewers already have the GPU in their possession, the review embargo has passed, and Nvidia officially started taking orders for graphics cards featuring the GPU earlier today. But you'll notice that neither Tom's Hardware nor other publications have reviewed the GTX 1650, and that's because Nvidia decided not to provide the appropriate driver ahead of the product's launch.

That decision has already prompted speculation from some reviewers and social media users. Many seem to have decided that Nvidia withheld the driver because it's worried that GTX 1650 reviews will be unfavorable, with the hope being that it could sell as many units as possible before reviews came out. (This is common with game reviews, for example, where publishers accept pre-orders but decline to offer reviewers access to the title.)
...
An alternative explanation could be found via a more diplomatic version of Hanlon's razor, which says "Never attribute to malice that which is adequately explained by stupidity." Nvidia isn't stupid, and we suspect the vast majority of its employees aren't, either. But things do fall through the cracks--it's entirely possible that the GTX 1650 driver was supposed to be sent to reviewers a while ago but wasn't because of technical difficulties.

Either way, a public release driver for the GTX 1650 should arrive this week. Our review for the GTX 1650 will be published as soon as possible.

[SteelCrysis]

https://www.extremetech.com/gaming/29011...ch-reviews

The RX 570 is certain to burn far more power than the GTX 1650 — we noted during the 1660 launch that the RX 590 burned almost literally twice as much power as the GTX 1660, and the RX 570 could well hit 2x the power draw of the GTX 1650. But launching a brand-new GPU that loses to your competitor’s two-year-old part isn’t a particularly good look either, and it’s likely why Nvidia refused to sample the GTX 1650 driver until today, thereby ensuring that even if hardware sites got their hands on a GPU, they wouldn’t be able to test it.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...39161.html

According to Nvidia's official specifications, the GeForce GTX 1650 employs the Volta NVENC instead of the Turing NVENC. Nvidia's decision makes no sense at all especially when the GeForce GTX 1660 and GTX 1660 Ti both feature the Turing NVENC. A possible explanation could be that Nvidia is trying to minimize costs to maximize the profits on the GeForce GTX 1650.

Nvidia states that the Volta NVENC's performance is right in the alley of the Pascal NVENC. However, the Turing NVENC is up to 15 percent more efficient than the Volta NVENC and also brings a few improvements to combat artifacting. It isn't a huge deal-breaker if you don't plan on using the NVENC feature on the GeForce GTX 1650. But then again, if you're paying for a next-generation graphics card, you would expect it to come with the latest features.

https://www.techpowerup.com/254861/nvidi...dia-engine

Turing's NVENC is known to have around 15 percent performance uplift over Volta's, which means the GTX 1650 will have worse game livestreaming performance than expected. The GTX 1650 has sufficient muscle for playing e-Sports titles such as PUBG at 1080p, and with an up-to-date accelerated encoder, would have pulled droves of more amateur streamers to the mainstream on Twitch and YouTube Gaming. Alas, the $220 GTX 1660 would be your ticket to that.

https://www.guru3d.com/articles_pages/zo...ew,29.html

Little is wrong with what ZOTAC brought to the table with his card, however, Nvidia's price level might once again be bothersome. Right now if you browse around a bit you can get an AMD Radeon RX 570 4GB for 139 EUR/USD and the 8GB version at 149 EUR/USD. Especially the 8GB on that RX 570 gives it a massive advantage. Design wise ZOTAC did well, I mean this is a lovely looking and very compact product. It does not run hot, it's fairly silent and power consumption is low. So once again it's pricing that is going to be a problem. In reality, I find this to be a 99 USD domain card, but yeah we'll have to wait and see how prices are going to develop in the months to come.

https://www.techpowerup.com/reviews/MSI/..._X/34.html

MSI is pricing their GTX 1650 Gaming X at "$150-$160", which is only a small increase and very reasonable. In return, you get a much better cooler that is really quiet, idle fan-stop, a large overclock out of the box, and an increased power limit. However, at that price point, the GTX 1650 is simply way too expensive. AMD's Radeon RX 570 can be found online for $130 and offers significantly higher performance. It also comes bundled with two AAA games, which further offsets the card's cost. MSI isn't to blame for this as the problem is simply NVIDIA setting their MSRP much too high. A more realistic price point for the GTX 1650 would be $120, at which point it would achieve price/performance parity with the Radeon RX 570. Compared to AMD's RX 570, NVIDIA's GTX 1650 definitely wins with impressive power efficiency and better temperature and noise levels, but pricing is just too high for that to matter. It also faces strong competition from NVIDIA's own GTX 1060 series, which offers similar or better performance and can be found used for bargain prices. Another competitor is the GTX 1660, which is faster and actually has a 5% better price/performance ratio than the GTX 1650.

https://www.tomshardware.com/reviews/gef...096-6.html

The mainstream graphics market is especially unforgiving. A few frames per second, a few dollars, a few watts of power consumption, or a de-featured encoder can sway comparisons one way or another. That’s bad news for Nvidia’s new GeForce GTX 1650. Depending on the card it’s up against, the 1650 can be made to look great or downright lackluster, which is a problem. The TU117-based GeForce shouldn’t come up so obviously short in any metric.

There’s no doubt that the 1650 is slower than Radeon RX 570. Worse, you can find the Radeon in 4GB trim for $20 less than the least-expensive 1650. We recommend stepping up to an 8GB Radeon RX 570 for a few bucks more. After all, even the higher-capacity version starts $10 cheaper than the base-level GeForce GTX 1650.

When you start eyeballing higher-end partner cards like our GeForce GTX 1650 Gaming OC 4G, Nvidia’s positioning really gets out of whack. Rather than spending $180 on the same class of card, why not spring for a Radeon RX 580 instead? Or, go with a GeForce GTX 1660 for $220. It’s a far better-performing card. This isn't the segment able to tolerate $10 bumps in exchange for a few more megahertz and RGB lighting.
...
GeForce GTX 1650 does well compared to the performance of GeForce GTX 1050 Ti, the original price of GeForce GTX 1060, and Radeon RX 570’s power consumption. It’s just not the all-around winner it could have been. We were big fans of the GeForce GTX 1050, so maybe a card without the six-pin connector would change our minds.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...39250.html

Our colleague Igor Wallossek over at Tom's Hardware Germany has received some breaking news from various sale channel representatives and Nvidia partners. Those partners claim that Nvidia will no longer offer A and non-A Turing dies for the GeForce RTX 2080 and RTX 2070 graphics cards after May.
...
By delivering two variants of the same product, Nvidia paved the way for the chipmaker's partners to use the A variants in the top models and the cheaper non-A variants for the more austere models. In a way, it gave manufacturers certain flexibility to bring different flavors of the same graphics card to the market at different price points. Unfortunately, it also meant that the consumer, who purchased a graphics card with the non-A die, is essentially paying for an inferior product. While Nvidia forbids partners to overclock non-A dies, you, as a consumer, could still overclock the graphics card yourself. However, the chances of you matching the operating speeds of an A die part is pretty slim.

