Inno3D, a devoted NVIDIA partner showcased at CES 2025 its own implementations of NVIDIA GeForce RTX 50 Series GPUs powered by NVIDIA Blackwell architecture. Inno3D custom cooling solutions include the impressive iChill Frostbite liquid cooling and the huge iChill air cooling alternative complemented by the X3 and Twin X2 solutions. Before moving on we like to mention that all the graphics cards, even the RTX 5090 are powered by a single 12-pin connector.

Immediately our eyes observed the iChill Frostbite GeForce RTX 5090 dual-slot graphic cards with an impressive liquid cooling system, something enthusiasts looking for a compact yet high-performance system will surely desire. Next to it, even a blind man could not miss the huge iChill X3 RTX 5080, a 3.5-slot monster. The gargantuan proportions are due to the oversized cooling solution that relies on a combination of three fans (with the outer fans rotating in an anticlockwise direction to reduce turbulence and improve airflow), a huge vapor chamber complemented by a heatsink and a custom backplate that aid hot air extraction to protect the PCB from overheating.

Once every two years or so, technology enthusiasts like ourselves have our sights pinned on what the GPU giants have in store for us. That moment is here, with both NVIDIA and AMD unveiling their Blackwell and RDNA 4 products respectively. NVIDIA has also announced its laptop offerings, with the RTX 5080 Laptop attempting to rule the mainstream high-performance segment. Now, barely a day or two after launch, we already have a rough idea of how mobile Blackwell is going to perform.

The leaked Geekbench OpenCL results, which comes courtesy of the Alienware Area-51 laptop, reveals how well the RTX 5080 Laptop GPU performs in a 175-watt configuration. According to the numbers, the RTX 5080 Laptop managed to barely exceed the 190,000-points barrier, putting it miles ahead of its predecessor which managed around 160,000. Interestingly, as the headline notes, the RTX 4090 Laptop was also left behind, which scores around 180,000 points on average, although systems with beefier cooling setups can post higher numbers.

GIGABYTE, the world's leading computer brand, announced the launch of NVIDIA GeForce RTX 50 Series graphics cards powered by NVIDIA Blackwell and AI, including the GeForce RTX 5090, RTX 5090 D, RTX 5080, RTX 5070 Ti, and RTX 5070, at CES. The latest GIGABYTE graphics card models leverage cutting-edge cooling solutions designed for NVIDIA GeForce RTX 50 Series GPUs to enhance performance in demanding games. Accomplished with the product design concept "Evolution of Ten", GIGABYTE's GeForce RTX 50 Series graphics cards improve thermal performance by up to 10% and reduce overall card volume by up to 10% for a more compatible PC-building experience.

Powered by NVIDIA Blackwell, GeForce RTX 50 Series GPUs bring game-changing capabilities to gamers and creators. Equipped with a massive level of AI horsepower, the RTX 50 Series enables new experiences and next-level graphics fidelity. Multiply performance with NVIDIA DLSS 4, generate images at unprecedented speed, and unleash creativity with NVIDIA Studio. Plus, access NVIDIA NIM microservices - state-of-the-art AI models that let enthusiasts and developers build AI assistants, agents, and workflows with peak performance on NIM-ready systems.

MSI introduced its groundbreaking NVIDIA GeForce RTX 50 Series graphics card range earlier today—hot on the heels of NVIDIA's formal unveiling of its brand-new Blackwell architecture. TechPowerUp staffers on the ground at this year's CES managed to wade through MSI's enviable collection of custom cards showcased on the showroom floor—taking up-close photos and discussing the latest designs with company representatives. An entry-level family has debuted—in the form of INSPIRE—alongside a new high-end series dubbed "VANGUARD." In addition, TPU was treated to the usual suspects including top-of-the-range SUPRIM cards. MSI has also tinkered with its design aesthetic for the VENTUS GAMING product line.

