Danish online retailer Proshop.dk jumped the gun with listing its stock of NVIDIA GeForce RTX 4080 "Ada" graphics cards. These include custom-design cards from ASUS, MSI, GIGABYTE, and Inno3D. While pictures of GIGABYTE and Inno3D cards weren't up, the site put out pictures of ASUS ROG Strix, TUF Gaming; and MSI Gaming X and Ventus 3X models. Prices range between 12,190 Danish Krone (US $1,626) to 14,590 Krone ($1,946), including taxes. The pictures give us our first look at MSI's new cost-effective Ventus 3X custom-design, which will be positioned close to NVIDIA's baseline MSRP (of $1,199 for the RTX 4080). The ASUS ROG Strix, TUF Gaming, and MSI Gaming X designs appear identical their RTX 4090 siblings. These listings appear at least 16 days ahead of when these cards are expected to be available from.

Despite sticking with PCI-Express Gen 4 as its host interface, the NVIDIA GeForce RTX 4090 "Ada" graphics card standardizes the new 12+4 pin ATX 12VHPWR power connector, even across custom-designs by NVIDIA's add-in card (AIC) partners. This tiny connector is capable of delivering 600 W of power continuously, and briefly take 200% excursions (spikes). Normally, it should make your life easier as it condenses multiple 8-pin PCIe power connectors into one neat little connector; but in reality the connector is proving to be quite impractical. For starters, most custom RTX 4090 graphics cards have their PCBs being only two-thirds of the actual card length, which puts the power connector closer to the middle of the graphics card, making it aesthetically unappealing, but then there's a bigger problem, as uncovered by Buildzoid of Actually Hardcore Overclocking, an expert with PC hardware power-delivery designs.

CableMod, a company that specializes in custom modular-PSU cables targeting the case-modding community and PC enthusiasts, has designed a custom 12VHPWR cable that plugs into multiple 12 V output points on a modular PSU, converting them to a 16-pin 12VHPWR. It comes with a pretty exhaustive set of dos and don'ts; the latter are more relevant: apparently, you should not try to arm-wrestle with an 12VHPWR connector: do not attempt to bend the cable horizontally or vertically close to the connector, but leave a distance of at least 3.5 cm (1.37-inch). This ensures reduced pressure on the contacts in the connector. Combine this with the already tall RTX 4090 graphics cards, and you have yourself a power connector that's impractical for most standard-width mid-tower cases (chassis), with no room for cable-management. Attempting to "wrestle" with the connector, and somehow bending it to your desired shape, will cause improper contacts, which pose a fire-hazard.Update Oct 26th: There are multiple updates to the story.

NVIDIA is responding to reports of the 12+4 pin ATX 12VHPWR power connector of its new GeForce RTX 4090 "Ada" graphics cards being unreliable, and posing a potential fire hazard. The company has reportedly instructed its add-in card (AIC) partners, companies that sell custom-design graphics cards; to collect all retail graphics cards with burnt power connectors, and send them over directly to NVIDIA HQ for investigation. Reports of the 12VHPWR connectors overheating due to improper terminal contact aren't new, but this is the first time a retail product implementing the connector is experiencing reliability issues.

It came to light when a Reddit user posted pictures of a melted 12VHPWR connector from an NVIDIA-supplied adapter that converts four 8-pin PCIe to one 600 W-capable 12VHPWR. There is also charring on the female connector on the card, but the user claims that the card is functional. Later this week, another Reddit user posted similar pictures of a burnt connector for their RTX 4090 card. NVIDIA director of global PR for GeForce, Bryan Del Rizzo, in a statement to The Verge, said that the company is in touch with the first owner who reported this problem, and is reaching out to the other, as part of their investigation.

EK, the premium liquid cooling gear manufacturer, is now introducing the ultimate water cooling solution for MSI Trio and Suprim GeForce RTX 4090 graphics cards. The EK-Quantum Vector² Trio RTX 4090 D-RGB comes in the form of a water block with a passive backplate or a cooling solution with an active backplate that's sandwiching the card between two water blocks for the best possible cooling results.