After this month, Nvidia will only offer a single Turing silicon, which is already reportedly in production, for the GeForce RTX 2080 and RTX 2070. That would be the TU104-400 and TU106-400 dies, respectively. Nvidia's latest move could indicate that Turing yields are improving, which is great news for consumers since this theoretically means that faster graphics cards will be arriving on the market at the same price point. For the time being, there is no clue if Nvidia will continue offering the TU102 die, which is used in the GeForce RTX 2080 Ti, in A and non-A variants. Wallossek's sources only mentioned the GeForce RTX 2080 and RTX 2070.

As a last note, Wallossek's information confirms that Nvidia will be rolling out a new vBIOS for the non-A Turing dies so board partners can flash the existing graphics card before sending them off to retailers.

[SteelCrysis]

https://www.tomshardware.com/reviews/gig...118-7.html

When the GeForce GTX 1660 Ti first launched, we looked at EVGA’s XC Black Gaming version of the card and observed that it bested the more expensive (and power-hungry) GeForce GTX 1070, along with AMD’s Radeon RX 590. It also served up better value than the remaining GeForce GTX 1060s and higher-end Radeon RX Vega 56. But today’s landscape looks a little different. The 1060s are almost gone, the remaining 1070s make no sense at all, Radeon RX 590s now start in the $210 range ($50 cheaper than before), and Vega 56 boards dip as low as $300.

In that context, asking an extra $20 for a premium GeForce GTX 1660 Ti may be difficult to justify. After all, the Vega 56 is faster, and Radeon RX 590 is quite a bit cheaper. Nvidia recently followed the 1660 Ti up with a vanilla 1660, compounding the Ti’s predicament by starting down at $220 and still serving up playable FHD performance. Or, for $50 more, you can get your hands on a GeForce RTX 2060 with dedicated RT and Tensor cores, plus enough performance in today’s games to make 2560 x 1440 a playable resolution.

In the end, then, GeForce GTX 1660 Ti’s biggest victory happens when you focus most intently on performance per watt and efficiency. The competition from AMD can’t touch it in those disciplines. If you instead turn to absolute performance, the Radeon RX Vega 56 is the faster choice, despite gobbling up more than 200W of power. Or, if you’re on a budget, the Radeon RX 580 sells for less than $200 and fares really well at 1920 x 1080.

[SteelCrysis]

https://www.extremetech.com/gaming/29160...x-1080-did

Up until now, none of these comparisons have been good for Turing. While we acknowledge that the Steam Hardware Survey is not a perfect metric for GPU market share, it remains the best overall guide that we have to what consumers are purchasing for themselves. This month, finally, Nvidia can claim a single data point moving in its favor: RTX 2070 adoption is happening more quickly than GTX 1080 adoption in 2016, according to reports from Steam in both time periods. We are far enough away from the Pascal launch that we expect GPU availability to be strong in this time period, and the cryptocurrency market hadn’t yet exploded, sending GPU prices through the stratosphere.
...
An ecosystem of words is not the same as an ecosystem of games you can purchase and enjoy right now, and there aren’t enough RT-enabled titles to qualify as an ecosystem — not by a long shot. By the time there are, Turing will have long since been replaced by something new. And claiming major victory for pushing ray tracing into corporate and 3D rendering markets, where it’s existed for years, isn’t much of an achievement to crow about. Nvidia and other companies have worked to improve ray tracing for a decade and more. It’s the availability of real-time ray tracing for gaming, specifically, that would be such a change from the current rasterized status quo. Turing adoption has lagged Pascal, almost across the board. Only now, eight full months after launch, do we see signs that any Turing SKUs are outperforming a Pascal equivalent — and we see it, thus far, in just one GPU SKU.

[SteelCrysis]

https://www.techpowerup.com/256501/nvidi...for-turing

NVIDIA's SUPER teaser hasn't crystallized into something physical as of now, but we know it's coming - NVIDIA themselves saw to it that our (singularly) collective minds would be buzzing about what that teaser meant, looking to steal some thunder from AMD's E3 showing. Now, that teaser seems to be coalescing into something amongst the industry: an entire lineup upgrade for Turing products, with NVIDIA pulling their chips up one rung of the performance chair across their entire lineup.

Apparently, NVIDIA will be looking to increase performance across the board, by shuffling their chips in a downward manner whilst keeping the current pricing structure. This means that NVIDIA's TU106 chip, which powered their RTX 2070 graphics card, will now be powering the RTX 2060 SUPER (with a reported core count of 2176 CUDA cores). The TU104 chip, which power the current RTX 2080, will in the meantime be powering the SUPER version of the RTX 2070 (a reported 2560 CUDA cores are expected to be onboard), and the TU102 chip which powered their top-of-the-line RTX 2080 Ti will be brought down to the RTX 2080 SUPER (specs place this at 8 GB GDDR6 VRAM and 3072 CUDA cores). This carves the way for an even more powerful SKU in the RTX 2080 Ti SUPER, which should be launched at a later date. Salty waters say the RTX 2080 Ti SUPER will feature and unlocked chip which could be allowed to convert up to 300 W into graphics horsepower, so that's something to keep an eye - and a power meter on - for sure. Less defined talks suggest that NVIDIA will be introducing an RTX 2070 Ti SUPER equivalent with a new chip as well.

This means that NVIDIA will be increasing performance by an entire tier across their Turing lineup, thus bringing improved RTX performance to lower pricing brackets than could be achieved with their original 20-series lineup. Industry sources (independently verified) have put it forward that NVIDIA plans to announce - and perhaps introduce - some of its SUPER GPUs as soon as next week.