We will start with the upper echelon—MSI's new GeForce RTX 5090 32G SUPRIM LIQUID SOC model. Well-heeled gaming enthusiasts will likely be handing over a hefty premium for this well appointed dual-slot card. Its stealthy "diamond-cut inspired" premium shroud houses a micro-fin copper base and MSI's patented water block array—ensuring that the GPU and VRAM are kept cool during the most demanding of gaming or professional sessions. The 360 mm aluminium radiator is coupled with the latest STORMFORCE fans in a triple unit configuration. It should be noted that the Ada Lovelace generation MSI RTX 4090 SUPRIM LIQUID model sported a dual-fan radiator setup. MSI reps did not divulge any details regarding extreme overclocking and boost figures for this top-tier Blackwell model, but we suspect that the extra fan will be doings its best to temper Team Green's "massive" GB202 die.

Alphacool International GmbH, based in Germany and a leading manufacturer in the field of water cooling, proudly presents the new graphics card water coolers for the latest NVIDIA Geforce RTX 5080 and 5090 generation.

The Alphacool Core GPU water cooler has been completely redesigned, and its appearance has also been adjusted. To further optimize cooling, all distances between the cooler and the components of the PCB were redefined and tested. The entire water flow was simulated using the latest software and then practically verified.

GIGABYTE revealed its GeForce RTX 50 Series graphics cards and new laptops at CES 2025 in Las Vegas. The product line features NVIDIA's Blackwell architecture and includes several GPU models for different market segments. The company's main release, the AORUS GeForce RTX 5090 Master 32G, uses the WINDFORCE cooling system with a new Hawk fan design. The card incorporates server-grade thermal materials and includes an LCD display on its edge. Other models in the lineup include the RTX 5080 Master 16G and RTX 5070 Master 12G, both featuring dual BIOS options for switching between performance and quiet operation. For smaller PC builds, GIGABYTE introduced the RTX 5070 EAGLE OC ICE SFF 12G. The company also showed a water-cooled variant, the AORUS RTX 5090 XTREME WATERFORCE WB 32G, which includes leak detection features.More pictures follow

PNY announced today the arrival of the GeForce RTX 5090, RTX 5080, RTX 5070 Ti, and RTX 5070 graphics cards to its lineup of NVIDIA GeForce RTX GPUs. Powered by NVIDIA Blackwell architecture, GeForce RTX 50 Series GPUs bring game-changing capabilities to gamers and creators. Equipped with a massive level of AI horsepower, the RTX 50 Series enables new experiences and next-level graphics fidelity. Multiply performance with NVIDIA DLSS 4, generate images at unprecedented speed, and unleash creativity with NVIDIA Studio. Plus, access NVIDIA NIM microservices - state-of-the-art AI models that let enthusiasts and developers build AI assistants, agents, and workflows with peak performance on NIM-ready systems.

The new GeForce RTX 50 Series GPUs are designed to provide an exceptional AI experience on PCs. Specialized AI Tensor Cores deliver a massive leap in AI TOPS, enabling transformative capabilities for AI in gaming, creative tasks, and daily productivity. Gamers demand the highest visual quality, and the AI-enhanced NVIDIA DLSS 4 technology combined with full ray tracing creates breathtaking in-game worlds.

MSI has unveiled its groundbreaking NVIDIA GeForce RTX 50 Series graphics cards, featuring cutting-edge designs including Suprim Liquid, Suprim, Vanguard, Gaming Trio, Ventus, and Inspire. Engineered with enhanced thermal solutions, the cards are crafted to meet the high-performance demands of next-gen GPUs, delivering advanced cooling and peak performance.

Powered by NVIDIA Blackwell, GeForce RTX 50 Series GPUs bring game-changing capabilities to gamers and creators. Equipped with a massive level of AI horsepower, the RTX 50 Series enables new experiences and next-level graphics fidelity. Multiply performance with NVIDIA DLSS 4, generate images at unprecedented speed, and unleash creativity with NVIDIA Studio. Plus, access NVIDIA NIM microservices - state-of-the-art AI models that let enthusiasts and developers build AI assistants, agents, and workflows with peak performance on NIM-ready systems.