The all-new RTX 4090 Trio and Suprim cards are expected to be significantly more powerful than their 30-series counterparts. And while their efficiency is also improved, these cards still pull over 450 W, which means they will benefit even more from water cooling, mainly due to the 4 nm production process, which increases thermal density compared to the previous 8 nm chips.

Phanteks announces the release of the new Glacier G40 ASUS GPU block, designed specifically for the ASUS STRIX/TUF RTX 4090/4080 Series cards. The Glacier G40 ASUS GPU Block brings the ultimate cooling performance with unique water block features to build stunning water-cooled systems. Phanteks' Glacier G40 ASUS GPU Block provides a high-performance water-cooling solution custom-designed for the latest ASUS GeForce RTX 4000 series cards. A transparent acrylic jet plate is positioned directly over the GPU die to ensure optimal heat transfer. The backplate is included with all models to provide a smooth integrated look while providing additional cooling to the back of the PCB. In addition, the full-cover copper cold plate directly cools all memory and power delivery components on the PCB with high-quality thermal pads.

The Glacier G40 ASUS GPU Block also introduces some new features, such as a new fitting location to allow for better tube routing, even in the most compact chassis. To improve stability, a GPU support bracket is added to connect the water block to the rear PCI bracket so that the GPU is mounted more securely. Like all Phanteks' Glacier products, the Glacier G40 ASUS GPU Block is made in Taiwan using only 100% copper, cast acrylic, anodized aluminium covers, and extremely durable Viton O-rings known from the automotive and aerospace industries. The integrated D-RGB lighting beautifully complements the water block and can synchronize with Phanteks' D-RGB accessories and compatible motherboards.

ARCTIC, leading manufacturer of silent PC coolers and components, today informs about the new INNO3D GEFORCE RTX 4090 ICHILL BLACK graphics card. We love innovative projects and enthusiastically took on the challenge of taming NVIDIA's power-hungry new release. Working closely with INNO3D, we created an efficient liquid cooling system for graphics cards. With technical know-how, in cooling combined with a clean, uncluttered design and straight lines and edges, the INNO3D GEFORCE RTX 4090 ICHILL BLACK graphics card is rightfully called "Brutal by Nature".

We would like to thank INNO3D for their very good collaboration. The new GeForce RTX 4090 iCHILL BLACK features a hybrid cooling solution that combines quiet airflow with powerful water cooling. We have been creating cooling solutions for computer hardware for over 20 years. Successful products such as the P-Series fans as well as the ARCTIC Liquid Freezer II have found enormous appeal in the hardware industry. We set out to combine the successful aspects of these products to create a high-performance GPU cooler.

It turns out that NVIDIA didn't just cancel (unlaunch) the GeForce RTX 4080 12 GB last week, but also shelved the SKU until it is needed in the product stack. This is probably because NVIDIA intended to sell it at $900, and will find it difficult to justify a xx70-class SKU at this price-point. A Moore's Law is Dead report goes into the possible reasons NVIDIA shelved the RTX 4080 12 GB, and why it won't be rebadged as the RTX 4070.

The RTX 4070, although expected to be based on the same AD104 silicon as the RTX 4080 12 GB, won't have the same configuration. The RTX 4080 12 GB maxed out the AD104, enabling all 7,680 CUDA cores on the silicon. It's likely that the RTX 4070 will have fewer CUDA cores, even if it retains the 192-bit memory interface and 12 GB memory size. The memory clock could be changed, too. The RTX 4080 12 GB was essentially NVIDIA trying to upsell the successor of the RTX 3070 Ti (maxed out GA104) as an xx80-class SKU, at a higher price-point. Moore's Law is Dead also showed off possible designs of the RTX 4070 Founders Edition, revealing a compact design with many of the same design improvements implemented with the RTX 4090 FE. This card comes in a strictly 2-slot design.