[SteelCrysis]

https://www.tomshardware.com/news/amd-li...39659.html

Speaking of Nvidia, Barron's actually removed Nvidia CEO Jensen Huang off of its "World's Best CEOs" list because the company "has disappointed on new Turing graphics chips, key to its gaming business."

[SteelCrysis]

https://www.techpowerup.com/256853/nvidi...n-july-2nd

NVIDIA has confirmed that they will be launching a new RTX series of gaming graphics cards, called RTX Super, on July 2nd. According to info from VideoCardz, there will be three models of the new GPUs at launch - RTX 2060 SUPER, RTX 2070 SUPER and RTX 2080 SUPER. The review embargo will lift on the same day as launch day for RTX 2060 SUPER and RTX 2070 SUPER, but the embargo for RTX 2080 SUPER will prevail until "later in July".

The embargo for custom cards based on the new SUPER GPUs will be delayed until July 9th, when we will get the first wave of new cards. There is no apparent reason for the delay, so we will need to find out more about that. Pricing is yet to be announced, but according to the source, it will be "later this week".

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...39752.html

Graphics card news and rumor website VideoCardz has snapped a photograph of what it claims is an upcoming rewarmed GeForce RTX 20-series Super graphics card from Nvidia.
...
According to VideoCardz, the RTX 2080, RTX 2070 and RTX 2060 will get a Super version. Twitter user KOMACHI_ENSAKA found a couple of recent Eurasian Economic Commission (EEC) listings from Asus, Gigabyte and Palit that lend some credence to the rumor. The registered models include the RTX 2080S, RTX 2070S and RTX 2060S models with S reportedly standing for Super. However, it's important to note that the models could be placeholders and not all will make it to market.

[SteelCrysis]

https://techreport.com/news/34655/evga-i...900-series

Most folks don't buy a lot of graphics cards, and if you're anything like me, it's likely that your last graphics purchase was based less on brand and more on price. Still, there's a fair few folks here at TR that like EVGA, and we reckon there's a goodly number of gerbils who feel the same way. If you purchased an EVGA GeForce card from the 700 or 900 series, the company may have a new offer for you.

Here's how it works: if you bought an EVGA GeForce GTX 770, 780, 780 Ti, 970, 980, or 980 Ti card, you can get a credit toward your purchase of a GeForce RTX 2060, 2070, or 2080 direct from EVGA. You have to still have the card, and you need to register it at EVGA's website. If you bought a recertified card, you are explicitly excluded from the promotion, sadly. Likewise, if you live outside of the USA (including Puerto Rico) or Canada, you aren't eligible for the trade-up. Sorry, Bruno.
...
EVGA also says that if you tack on a new EVGA power supply to your RTX graphics card upgrade, you'll get a "significant" discount on that as well, but since I'm not signing up for the program myself, I don't have any way to check what that discount would be. If you decide to take advantage of the offer, keep in mind you'll have to send in your graphics card for EVGA to check first, which means you will be without it for a short while.
...
To be sure, this deal is welcome news for PC gaming enthusiasts who have balked at the high prices of Nvidia's GeForce RTX graphics cards. However, EVGA's other offer—optional interest-free financing with Affirm—may not be as altruistic as it seems. If you're not familiar, Affirm is a service that offers financing on small purchases just like this. The company says it is a more transparent and easier-to-use alternative to credit cards. I haven't used Affirm, so I won't remark on the company's services, but the very idea that buyers might take out a loan to buy a graphics card is rather biting commentary on society in 2019, the prices of said products, or perhaps both.

[SteelCrysis]

https://techreport.com/news/34660/report...due-july-9

A little over a month ago, Nvidia released a YouTube teaser indicating that something "super" was on the way. According to a report from Rock, Paper, Shotgun, that something super is a refresh of Nvidia's entire GeForce RTX lineup. They're reportedly coming soon, too—July 9th, to be exact.

Thanks to some early product listings on Amazon, several RTX 2000-series Super cards will be available in just over a week. Listings for GeForce RTX 2060, 2070, and 2080 Super editions showed up on the e-tailer earlier today but they've since been removed. From the screenshots captured by RPS, it seems that these cards will have just a slight clock boost from their non-Super forms, but the RTX 2060 Super's listing also showed 8 GB of VRAM instead of the vanilla 2060's six. In the listing for EVGA's purportedly upcoming (and sexily-named) GeForce RTX 2070 Super XC Ultra Gaming, the GPU's max boost tops out at 1.8 GHz, vs 1.725 GHz for the non-Super 2070.

[SteelCrysis]

https://www.techpowerup.com/256906/nvidi...cing-outed

NVIDIA has officially confirmed pricing and SKU availability for its refreshed Turing lineup featuring the SUPER graphics cards we've been talking about for ages now. Primed as a way to steal AMD-s Navi release thunder, the new SUPER lineup means previously/released NVIDIA grpahics cards have now hit an EOL-status as soon as their souped/up, SUPER versions are available, come July 2nd.

[SteelCrysis]

https://www.techpowerup.com/256951/nvidi...y-revealed

NVIDIA is giving its GeForce RTX 20-series product family a mid-cycle refresh with the RTX 20 Super-series, which sees performance uplifts across price-points, in the wake of AMD's Radeon RX 5000 "Navi" series. The company is expected to formally announce the series tomorrow (2nd July, 2019), but VideoCardz has learned the dates more relevant to you: market availability. According to them, while all three Super SKUs will be announced on July 2nd, namely the RTX 2060 Super, the RTX 2070 Super, and the RTX 2080 Super; market availability will vary.

https://www.techpowerup.com/256954/evga-...r-pictured

Ahead of formal launch tomorrow (2nd July), followed by a 9th July market availability, pictures of EVGA's custom-design GeForce RTX 2060 Super and RTX 2070 Super graphics cards were leaked to the web by an early listing on Amazon. The RTX 2060 Super XC Ultra is a triple-slot monstrosity, which uses a thicker heatsink ventilated by a pair of spinners; while the RTX 2070 Super XC is the base variant of an entire stack of cards based on this chip. The boxes have clear "Super" branding in the SKU detail corners. This branding is also found on metal decals on the cooler shrouds, and printed on the back-plates.