As a leading innovator in graphics card technology, Palit Microsystems Ltd. today proudly announces the launch of the new NVIDIA GeForce RTX 50 Series graphics cards, featuring the RTX 5090, RTX 5080, RTX 5070 Ti, and RTX 5070 GPUs.

Powered by NVIDIA Blackwell, GeForce RTX 50 Series GPUs bring game-changing capabilities to gamers and creators. Equipped with a massive level of AI horsepower, the RTX 50 Series enables new experiences and next-level graphics fidelity. Multiply performance with NVIDIA DLSS 4, generate images at unprecedented speed, and unleash creativity with NVIDIA Studio. Plus, access NVIDIA NIM microservices - state-of-the-art AI models that let enthusiasts and developers build AI assistants, agents, and workflows with peak performance on NIM-ready systems.

ASUS today announced its NVIDIA GeForce RTX 50 Series graphics cards, including the all-new ROG Astral line that ushers in a new generation of performance. The ROG Astral series combines top-tier frame rates and a sleek, stellar design, with both air-cooled and liquid-cooled variants ready to power a user's next PC.

Powered by NVIDIA Blackwell, GeForce RTX 50 Series GPUs bring game-changing capabilities to gamers and creators, Equipped with a massive level of AI horsepower, the RTX 50 Series enables new experiences and next-level graphics fidelity. Users can multiply performance with NVIDIA DLSS 4, generate images at unprecedented speed, and unleash creativity with NVIDIA Studio. Plus, they can access NVIDIA NIM microservices—state-of-the-art AI models that let enthusiasts and developers build AI assistants, agents, and workflows with peak performance on NIM-ready systems.

NVIDIA today unveiled the most advanced consumer GPUs for gamers, creators and developers—the GeForce RTX 50 Series Desktop and Laptop GPUs. Powered by the NVIDIA Blackwell architecture, fifth-generation Tensor Cores and fourth-generation RT Cores, the GeForce RTX 50 Series delivers breakthroughs in AI-driven rendering, including neural shaders, digital human technologies, geometry and lighting.

"Blackwell, the engine of AI, has arrived for PC gamers, developers and creatives," said Jensen Huang, founder and CEO of NVIDIA. "Fusing AI-driven neural rendering and ray tracing, Blackwell is the most significant computer graphics innovation since we introduced programmable shading 25 years ago." The GeForce RTX 5090 GPU—the fastest GeForce RTX GPU to date—features 92 billion transistors, providing over 3,352 trillion AI operations per second (TOPS) of computing power. Blackwell architecture innovations and DLSS 4 mean the GeForce RTX 5090 GPU outperforms the GeForce RTX 4090 GPU by up to 2x.

According to two of the most accurate leakers, kopite7kimi and hongxing2020, NVIDIA's GeForce RTX 5090 and RTX 5080 will feature 575 W and 360 W TDP, respectively. Previously, rumors have pointed out that these GPU SKUs carry 600 W and 400 W TGPs, which translates into total graphics power, meaning that an entire GPU with its RAM and everything else draws a certain amount of power. However, TDP (thermal design power) is a more specific value attributed to the GPU die or the specific SKU in question. According to the latest leaks, 575 Watts are dedicated to the GB202-300-A1 GPU die in the GeForce RTX 5090, while 25 Watts are for GDDR7 memory and other components on the PCB.