Earlier today, we reported a story on MSI increasing the price of its GeForce RTX 4090 SUPRIM Liquid X graphics card from its $1,750 MSRP to $1,935 in just a week following its market availability (more importantly, just a week after the card heaped praise for being a well-priced custom flagship product). This development first appeared on Reddit, where keen-eyed Redditor Jmulcahey spotted the price-hike first. The $1,935 price appeared on the company's US online store (a first-party retailer), where any changes in prices are taken to be changes in the MSRP.

MSI responded to the development, saying that the increased price on the MSI US online store was the result of a miscommunication, that the MSRP of the RTX 4090 SUPRIM Liquid X remains at $1,750; and more importantly, that the company will refund the price-difference to anyone who purchased the card at the increased price of $1,935 (i.e. a $185 refund). The company released a statement to TechPowerUp, which we are posting verbatim below.

MSI debuted its flagship GeForce RTX 4090 SUPRIM Liquid X graphics card to gleaming reviews as one of the few custom-design RTX 4090 cards that fits inside a 2-slot design, freeing up expansion slots on your motherboard, thanks to its AIO liquid cooling solution. One of the most striking aspects of the card was its MSRP of USD $1,750, a mere $150 (9% premium), which made the card look like an overall great purchase, given that ASUS was pricing the air-cooled 4-slot ROG Strix RTX 4090 OC at $2,000 (25% premium). This had an impact on the SUPRIM's performance-per-Dollar calculations in reviews, in relation to pricier options. It turns out that MSI has revised the price of the RTX 4090 SUPRIM Liquid X a mere 1 week into its launch, increasing it by $185.

The card is now priced at $1,935 (20% premium over NVIDIA baseline), closer to that of the ASUS ROG Strix. This isn't a marked-up price by a retailer, but the latest price on MSI's first-party online store for the US market. Price premiums of custom-design RTX 4090 cards have been conservative across the board, particularly for a flagship card like the SUPRIM Liquid X, but a price revision this soon after launch doesn't go down well with buyers, and makes it seem like the card was underpriced just for the reviews—something ASUS avoided when they priced the ROG Strix at $2,000 right off the bat. And yes, we are mindful of the concept of introductory pricing, but only when it's advertised as such. An introductory price is meant to be disclosed to be as such, making buyers [and reviewers] aware that the card's actual MSRP is something else/higher. When we contact companies for pricing to use in our reviews, we clearly seek MSRPs, and not introductory pricing for the company to sell a tiny inventory at low prices, only to raise them a week later.Update 10:26 UTC: The price on the MSI website is now back to $1750. Those who bought the card at $1,935 are being refunded the price difference. More on this here.

Intel Arc A770 "Alchemist" graphics card has an idle power-draw problem. It pulls 44 W (card-only) power when idling. This used to be acceptable some 15 years ago, but GPU idle power-draw has come a long way since. The reigning Goliath GeForce RTX 4090 pulls just 21 W when idling, and the RTX 3070, the card the A770 was extensively compared against, only pulls 9 W—that's 7 LED downlights worth power-difference between the A770 and RTX 3070. Intel has a workaround to this problem: enable the PCI-Express active state power management (ASPM) setting to L1 mode in your motherboard's UEFI BIOS setup program.

The Intel Xe-HPG "Alchemist" graphics architecture reportedly uses PCIe Gen 2-era L0 and L1 ASPM, which needs to be forced via software settings. To do this, find the PCIe ASPM settings in your BIOS setup, and enable them with the "L1" setting. You then make your way to Power Options in the Windows Control Panel, edit your active power scheme, and manually set the PCI-Express "Link state power-management" to "Maximum." This affects the power-management behavior and performance of all PCIe devices in your system, including NVMe SSDs, not just the graphics card. Intel did not put out its power-draw numbers for this workaround, but we intend to test it as soon as we can.