The listing also confirms that the RTX 2060 Super features 8 GB of GDDR6 memory. This move was necessitated by AMD equipping its Radeon RX 5700 graphics card with 8 GB of 14 Gbps GDDR6 memory. The RTX 2060 Super makes full use of the 256-bit memory bus width of the "TU106" silicon. The RTX 2070 Super, on the other hand, is based on the larger "TU104" chip, and has 2,560 of the 3,072 CUDA cores present on the chip enabled. This SKU is designed to compete with the RX 5700 XT. We'll find pictures of custom-design RTX 2080 Super as we inch closer to its July 23 market availability date.

[SteelCrysis]

https://www.techpowerup.com/257012/nvidi...a-fab-11nm

During our disassembly of the GeForce RTX 2060 Super, we noticed a shocking detail. The 12 nm "TU106" GPU on which it is based, has the marking "Korea." We know for a fact that TSMC does not have any fabs there. The only Korean semiconductor manufacturer capable of contract-manufacturing a piece of silicon as complex as a GPU, for a designer with the energy-efficiency OCD as NVIDIA, is Samsung.

What makes this interesting is that Samsung does not officially have a 12 nm FinFET process. It has 14 nm, and the 11LPP, a 11 nm nodelet, which the company designed to compete with TSMC 12 nm. It would hence be really interesting to hear from NVIDIA on whether they've scaled out the "TU106" to 14LPP, or down to 11LPP at Samsung. It's interesting to note that the shrink in transistor sizes in these nodelets doesn't affect die-sizes. We hence see no die-size difference between these Korea-marked chips, and those marked "Taiwan." We've reached out to NVIDIA for comment.

https://www.tomshardware.com/reviews/nvi...207-8.html

GeForce RTX 2070 Super is an attempt to improve Turing’s standing among gamers who turned their noses up at GeForce RTX 2070 last year. The Founders Edition model we tested is almost 13% faster than its predecessor at a more attractive $500 price point. Nvidia’s partners probably aren’t pleased that they’re now battling a beefy reference design. But gamers benefit, which is what we want to see.

Inserted right between the $350 GeForce RTX 2060 and $500 RTX 2070 Super, we can’t imagine that anyone actually asked for a $400 GeForce RTX 2060 Super. However, if it’s able to outperform Radeon RX 5700 when the vanilla 2060 would have lost, then you know the game Nvidia is playing. Sandwich AMD’s card between a slightly slower and a slightly faster GeForce, then use “but ours has ray tracing” as the coup de grâce to dissuade potential customers. That’s a tough argument to beat, except with a lower price. We’ll have to see how AMD responds.
...
Cramming today’s launch in between AMD’s Radeon RX 5700-series announcement and performance review means that we can’t give the GeForce RTX 2060 Super or 2070 Super a recommendation one way or the other, though. In a few short days, the results from both Navi-based boards will go live. At that point we’ll have a more complete picture of high-end graphics performance in 2019. Count on us to declare a winner once the smoke clears.

https://www.techpowerup.com/review/nvida...er/34.html
The RTX 2070 Super is performing similarly to the Radeon VII at a lower price point.

Results from our new graphics card test suite with all the latest games and a new Core i9-9900K paired with an EVGA Z390 DARK motherboard, show a solid 14% performance improvement over RTX 2070. Last generation's flagship, the GeForce GTX 1080 Ti is 2% slower, so RTX 2070 Super can be considered to deliver equal performance. AMD's fastest, the Radeon VII is 3% behind, and we expect the Navi-based Radeon RX 5700 XT to be around 10% slower than RTX 2070 Super, too. The next step up is the GeForce RTX 2080, which is 7% faster, not that much, especially when you consider the price difference. With those performance numbers, we can recommend RTX 2070 Super for highest detail gaming at 1440p.
...
Releasing on July 9th, priced at $500, the RTX 2070 Super is not cheap. It comes at the exact same price point as RTX 2070 — but with more performance, so effectively price/performance does go up. We reached out to NVIDIA whether the MSRP of RTX 2070 Non-Super will go down now, and they responded that they haven't made any changes to the RTX 2070 MSRP, but naturally its price will go down, following the changed market conditions. In order to sweeten the deal, NVIDIA will include a two-game bundle with all RTX Super cards: "Wolfenstein Youngblood" and "Control." Both titles come with support for NVIDIA RTX raytracing, to show off the capabilities of their new technology. The new RTX Super lineup will definitely turn up the heat on AMD, who are pricing their RX 5700 XT at $450 (for now). RX 5700 XT will definitely not be able to match RTX 2070 Super performance, and we're having doubts whether thermals or noise will end up being comparable, which could make the $50 price increase to Super very justifiable for potential customers, or force AMD to reduce their pricing.

https://www.techpowerup.com/review/nvidi...er/34.html

With these hardware changes, the RTX 2060 Super is a significantly faster card than the original RTX 2060, without beating the original RTX 2070. It gets danger-close, though. At 1080p, the card is 10% faster than the original RTX 2060, and 12% faster at 1440p. The gap increases at 4K to 13%, and although not its forte, quite a few games in our bench are playable with this card. It's hard to say just how much of a dividend the 33% faster and larger memory setup pays, but it could certainly do wonders for the card's future-proofing, especially when you take into account memory sizes on future consoles. Across the competitive landscape, the RX Vega 64 is the closest AMD card to the RTX 2060 Super, and trails by 11% across all resolutions. This landscape will probably change when AMD formally launches the RX 5700 and RX 5700 XT next week.
...
Overclocking the RTX 2060 Super is a breeze, with a massive 29 percent GPU overclock (+200 MHz) without tinkering with the voltages, and an easy-peasy memory overclock to 16 Gbps. It's astonishingly easy to get this card up to RTX 2070 performance-levels, which is probably why we hear rumors of NVIDIA retiring the RTX 2070 from its product stack. What the RTX 2060 Super doesn't do, however, is replace the RTX 2060 at $349. It instead comes at a $50 premium, which is probably NVIDIA trying to offset the cost of more memory chips and stronger VRM solutions. Unless NVIDIA cuts prices of the original RTX 2060, it risks the RTX 2060 Super cannibalizing not just the RTX 2070, but also the original RTX 2060. We would definitely spend the extra $50 on the Super not just for the added performance, but also the future-proofing the added memory brings to the table.

[SteelCrysis]

https://www.extremetech.com/gaming/29440...same-price

Rise of the Tomb Raider shows the RTX 2070 and RTX 2070S more-or-less tying at 1080p, but the faster GPU sharply distinguishes itself at higher resolutions. The gains here are significant enough to represent a reset for Nvidia’s overall product stack. The RTX 2070S is effectively an RTX 2080, for $500 instead of $700. The RTX 2060S is an RTX 2070 at $400 instead of $500. When the RTX 2080S arrives, it’ll presumably improve on the RTX 2080 by 8-10 percent as well given the boost to its overall specs.