For the RTX 5080, the GB203-400-A1 chip is supposedly drawing 360 Watts of power alone, while 40 Watts are set aside for GDDR7 memory and other components in the PC. The lower-end RTX 5080 uses more power than the RTX 5090 because its GDDR7 memory modules reportedly run at 30 Gbps, while the RTX 5090 uses GDDR7 memory modules with 28 Gbps speeds. Indeed, the RTX 5090 uses more modules or higher capacity modules, but the first-generation GDDR7 memory could require more power to reach the 30 Gbps threshold. Hence, more power is set aside for that. In future GDDR7 iterations, more speed could be easily achieved without much more power.

NVIDIA uploaded a new video on its GeForce YouTube channel called "GeForce LAN 50 Missions and Rewards" that not only features the objective of its upcoming LAN party but also a teaser of one of the GeForce RTX 50 series "Blackwell" GPUs. Pictured below, NVIDIA purposely left it extremely dark to hide as much information as possible. However, we uncovered what was hiding in the dark PC enclosure thanks to some image processing. From the picture alone, we cannot determine what possible SKU this would represent. Potential candidates include the GeForce RTX 5090 or 5080, which are rumored to appear first at the CES launch event, with NVIDIA's CEO Jensen Huang taking the stage and unveiling details about the new GPU generation.

In regards to the design itself, the GPU features a two-fan configuration, meaning that the founder's edition has undergone a redesign from the traditional push/pull cooler design. If this is only a render, time will tell. However, we look forward to seeing more leaks, details, and hints just like NVIDIA left.

Here are some of the first pictures of the MSI GeForce RTX 5080 Gaming Trio "Blackwell" graphics card. The box art provides the first look at what the actual product will look like. It remains over 2 slots thick, and is fairly long; with a trio of what look like 100 mm fans. As with all past generation Gaming X series graphics cards from MSI, this product will feature extensive RGB LED lighting, and a fairly generous factory overclock among its base and premium tiers (Gaming and Gaming X); although historically, MSI's Suprim series has had the company's highest factory tuning.

The back-side of the box lists out MSI-specific and NVIDIA "Blackwell" specific features. The RTX 50-series will see MSI debut the new StormForce fan, which appears to have greater impeller depth than the previous-generation TorX 3.0 fan. The impellers are webbed for axial airflow, the fan blades have ridges. The box also says that the cooler features a nickel-plated copper base-plate, which makes contact with the GPU, which probably has mirror finish. This baseplate will be backed by Core Pipes, which are the cooler's copper heatpipes made square near the baseplate for better contact with it. Lastly, the card comes with a metal backplate—no surprises there. NVIDIA is looking to debut the GeForce RTX 5080 on January 21.

-- images removed at request of MSI --

Hong Kong tech media HKEPC report indicates that NVIDIA's GeForce RTX 5080 graphics card will launch on January 21, 2025. The release follows a planned announcement event on January 6, where CEO Jensen Huang will present the new "Blackwell" architecture. Anticipated specifications based on prior rumors point to RTX 5080 using GB203-400-A1 chip, containing 10,752 CUDA cores across 84 SM. The card maintains 16 GB of memory but upgrades to GDDR7 technology running at 30 Gbps, while other cards in the series are expected to use 28 Gbps memory. The graphics card is manufactured using TSMC's 4NP 4 nm node. This improvement in manufacturing technology, combined with architectural changes, accounts for most of the expected performance gains, as the raw CUDA core count only increased by 10% over the RTX 4080. NVIDIA is also introducing larger segmentation between its Blackwell SKUs, as the RTX 5090 has nearly double CUDA cores and double GDDR7 memory capacity.

NVIDIA is organizing a GeForce LAN event two days before the announcement, marking the return of this gathering after 13 years, so the timing is interesting. NVIDIA wants to capture gamer's hearts with 50 hours of non-stop gameplay. Meanwhile, AMD currently has no competing products announced in the high-end graphics segment, leaving NVIDIA without direct competition in this performance tier. This market situation could affect the final pricing of the RTX 5080, which will be revealed during the January keynote. While the January 21 date appears set for the RTX 5080, launch dates for other cards in the Blackwell family, including the RTX 5090 and RTX 5070 series, remain unconfirmed. NVIDIA typically releases different models in their GPU families on separate dates to manage production and distribution effectively.