In the run up to the November 3 reveal of the next-generation RDNA3 architecture, and with the 43% faster RTX 4090 mauling away its appeal to the enthusiast crowd, the AMD Radeon RX 6900 XT got another round of price-cuts, and can be had for as low as $669. Prices are down on both sides of the big pond, with European retailers listing it for as low as 699€. Although not technically AMD's flagship graphics card, with the RX 6950 XT (starts at $869); the RX 6900 XT is a formidable 4K gaming graphics card with a high performance-per-Dollar at its new price (roughly 35% higher than the RTX 4090). AMD's latest round of official price-cuts happened around mid-September as the company was bracing for the RTX 4090 "Ada."

It's barely been a day, and the $1600-2000 NVIDIA GeForce RTX 4090 "Ada" graphics card is already out of stock in leading US retailers such as Newegg. Either the inventory was barely large enough to last a day; or the scourge of scalping is back to haunt graphics card buyers. NVIDIA is testing a new way of retail for its GeForce RTX 4090 Founders Edition graphics cards, which it calls "Verified Priority Access." From the looks of it, this is copied from EVGA ELITE Priority Access program; where you reserve a graphics card purchase regardless of it being in stock, and when the reseller is ready with some stock, you're given a position in a queue, with a limited time window in which to complete your purchase and place the order.

NVIDIA Verified Priority Access (VPA) program is only available in a handful markets, and through participating retailers. These include Best Buy (United States), Scan (United Kingdom), NBB (Germany & Netherlands), and LDLC (France, Italy, and Spain). One big catch with VPA is that you should already be owning a GeForce GTX 10-series, GTX 16-series, RTX 20-series or RTX 30-series graphics card; and be signed in with GeForce Experience. When your position in the purchase queue advances or opens up for you to make the purchase, you will be notified through a pop-up in GeForce Experience, from where you'll be taken to a page by the participating online retailer to buy your card.

With the Eiswolf 2 AiO and the ES cooling block from the Enterprise Solutions series, Alphacool presents two innovative solutions for cooling Nvidia's latest graphics card. In order to dissipate the enormous waste heat of this graphics card generation in the best way possible, numerous optimizations have been made to the water cooler compared to the previous models. The fin structure has been adapted and allows a very high water flow while increasing the cooling surface. The modified jet plate with improved inflow engine ensures optimum distribution of the water over the cooling fins. The fully chrome-plated copper base is resistant to acids, scratches and damage. Additionally, the chrome plating provides homogeneity and shine that cannot be achieved by nickel plating.

This improved cooler is now being used in Alphacool's Eiswolf All-in-One as well as Enterprise Solutions in the industrial sector. The expandable Eiswolf 2 still scores as an all-round carefree package and ensures that its user can easily take advantage of a performance-oriented GPU water cooling with digital aRGB illumination out-of-the-box. Also still included are the NexXxos ST30 360 mm radiator, the Rise Aurora 120 mm fans, the Eiswolf 2 pump top with integrated DC-LT 2 pump and the TPV hoses with quick-release coupling. The ES cooler impresses with its technical design. The positioning of the connections at the rear allows easy integration of the GPU cooler into the water loop even in the tightest server cases. The discreet material mix of carbon and chrome-plated copper makes it additionally interesting for private users who can do without aRGB lighting.

EK, the premium liquid cooling gear manufacturer, is now introducing the ultimate water cooling solution for ROG Strix and ASUS TUF GeForce RTX 4090 graphics cards. EK-Quantum Vector² Strix/TUF RTX 4090 D-RGB comes in the form of a water block with a passive backplate or a cooling solution with an active backplate that's sandwiching the card between two water blocks for the best possible cooling results.

The all-new RTX 4090 Strix and TUF cards are expected to be significantly more powerful than their 30-series counterparts. And while their efficiency is also improved, these cards still pull over 450 W, which means they will benefit even more from water cooling, mainly due to the 4 nm production process, which increases thermal density compared to the previous 8 nm chips.