This is what Nvidia should have launched last September. It’s a significant enough price improvement to change the overall value proposition of the cards. We’ll be revisiting the question of ray tracing support and the overall value these cards offer against AMD’s upcoming GPUs in weeks to come, but one thing is clear: AMD probably needs to tweak its own launch pricing.
...
AMD may or may not respond with price adjustments before launch, but the 2070S looks like stiffer competition. And these new Turing cards are a noticeable improvement on previous models. All of this sets up a very interesting comparison coming in just a few more days.

[SteelCrysis]

https://techreport.com/news/34666/nvidia...pu-monitor

Hardware isn't all that Nvidia announced today. When the RTX 2060 and 2070 Super cards hit retail on July 9, the company will also release FrameView, a GPU monitoring utility for its graphics cards. The application will report framerates and frame times, along with power consumption and performance per watt metrics. Nvidia says FrameView will support all of the major Windows APIs and UWP games, and will be available for download from Nvidia's site.

[SteelCrysis]

https://www.techpowerup.com/257012/nvidi...a-fab-11nm

Update July 3rd: NVIDIA got back to us

The answer is really simple and these markings are not new. Other Turing GPUs have had these markings in the past. The chip is made at TSMC, but packaged in various locations. This one was done in Korea, hence why his says "Korea".

On an unrelated note: We already use both TSMC and Samsung, and qualify each of them for every process node. We can't comment in any further detail on future plans, but both remain terrific partners.

[SteelCrysis]

https://www.techpowerup.com/257410/nvidi...ter-memory

NVIDIA's upcoming GeForce RTX 2080 Super graphics card doesn't just max out the 12 nm "TU104" silicon and add higher GPU clock-speeds, but also features the highest-clocked GDDR6 memory solution on the market, to make the most of the 256-bit wide memory bus of the silicon. NVIDIA deployed 15.5 Gbps GDDR6 memory, which is 10.7 percent faster than the 14 Gbps memory used on the original RTX 2080 and other RTX 20-series graphics cards. The memory real-clock is set at 1937 MHz compared to 1750 MHz on the original RTX 2080. At this memory frequency, the RTX 2080 Super enjoys a memory bandwidth just a touch short of 500 GB/s, at 496 GB/s.

Besides memory, the RTX 2080 Super maxes out the "TU104" silicon by enabling all 3,072 CUDA cores physically present, as opposed to just 2,944 of them being enabled on the original RTX 2080. The card is also endowed with 192 TMUs, 64 ROPs, 384 Tensor cores, and 48 RT cores. The GPU frequencies are set at 1650 MHz with 1815 MHz GPU Boost, compared to 1515/1710 MHz of the original RTX 2080. NVIDIA is launching the RTX 2080 Super at an MSRP of USD $699, with availability slated for July 23. The company's add-in card (AIC) partners are allowed to design custom-design cards that come with improved cooling solutions and higher clocks.

https://www.techpowerup.com/257419/nvidi...0-ti-super

However, according to Anthony Garreffa from TweakTown, during the GeForce RTX Super release event, he asked senior vice president of NVIDIA's PC business, Jeff Fisher, will there be a release of GeForce RTX 2080 Ti Super graphics card. To that question Jeff replied with a simple "would not be", meaning that the current "Super" lineup is complete as it is. Additionally, it is quite reasonable for NVIDIA not to give RTX 2080 Ti a "Super" treatment as that would close the performance gap between GeForce RTX 2080 Ti and RTX Titan graphics cards, possibly compromising the sales of the RTX Titan GPU.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...39988.html

The benchmark submission in question comes from the Final Fantasy XV Benchmark database, which has listed other pre-release GPUs in the past. While the 2060 Super and 2070 Super were significantly faster than their predecessors, the 2080 Super only scored 8% higher than the original 2080 in the FFXV benchmark. The 2080 Ti clearly remains the fastest GPU for gaming, being about 14% faster than the 2080 Super in this benchmark.

A Geekbench result for the 2080 Super was also found, and the performance difference between the 2080 Super and the original 2080 is similar to what we see in the leaked FFXV benchmark. Geekbench also reports that the 2080 Super will have the full 3072 CUDA cores of the TU104 die the 2080 is based on (as was rumored) and a maximum frequency of 1.81 GHz (it should be noted however that benchmarks like Geekbench aren't always entirely accurate at reporting clock speed).

[SteelCrysis]

https://www.tomshardware.com/reviews/nvi...243-6.html

Although GeForce RTX 2080 Ti lives on at $1,200, Nvidia’s new GeForce RTX 2080 Super improves upon the 2080 Founders Edition’s price and performance by posting average frame rates 6% higher for $100 less. The result is a 21% improvement in performance per dollar. Of course, less expensive 2080s make the 2080 Super a more modest proposition; they already sell for as little as $700, leaving you with just the 6% speed-up to speak of.
...
In the end, GeForce RTX 2080 Super represents a slight performance improvement over GeForce RTX 2080. It’s quite a bit less expensive than GeForce RTX 2080 Founders Edition but matches the price of third-party 2080s already available. The 2080 Super was designed in a way that allows Nvidia’s partners to repurpose their existing efforts. Its power consumption is 25-35W higher though, so expect warmer GPU temperatures, faster fan speeds, and slightly elevated acoustics. Although the GeForce RTX 2080 Super isn’t perfect, enthusiasts who crave more performance than what GeForce RTX 2070 Super or Radeon RX 5700 XT offer only have one option under the opulent GeForce RTX 2080 Ti, and that’s the 2080 Super. If you need the frame rates it offers, this is your only choice short of Nvidia's flagship.