By now, it's a surprise to almost nobody that NVIDIA plans to launch its next-generation RTX 5000-series "Blackwell" gaming graphics cards at the upcoming CES 2025 event in Las Vegas in early January. Previously, leaks and rumors gave us a full run-down of expected VRAM amounts and other specifications and features for the new GPUs, but these have yet to be confirmed by NVIDIA—for obvious reasons. Now, though, it looks as though iBuyPower has jumped the gun and prematurely revealed the new specifications for its updated line-up of pre-built gaming PCs with RTX 5000-series GPUs ahead of NVIDIA's official announcement. The offending product pages have since been removed, but they both give us confirmation of the previously leaked VRAM amounts and of the expected release cadence for RTX 5000, which will reportedly see the RTX 5070 Ti and RTX 5080 launch before the RTX 5090 flagship.

On iBuyPower's now-pulled pages, the NVIDIA GeForce RTX 5080 16 GB and GeForce RTX 5070 Ti 16 GB can be seen as the GPUs powering two different upcoming Y40 pre-built gaming PCs from the system integrator. The VRAM specifications here coincide with what we have previously seen from other leaked sources. Unfortunately, while an archived version of the page for the pre-built containing the RTX 5080 appears to show the design for an ASUS TUF Gaming RTX 5080 with a triple-fan cooler, it looks like iBuyPower is using the same renders for both the 5080 and 5070Ti versions of the pre-built PCs. What's also interesting is that iBuyPower looks to be pairing the next-gen GPUs with 7000-series AMD X3D CPUs, as opposed to the newly released AMD Ryzen 9000 X3D chips that have started making their way out into the market.

What we've all been waiting for, might just have appeared and what we're talking about is of course the pricing of NVIDIA's upcoming graphics cards. @wxnod has posted a single screenshot on X/Twitter of what could be the MSRP of the RTX 5090 and RTX 5080 in China. The MSRP of the RTX 4080 was 9,499 RMB and the RTX 5080 appears to be not that much higher, at 9,999 RMB, but this still equates to about US$1,370, although do note that there's 13 percent sales tax/VAT in China.

Now as for the RTX 5090, things won't be as rosy. The RTX 4090 had an MSRP of 12,999 RMB in China and the RTX 5090 comes in at an insane 18,999 RMB or US$2,600. That's a price hike of a not insignificant 46 percent over the RTX 4090 and this might make it the most expensive consumer graphics card ever released. We'd suggest taking these prices with a helping of NaCl just to be on the safe side. The cards are expected to be available some time in January according to the screenshot.

Update 15:34 UTC: A second picture was posted in the same thread on X/Twitter that shows the expected launch months of the lower-tier RTX 5000-series cards as well and it appears to be taken from a video.

The box-art of upcoming GALAX GeForce RTX 5080 "Blackwell" graphics card has much to say about the graphics architecture, without dropping any explicit mentions. The front-face of the box features a hooded human face—nothing fancy, until you begin to pay attention to the details. Half the face is composed of triangles streaming toward the face, while the other half is composed in place by a blue stream of light, as if to denote that it's being drawn by a fundamentally different method than "triangles."

The triangles here represent classic raster 3D graphics, while the other spirit-like half denotes neural rendering. Here's where it gets interesting. Both kinds of rendering are being applied to the same frame, and so neural rendering is fundamentally different from DLSS 3 Frame Generation, a technology that draws alternate frames using optical flow, motion vectors, and AI. Neural rendering appears to be, at least from this GALAX box-art, a technology that runs in real-time, where some elements, portions, or details of a frame are rendered by a generative AI, and others by raster 3D graphics.