Rather than another leak of performance numbers for NVIDIA's upcoming RTX 4090 and 4080 cards, the company has shared some performance numbers of their upcoming cards in Overwatch 2. According to NVIDIA, Blizzard had to increase the framerate cap in Overwatch 2 to 600 FPS, as the new cards from NVIDIA were simply too fast for the previous 400 FPS framerate cap. NVIDIA also claims that with Reflex enabled, system latency will be reduced by up to 60 percent.

As for the performance numbers, the RTX 4090 managed to 507 FPS at 1440p in Overwatch 2, using an Intel Core i9-12900K CPU, at the Ultra graphics quality setting. The RTX 4080 16 GB manages 368 FPS, with the RTX 4080 12 GB coming in at 296 FPS. The comparison to the previous generation cards is seemingly a bit unfair, as the highest-end comparison we get is an unspecified RTX 3080 that managed to push 249 FPS. This is followed by an RTX 3070 at 195 FPS and an RTX 3060 at 122 FPS. NVIDIA recommends an RTX 4080 16 GB if you want to play Overwatch 2 at 1440p 360 FPS+ and an RTX 3080 Ti at 1080p 360 FPS+.

TechPowerUp today released the latest version of TechPowerUp GPU-Z, the popular graphics sub-system information and diagnostics utility for PC enthusiasts and gamers. Version 2.50.0 adds support for NVIDIA GeForce RTX 4090, and improved support for Intel Arc A770, and Arc A750. Video memory chip temperature monitoring has been added for Arc "Alchemist," and the overall temperature sensors for the Arc series have been fixed. The DLSS game scanning has been improved to tell if no DLSS-compatible games have been found.

While the number "2.50" might sound special, it is not, it is simply 49+1. We also had to get this build out quickly so that reviewers can do their jobs playing with the latest hardware from Intel and NVIDIA.

Grab GPU-Z from the link below.

DOWNLOAD: TechPowerUp GPU-Z 2.50.0

With the new generation of GeForce RTX 4000 cards announcement, Phanteks today unveils the upcoming Glacier G40 GPU water blocks for RTX 4000 Series cards. The Glacier G40 GPU blocks bring the ultimate cooling performance with some unique water block features to build amazing water-cooled system. With the new generation of GeForce RTX 4000 cards announcement, Phanteks today unveils the upcoming Glacier G40 GPU water blocks for RTX 4000 Series cards. The Glacier G40 GPU blocks bring the ultimate cooling performance with some unique water block features to build amazing water-cooled systems.

Phanteks' Glacier G40 GPU blocks provide a high-performance water-cooling solution custom-designed for the latest GeForce RTX 4000 series cards. A clear acrylic jetplate directly over the GPU die ensures optimal heat transfer. The backplate is included with all models to provide a clean integrated look while also providing additional cooling to the back of the PCB. In addition, the full cover copper coldplate directly cools all memory and power delivery components on the PCB with high quality thermal pads.

As with the recent pre-launch listing of 13th Gen Core desktop processors, US retailer Newegg put out listings of various custom-design NVIDIA GeForce RTX 4090 "Ada" graphics cards, revealing their launch prices. These prices appear close enough to the $1,599 baseline price set by NVIDIA, to conclude that pricing sanity is slowly returning to the graphics card market. A lot will however depend on how the market behaves on October 12, when the RTX 4090 goes on sale; particularly whether scalpers vaporize inventory within minutes. Even if they did, scalpers would only see demand from the niche that actually wants to spend north of $1,600 on a graphics card, there are no crypto-currency miners lining up to buy graphics cards. Especially not after the Ethereum merge.

To illustrate that AIC prices are beginning to appear normal, one needs to look at the pricing of the MSI RTX 4090 SUPRIM Liquid, supposedly MSI's most premium RTX 4090 product, which is priced at $1,749, or just a $150 premium over the NVIDIA baseline. Several cards such as the ASUS TUF Gaming, GIGABYTE WindForce OC, and MSI Gaming (standard), are listed bang on the $1,599 baseline, while their OC siblings are at a small premium. The ASUS ROG Strix O24G is the most expensive card of the lot, priced at $1,999, or a $400 premium.