https://www.techpowerup.com/review/nvidi...on/34.html

Overall, when averaged over our benchmarks at 4K resolution, the RTX 2080 Super Founders Edition is 8% faster than the original RTX 2080, which widens the gap to RTX 2070 Super to 16%, restoring the balance in this market segment. NVIDIA's flagship, the RTX 2080 Ti, is 18% faster than the RTX 2080 Super. AMD's fastest, the now end-of-life Radeon VII, is 15% behind, and the new Navi-based RX 5700 XT is 25% slower — not even close. Performance numbers of RTX 2080 Super are good, the higher FPS rates definitely help improve the gaming experience at 4K. While not a 4K60 max details card, it is good enough for solid 4K gaming with decent frame rates, if you are willing to sacrifice some details settings (depending on the game). Its high performance will also help gamers looking to drive a high-refresh-rate monitor beyond 60 Hz, on 1440p, with highest details.
...
Priced at $699, the RTX 2080 Super Founders Edition actually launches at a lower price than the RTX 2080, which launched "starting at $699," with no cards in sight, and even the Founders Edition ended up being priced at $799. Combined with RTX 2080 Super's performance, I think this (small) change in NVIDIA's pricing strategy definitely helps make the RTX 2080 Super more attractive against the in-house competition from both sides of the spectrum. The RTX 2070 Super can be had for $500—around 20% better price/performance, but simply not enough horsepower for 4K gaming. If you have the money, then RTX 2080 Ti at $1100 will get you the 4K60 performance, but at a hefty price premium.

Let's not kid ourselves, $700 is a lot of money, but the options are limited: the Radeon VII is end-of-life, the RTX 2070 Super is too slow for 4K, the original RTX 2080 has only marginally better price/performance and lacks the memory OC headroom of RTX 2080 Super, and the RTX 2080 Ti is probably too expensive for people with just $600-$800 to spend. In order to offset the cost a little bit, and to show off the benefits of their RTX technology, NVIDIA includes a two-game bundle with all RTX Super cards consisting of Wolfenstein: Youngblood and Control, which both support raytracing.

[SteelCrysis]

https://www.techpowerup.com/257776/nvidi...g-possible

While working on GPU-Z support for NVIDIA's new GeForce RTX Super cards, I noticed something curious. Each of the RTX 2060 Super and RTX 2070 Super is listed with three independent device IDs in the driver: 1F06, 1F42, 1F47 for the former and 1E84, 1EC2, 1EC7 for the latter. GeForce RTX 2080 Super on the other hand, like nearly every other NVIDIA SKU, uses only a single device ID (1E81). The PCI device ID uniquely identifies every GPU model, so the OS and driver can figure out what kind of device it is, what driver to use, and how to talk to it. I reached out to NVIDIA, for clarification, and never heard back from them besides an "interesting, I'll check internally" comment.

With no official word, I took a closer look at the actual values and remembered our NVIDIA segregates Turing GPUs article, that was part of the launch coverage for the initial GeForce RTX unveil. In that article, we revealed that NVIDIA is creating two models for each GPU, that are identical in every regard, except for name and price. If board partners want to build a factory-overclocked card, they have to buy the -A variant of the GPU, because only that is allowed to be used with an out of the box overclock. Manual overclocking by the users works exactly the same on both units.
...
Now of course you're wondering what reasons board partners would have to change existing, perfectly fine cards sitting in warehouses, ready to go, into slower (= cheaper) variants. Let's look at the financials quickly. RTX 2080 currently sells for $640, flashing that into RTX 2070 SUPER, selling for $520 makes little economic sense. Things don't look that much better for the $450 RTX 2070 to $400 RTX 2060 SUPER transition. Even if a board partner is having a hard time selling RTX 2080 for $630, they could definitely sell them for $600, and still make a better profit than turning it into a $520 RTX 2070 SUPER. My guess is that NVIDIA wants to increase available inventory for SUPER, reduce inventory of non-SUPER at the same time, and avoid a price fight between non-SUPER end-of-life models and brand-new SUPER cards. If that is the case, then maybe the company is giving its partners some sort of compensation, either as cash, or as rebate for future GPU purchases.

The only other explanation that I can imagine is that if demand for RTX SUPER greatly exceeds that for the older models, then partners could be tempted to convert just enough inventory to SUPER, to ensure that they don't lose sales to other AICs. People who want RTX SUPER now might look at options beyond their established hardware vendor, which could result in lost sales. This is more of a possibility nowadays, with differences between custom-design cards being minimal and coming down mostly to cosmetic differences and cooler / thermal settings. Flashing down small batches would ensure customers are kept happy, while vendors are able to reduce existing warehouse stocks a little bit quicker-essentially a balancing tool.

We're not sure exactly how the board modification is performed, whether solder rework and BIOS flash are done at the AIC factories, or whether NVIDIA does it in-house to protect their methods. We suspect this to be the case, because both the GPU and NVFlash BIOS flashing software have several layers of protection against GPU cross-flashing, and NVIDIA might not want to give its board partners the keys to the kingdom. On the other hand, if these cards can be flashed back (even if using an external flasher), and the solder mod is reasonably easy to do, then this could end up being an amazing method for people to flash their SUPER cards back to RTX 2070 / 2080 and gain some free performance in the process. These cards are full RTX 2080 / 2070 cards after all, there can be no defective units, otherwise that cards wouldn't have made it through QA in the first place.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...40166.html

Currently, there are four Nvidia graphics card that use the Turing TU102 silicon. The GeForce Titan RTX and GeForce RTX 2080 Ti hail from Nvidia's mainstream product line while the Quadro RTX 8000 and Quadro RTX 6000 belong to the enterprise side. Thanks to AIDA64's latest changelog, we're almost certain that Nvidia is working on another TU102-based graphics card behind the scenes.
...
During a secret E3 meeting outlining its Super cards, Nvidia stated that, at that time, the chipmaker had no short or long-term plans to introduce a GeForce RTX 2080 Ti Super. As you are well aware, plans are always changing. Now that AMD President and CEO Dr. Lisa Su has confirmed that high-end Navi, or commonly called Big Navi, are on the way, Nvidia could be gearing up new models to stand its ground.

[SteelCrysis]

https://www.techpowerup.com/258340/asset...nvidia-rtx

Assetto Corsa Competizione, the big AAA race simulator slated for a September release, will lack support for NVIDIA RTX real-time raytracing technology, not just at launch, but even the foreseeable future. The Italian game studio Kunos Simulazioni in response to a specific question on the Steam Community forums confirmed that the game will not receive NVIDIA RTX support.
...
This is significant, as Assetto Corsa Competizione was one of the posterboys of RTX, and featured in the very first list by NVIDIA, of RTX-ready games under development.