Last week, we reported on a massive leak that gave us a glimpse at a surfeit of upcoming ROG gaming laptops from ASUS boasting Arrow Lake-HX and Strix Halo processors, along with RTX 50-series Laptop GPUs. The company has now gone ahead and officially teased its ROG Flow Z13 laptop and the ROG Strix 18 laptop, the videos for which were obtained by VideoCardz. Considering that these were just teasers, the details shared were sparse, but undoubtedly intriguing regardless.

The ROG Strix 18, will likely boast the Core Ultra 9 285HX and Core Ultra 9 275HX CPUs along with NVIDIA "Blackwell" Laptop GPUs, with up to an 175-watt RTX 5090 Laptop with 16 GB of GDDR7 VRAM. Multiple other GPU options will also be available, with the RTX 5070 Ti Laptop GPU being the entry-point as per the leaked listings. Clearly, the Strix 18 is poised to be a powerful and power-hungry desktop replacement-class laptop, along with a massive 18-inch mini LED screen with a speedy 240 Hz refresh rate. Needless to say, the G18 has no intention of being affordable.

Moore's Law is Dead, a prominent YouTube channel specializing in computer hardware leaks, has revealed its expectations for the RTX 50-series Laptop GPUs. We have already reported on a massive product listing leak shedding light on almost every single "Blackwell" laptop GPU, but needless to say, more information is always welcome. According to Moore's Law is Dead, the RTX 5090 Laptop GPU, and this has been hinted at by the aforementioned prior leak, will only sport 16 GB of GDDR7 VRAM - the same as the RTX 5080 Laptop.

Moreover, his sources indicate that the RTX 5080 will drop with a 175-watt TGP and 7,680 CUDA cores, which is shockingly only a hair more than the 7,424 found in its predecessor. However, the source did state that the RTX 5080 will be around 40 to 60% faster than the RTX 4080, which is a massive generational leap in performance. It is not clear at this point how this number was arrived at, but it sure does seem rather utopian. Yet another source has also indicated that an RTX 5090 Laptop card with a whopping 24 GB of VRAM is also in the works which might launch down the line, but there is little else to be said about it. As MLID notes, NVIDIA has very little to no competition in the high-end laptop segment, which inevitably makes things worse for the end-user.

It is barely a surprise that the desktop RTX 50X0-series cards will be accompanied by laptop variants soon after their launch. And as such, multiple listings of upcoming gaming laptops have appeared on the internet. The information is rather intriguing, to say the least. The leaked listings indicate that both AMD and Intel will have products ready to join forces with Blackwell, with Intel's Arrow Lake-HX and, rather shockingly, Ryzen 7000HX-series from Team Red - much to the chagrin of those expecting Ryzen 9000HX to drop sooner. The listings also reveal some other specifications, such as the display, but there is nothing notable there - it's the typical spread of mini LED and OLED panels.

Among the leaked laptops, we have the ROG Strix Scar 18, which will pack up to a Core Ultra 9 285HX "Arrow Lake-HX" CPU, along with either an RTX 5080, or an RTX 5090 GPU. Disappointingly, it appears that the RTX 5090 will be limited to only 16 GB of VRAM, which is half that of its desktop counterpart. The RTX 5080 will also be available with 16 GB of VRAM, which certainly makes us sigh in relief. A lower-tier laptop, namely the Strix 16, has also been leaked, packing either an RTX 5070 Ti, or an RTX 5060 GPU with 12 GB and 8 GB of VRAM respectively. The laptops further down will also have the entry-level RTX 5050, with 8 GB of VRAM as well. Of course, all of this is on the new GDDR7 spec. The full-resolution image of the leaked listings can be found here.

Acer has jumped the gun and listed its ACER Predator Orion 7000 systems with the upcoming NVIDIA RTX 50 series graphics cards, namely the GeForce RTX 5080 and the GeForce RTX 5090. In addition, the listing confirms that the GeForce RTX 5080 will come with 16 GB of GDDR7 memory, while the GeForce RTX 5090 will get 32 GB of GDDR7 memory.