Today, Alienware continues its mission to create premier gaming experiences with a triple-threat of new devices, including: a revamped Aurora R15 Gaming Desktop, a compact and feature-rich Tenkeyless Gaming Keyboard, and our second QD-OLED Gaming Monitor positioned at a lower price point. The new Alienware Aurora R15 desktop sees a significant performance boost, thanks to the latest 13th Gen Intel Core processors, NVIDIA GeForce RTX 40 Series GPUs and an armada of system refinements.

The Aurora R15 now also includes up to a 1350 W power supply (up from 1000 W in the previous generation), designed to support NVIDIA's flagship GeForce RTX 4090 graphics card. Speaking of the graphics card, we repositioned the graphics slot to enable larger card designs (up to triple wide cards). In addition to offering NVIDIA GeForce RTX 40 Series and 30 Series GPUs, Aurora R15 will also be configurable with AMD Radeon RX 6000 Series graphics which pair nicely with our first QD-OLED AMD FreeSync monitor… more on that below.

Intel today launched its 13th Gen Core "Raptor Lake" desktop processors, and companion 700-series motherboard chipset. These processors are built in the same LGA1700 package as the previous generation "Alder Lake," and are backwards-compatible with 600-series chipset motherboards through a BIOS update. Likewise, 700-series chipset motherboards support older "Alder Lake" processors. With the new 13th Gen Core, Intel is broadly promising an up to 15% uplift in single-threaded performance, which has a bigger bearing on gaming performance; and an up to 41% multi-threaded performance uplift; over the previous-generation, when comparing the top Core i9-13900K with its predecessor, the i9-12900K. Intel also claims to have outclassed the AMD Ryzen 9 5950X in multi-threaded performance, and the Ryzen 7 5800X3D in gaming performance.

Intel's performance claims are backed by some impressive hardware changes despite the company sticking with the same Intel 7 (10 nm Enhanced SuperFin) foundry node as "Alder Lake." To begin with, the single-thread performance uplift comes from the new "Raptor Cove" performance-core, which promises an IPC uplift over the previous-generation "Golden Cove," comes with more dedicated L2 cache of 2 MB per core (compared to 1.25 MB per core in the previous-generation); and significantly higher clock-speeds, going all the way up to 5.80 GHz. "Raptor Lake" has up to 8 P-cores, but the company has put in a lot of work in improving the contribution of E-cores to the processor's overall multi-threaded performance uplift. This is achieved by doubling the E-core count to 16. These are the same "Gracemont" E-cores as previous-generation, but Intel has doubled the L2 cache that's shared in a 4-core Gracemont cluster, from 2 MB per cluster to 4 MB. There are upgrades to even the hardware prefetchers of these cores.

NVIDIA and CD Projekt RED (CDPR) came up with a unique giveaway for three GeForce RTX 4090 Founders Edition graphics cards that have been customized with an ultra-rare Cyberpunk 2077-themed backplate. Even if not for the backplate, just the fact that you could win an RTX 4090 should get you to "dust off" Cyberpunk 2077 from your Steam library, reinstall it with its latest update that includes a crazy graphics present aptly named "overdrive," and explore every nook and corner of Night City for clues leading up to the card. These cards are hidden away as premium gear, and everyone who finds the card in the game stands a chance to win one IRL. For more information, visit this page.

A GIGABYTE GeForce RTX 4090 Gaming OC "Ada" graphics card allegedly went on sale way ahead of the October 12 retail availability date. The person who bought it, "Likhg," posted unboxing pictures of the card, revealing a gigantic 4-slot design, a triple-fan cooling solution, with flow-through for the third fan; and accessories that include a reinforcement brace with some screws, an NVIDIA adapter that converts four 8-pin PCIe power connectors into an ATX 12VHPWR connector, and well, that's about it. Graphics card manufacturers have mostly weaned away from including a driver DVD in the box, and point you to download the latest compatible drivers from the NVIDIA website, which means this card is a paperweight until we're closer to the retail date. We're hearing that whoever bought this card managed to resell it for a hefty profit.