[SteelCrysis]

https://www.techpowerup.com/258404/nvidi...g-is-crazy
Before you believe what Jensen is saying, keep this reality in mind: https://www.extremetech.com/gaming/27589...-gaming-pc

Specifically, Huang said that "SUPER is off to a super start for and at this point, it's a foregone conclusion that we're going to buy a new graphics card, and it's going to the last 2, 3, 4 years to not have ray tracing is just crazy. Ray tracing content just keeps coming out. And between the performance of SUPER and the fact that it has ray tracing hardware, it's going to be super well positioned for throughout all of next year."

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...40235.html

When Nvidia launched their new Super RTX GPUs, it seemed like a given that we'd eventually see a 2080 Ti Super, especially because rumors pointed to its existence from the get-go. Earlier this month the program AIDA64 got an update for a new TU102-based GPU, and some thought this "T10-8" RTX GPU could be the new 2080Ti Super.

If this were the 2080 Ti Super, we'd expect this GPU not to be cut down, or at least less cut-down than the 2080Ti. "If" being the keyword, because it seems that this is not a consumer GPU, but a GPU specifically made for Nvidia's GeForce Now streaming service.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...40418.html

You might want to wait before emptying your wallet for an RTX 2060-equipped gaming laptop. Notebook Check reported on Sunday that Nvidia's latest entry-level mobile GPU performs worse than expected--anywhere from 20% to 25%--when compared to its desktop counterpart.
...
That's a little problematic when there are slight discrepancies between a GPU's mobile and desktop versions. But according to Notebook Check, the RTX lineup's performance gap is fairly large. The mobile RTX 2070 is 14% to 18% slower than its desktop counterpart; the mobile RTX 2080 narrows the gap to a difference of just 8% to 10%. That at least means the products get closer to parity as they rise in price.

It's unclear how much of this may be a difference may also be due to the difference in thermals from a laptop to a desktop or other circumstances .
...
We've reached out to Nvidia for comment on Notebook Check's claims and will update this post when the company responds.

[SteelCrysis]

https://www.techpowerup.com/259341/nvidi...tober-22nd

According to the recent round of rumors, NVIDIA could extend its budget GPU offering on October 22nd when it will launch the new GeForce GTX 1650 Ti graphics card. Expected to sit between GTX 1650 and GTX 1660, the new graphics card is supposed to be NVIDIA's answer to AMD's unannounced, low-end NAVI GPUs rumored to be called the RX 5600 series.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...40474.html

Nvidia GeForce GTX 1660 Super cards may be coming in October. A report from Videocardz claims that sources from Asus have confirmed the card is coming, including models from the Dual Evo, Phoenix and Tuf3 lines.
...
Update: 2:15 p.m. ET: Nvidia told Tom's Hardware that "we do not comment on rumors."

We reached out Asus about this story but did not hear back before publication. This story will be updated if and when the company responds.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...40528.html

A new Gigabyte filling with the Eurasian Economic Commission (EEC) hints that, in addition to the GeForce GTX 1660 Super, Nvidia could be preparing the GeForce GTX 1650 Super as well to face AMD's upcoming Radeon RX 5500.

[SteelCrysis]

https://www.tomshardware.com/news/gtx-16...lease-date

VideoCardz posted what it said are product images for the regular and AMP version of the Zotac Gaming GeForce GTX 1660 Super. Visually, both graphics cards resemble their non-Super counterparts. If it wasn't for the Super moniker on the packaging, we could swear we were looking at the GTX 1660.

[SteelCrysis]

https://www.tomshardware.com/news/two-of...n-the-wild

With rumors of Nvidia's GeForce GTX 1660 Super series launch in full force, we're hearing murmurs of third-party GTX 1660 Super graphics cards as well. Today, VideoCardz.com reported that MSI has got two GTX 1660 Super cards in the works, the GTX 1660 Super Ventus XS OC and the GTX 1660 Super Gaming X.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...lease-date

A listing on JD.com lists specifications for the unreleased Nvidia GeForce GTX 1660 Super graphics card, as spotted by longtime hardware leaker @momomo_us.
...
The most obvious difference between the two graphics cards is the memory type. The GTX 1660 debuted with GDDR5 memory clocked at 2,000 MHz (8,000 MHz effective). The Super version will have seemingly have speedier GDDR6 memory at 1,750 MHz (14,000 MHz effective). Although both graphics cards are still restricted to a 192-bit memory interface, the memory swap should let the GTX 1660 Super deliver 75% more memory bandwidth than its sibling.

JD lists the Maxsun GeForce GTX 1660 Super Terminator for 1,899 yuan (~$268.36). The non-Super Terminator version sells for 1,499 yuan (~$211.83). That's about a 26.7% premium over the GTX 1660.

With October 29 being the rumored launch date for the GTX 1660 Super, hopefully we'll soon find out if the upgrade to GDDR6 memory is enough deserve our consideration.

[SteelCrysis]

https://www.techpowerup.com/260391/nvidi...with-gddr6

NVIDIA's GeForce GTX 1660 SUPER, the first non raytracing-capable Turing-based SUPER graphics card from the company, is set to drop on October 29th. Contrary to other SUPER releases though, the GTX 1660 SUPER won't feature a new GPU ship brought down from the upwards performance tier. This means it will make use of the same TU116-300 as the GTX 1660 with 1408 CUDA cores, not the 1536 CUDA count of the GTX 1660 Ti. Instead, NVIDIA has increased performance of this SUPER model by endowing it with GDDR6 memory.
...
Of course, the fewer CUDA core resources compared to the GTX 1660 Ti mean it should still deliver lower performance than that graphics card. This justifies its price-tag set at $229 - $20 higher than the GTX 1660, but $50 less than the GTX 1660 Ti.

[SteelCrysis]

https://www.tomshardware.com/news/nvidia...s-graphics

Just this week we heard of a rumored launch date for the Nvidia GeForce GTX 1660 Super, October 29. But that's not all she wrote. VideoCardz has also reported that Nvidia's preparing a Super version of the GTX 1650 that's will arrive November 22.

https://www.neowin.net/news/nvidia-gefor...k-database

The 'SUPER' variant of Nvidia's GeForce GTX 1660 GPU is rumored to be launching on October 29, and as is often the case with a product launch, a few performance numbers almost always manage to pop up before the official embargo is lifted. The 1660 SUPER is no exception with the GPU seemingly making an appearance in the Final Fantasy XV Windows Edition benchmark database. The graphics card was measured at 2560x1440 resolution using the high quality preset. It has managed to score 4794 points making it significantly faster than the vanilla 1660 and is nearly as fast as the GTX 1660 Ti.
...
The 1660 SUPER allegedly has the same number of CUDA cores as its non-Super counterpart, so most of the improvement seen here can be attributed to the higher bandwidth provided by the addition of GDDR6 memory. It is important, however, to note that such gains may not be seen in every title.