Here's our first look at the "AD103" and "AD104" chips powering the GeForce RTX 4080 16 GB and RTX 4080 12 GB, respectively, thanks to Ryan Smith from Anandtech. These are the second- and third-largest implementations of the GeForce "Ada" graphics architecture, with the "AD102" powering the RTX 4090 being the largest. Both chips are built on the same TSMC 4N (4 nm EUV) silicon fabrication process as the AD102, but are significantly distant from it in specifications. For example, the AD102 has a staggering 80 percent more number-crunching machinery than the AD103, and a 50 percent wider memory interface. The sheer numbers at play here, enable NVIDIA to carve out dozens of SKUs based on the three chips alone, before we're shown the mid-range "AD106" in the future.

The AD103 die measures 378.6 mm², significantly smaller than the 608 mm² of the AD102, and it reflects in a much lower transistor count of 45.9 billion. The chip physically features 80 streaming multiprocessors (SM), which work out to 10,240 CUDA cores, 320 Tensor cores, 80 RT cores, and 320 TMUs. The chip is endowed with a healthy ROP count of 112, and has a 256-bit wide GDDR6X memory interface. The AD104 is smaller still, with a die-size of 294.5 mm², a transistor count of 35.8 billion, 60 SM, 7,680 CUDA cores, 240 Tensor cores, 60 RT cores, 240 TMUs, and 80 ROPs. Ryan Smith says that the RTX 4080 12 GB maxes out the AD104, which means its memory interface is physically just 192-bit wide.

NVIDIA GeForce RTX 4090 is turning out to be a cool operator, with the GPU reportedly boosting up to 2.8 GHz (2810 to 2850 MHz) at stock settings, when playing Cyberpunk 2077 at 1440p, in its "psycho" settings preset. with DLSS and Reflex disabled. At native resolution, the RTX 4090 scores 59 FPS (49 FPS at 1% lows), with a frame-time of 72 to 75 ms. With 100% GPU utilization, the card barely breaks a sweat, with GPU temperatures reported in the region of 50 to 55 °C. With DLSS 3 enabled, the game nearly doubles in frame-rate, to 119 FPS (1% lows), and an average latency of 53 ms. This is a net 2X gain in frame-rate with latency reduced by a third. The power-draw is also said to be significantly reduced. The card pulls up to 461 W when rendering at native-resolution, but this drops down to 348 W with DLSS 3 "quality," a 25% reduction.

PSUs with native 12+4 pin ATX 12VHPWR connectors are few and far between, which means some of the first adopters of the GeForce RTX 4090 "Ada" graphics cards will rely on the adapter cable that converts 8-pin PCIe power connectors into one 12VHPWR connector that plugs into the graphics card. Cards that stick to the baseline specs include adapters that convert three 8-pin PCIe connectors (for 450 W output that matches the RTX 4090 reference specs); whereas some premium overclocked RTX 4090 cards, such as the ZOTAC RTX 4090 AMP Extreme, include adapters that convert as many as four 8-pin PCIe to a 12VHPWR, maxing out its 600 W power delivery capability. The product page of the ZOTAC AMP Extreme has an interesting sentence describing this in-box adapter: "Limited service life with up to 30 connect / disconnects."

Apparently the adapter is only good for up to 30 connect/disconnect cycles safely, before you'll need another one. For most gamers who'll install the card and forget about it for years, this shouldn't be an issue. However, for overclockers and enthusiasts using the card on an open-air bench and who rely on a lot of moving cards around, this could be an irritant. Tech journalists (reviewers) swap graphics cards out a lot, too, but then they're likely to have several such adapters lying around from multiple samples, or a PSU with a native 12+4 pin connector.