[SteelCrysis]

https://techreport.com/news/3466434/nvid...ctober-29/

Pretty much everyone agrees: Nvidia’s Super cards are super good. They’re also super expensive. If you’re looking for a video card refresh and don’t need ray tracing, though, Nvidia’s latest Super cards might suit you. The rumored GeForce GTX 1660 Super will launch on October 29, Nvidia has confirmed to VideoCardz.com.

[SteelCrysis]

https://www.techpowerup.com/review/msi-g...-x/37.html

NVIDIA's new GeForce GTX 1660 Super is more of an incremental update than a radical redesign of the GTX 1660. While the GTX 1660 used GDDR5 memory, the GTX 1660 Super has GDDR6, which ticks at much faster speeds, too, increasing memory bandwidth by 75%. NVIDIA also priced the GTX 1660 Super much more aggressively, at $230 MSRP, to preempt AMD's Radeon RX 5500, which is expected to launch later this year.
...
We do know that AMD's Radeon RX 5500 comes with 8 GB of VRAM, whereas the GeForce GTX 1660 Super only has 6 GB. This will definitely become an important factor, mostly for marketing, to reel in less tech-savvy buyers, as upcoming consoles will have more VRAM than before and scaring people that what they buy today won't be enough tomorrow has always worked. However, our performance data shows that even the aging GeForce GTX 1060 3 GB doesn't see any significant performance deficits because of its 3 GB of VRAM—even at the highest settings with 1080p. The 4K results confirm that lack of memory will lower FPS, but 4K is unplayable either way on the GTX 1060 because of the lack of shading power; the same will be true for the 6 GB on the GTX 1660, so don't worry about that.

Turing has impressed everyone with its power efficiency, and the GTX 1660 Super can deliver here, too. Being based on the same TU116 GPU as GTX 1660 and GTX 1660 Ti, we've not been expecting anything else. Both older cards are a little bit more efficient, which is probably because the high-clocked GDDR6 memory on GTX 1660 Super draws a bit more power, but overall, the PSU requirements are minimal. This is an important factor because NVIDIA wants people to upgrade their prebuilt gaming PCs, too, and these often come with weak power supplies of questionable quality. Less than 150 W in gaming should be no problem for any PSU, and they all include a single 8-pin power connector, too. MSI increased their card's power limit a little bit, up to 130 W, which is only a 5 W increase. I would have wished for more as that would have yielded additional performance, because NVIDIA's Boost algorithm would boost higher for longer. The card's heatsink could certainly handle the extra heat.
...
NVIDIA's GTX 1660 Super reference pricing is $230, which is surprisingly competitive in the current landscape. This puts the card on the same price/performance level as AMD's Radeon RX 580 and RX 590 and dials up the heat on AMD and its board partners. I'm really looking forward to seeing what their new Radeon RX 5500 "Navi" card can bring to this price segment. MSI is asking for another $30 for their Gaming X, which I think is reasonable; you get a much better cooler, more monitor outputs, and a small performance increase for it. NVIDIA's new pricing kind of obsoletes the GTX 1660 Ti, which is currently $275 and doesn't really bring much to the table to justify the extra cost. Maybe NVIDIA could lower its price to $250 to clear existing inventory and then end-of-life the card. The next-fastest NVIDIA SKU, the RTX 2060, is $340 now, which leaves quite a big hole in the product stack, so maybe we'll see yet another card to fill this void at around $300, or AMD could try to capture it with an upcoming Radeon RX 5600. It looks like the $200–$300 market segment is suddenly going to become very interesting.

[SteelCrysis]

https://www.tomshardware.com/reviews/evg...sc-ultra/6

Nvidia set out to fill the gaps in the budget space and improve performance over the existing lineup. With the GTX 1660 Super, the company has accomplished just that. In our testing, we saw the 1660 Super generally trail the more expensive 1660 Ti. When testing higher, more GPU-dependent resolutions, the 1660 Ti slightly extends its lead in some of these titles, but not by much. The new Super card held its own next to the more expensive Ti, to say the least.

Price versus performance is an important factor for many users, and we’ve seen that improve with the previous high-end Super cards. The value of the Super cards in reference to performance is much better than the vanilla, non-Super versions that came before them. A quick trip to Newegg shows GTX 1660 cards priced at $225-plus, and the 1660 Ti’s cheapest entry is $259.99. Nvidia’s MSRP for the 1660 Super is $229, which is also the price of the EVGA card we tested today. Without a price adjustment, it’s hard to argue for the 1660 Ti given it lands just slightly ahead of the 1660 Super and currently retails for at least $30 more.
...
Overall, Nvidia continues to bring pricing down and performance up with its Super cards, which is certainly a welcome change. While performance was always there with Turing, the value aspect was lacking, and the Super cards have brought that concept more in line with the hopes of enthusiasts. The EVGA GTX 1660 Super SC we tested for this review will supposedly land at the $229 MSRP. As pricing remains significantly below the 1660 Ti’s price point, the GTX 1660 Super is going to be an appealing card for 1080p and budget 1440p gamers at or near $230.

[SteelCrysis]

https://www.techpowerup.com/review/msi-g...-x/35.html

Video memory size of 4 GB might sound low at first, but you have to consider that pricing matters a lot in this segment. Adding more memory would make the card more expensive with little or no performance difference at 1080p Full HD. Looking at our performance numbers, we can definitely see reduced FPS at 4K resolution compared to cards with more memory, but I'm not seeing anything in our data that would suggest these cards are memory-bound at 1080p. We're running maximum details, including optional HD texture packs when available, so there are lots of options to reduce memory requirements should it come to that. I rather buy a more affordable card now than waste money on future-proofing, at least in this segment.

NVIDIA has positioned their new GTX 1650 Super very aggressively. With an MSRP of $160, the card is priced similarly to AMD's aging Radeon RX 580, yet offers better performance, noise, and thermals. The Radeon RX 590 is $190 at the moment, down from its launch price of $280—I'd rather buy the GTX 1650 Super.