The ACER Predator Orion 7000 gaming PC was announced back in September, together with Intel's Core Ultra 200 series, and it does not come as a surprise that this high-end pre-built system will now be getting NVIDIA's new GeForce RTX 50 series graphics cards. In case you missed previous rumors, the GeForce RTX 5080 is expected to use the GB203-400 GPU with 10,752 CUDA cores, and come with 16 GB of GDDR7 memory on a 256-bit memory interface. The GeForce RTX 5090, on the other hand, gets the GB202-300 GPU with 21,760 CUDA cores and packs 32 GB of GDDR7 memory.

NVIDIA is preparing to unveil its "Blackwell" GeForce RTX 5080 graphics card, featuring cutting-edge GDDR7 memory technology. However, RTX 5080 is expected to be equipped with 16 GB of GDDR7 memory running at an impressive 30 Gbps. Combined with a 256-bit memory bus, this configuration will deliver approximately 960 GB/s bandwidth—a 34% improvement over its predecessor, the RTX 4080, which operates at 716.8 GB/s. While the RTX 5080 will stand as the sole card in the lineup featuring 30 Gbps memory modules, while other models in the RTX 50 series will incorporate slightly slower 28 Gbps variants. This strategic differentiation is possibly due to the massive CUDA cores gap between the rumored RTX 5080 and RTX 5090.

The flagship RTX 5090 is set to push boundaries even further, implementing a wider 512-bit memory bus that could potentially achieve bandwidth exceeding 1.7 TB/s. NVIDIA appears to be reserving larger memory configurations of 16 GB+ exclusively for this top-tier model, at least until higher-capacity GDDR7 modules become available in the market. Despite these impressive specifications, the RTX 5080's bandwidth still falls approximately 5% short of the current RTX 4090, which benefits from a physically wider bus configuration. This performance gap between the 5080 and the anticipated 5090 suggests NVIDIA is maintaining a clear hierarchy within its product stack, and we have to wait for the final launch to conclude what, how, and why of the Blackwell gaming GPUs.

Recent leaks have unveiled specifications for NVIDIA's upcoming RTX 5070 Ti graphics card, suggesting an increase in power consumption. According to industry leaker Kopite7kimi, the RTX 5070 Ti will feature 8,960 CUDA cores and operate at a 300 W TDP. In a departure from previous generations, the RTX 5070 Ti will reportedly share the same GB203 die with its higher-tier sibling, the RTX 5080. This architectural decision differs from the RTX 40-series lineup, where the 4070 Ti and 4080 utilized different dies (AD104 and AD103, respectively). This shared die approach could potentially keep NVIDIA's manufacturing costs lower. Performance-wise, the RTX 5070 Ti shows promising improvements over its predecessor. The leaked specifications indicate a 16% increase in CUDA cores compared to the RTX 4070 Ti, though this advantage shrinks to 6% when measured against the RTX 4070 Ti Super.

Power consumption sees a modest 5% increase to 300 W, suggesting improved efficiency despite the enhanced capabilities. Memory configurations remain unconfirmed, but speculations about the card indicate that it could feature 16 GB of memory on a 256-bit interface, distinguishing it from the RTX 5080's rumored 24 GB configuration. The positioning across the 50-series GPU stack of this RTX 5070 Ti appears carefully calculated, with its 8,960 CUDA cores sitting approximately 20% below the RTX 5080's 10,752 cores. This larger performance gap between tiers contrasts with the previous generation's approach, potentially indicating a more defined product hierarchy in the Blackwell lineup. NVIDIA is expected to unveil its Blackwell gaming graphics cards at CES 2025, with the RTX 5090, 5080, and 5070 series leading the announcement.

Q1-2025 promises to be an action-packed quarter for graphics cards, with NVIDIA introducing the bulk of its next-generation GeForce RTX 50-series "Blackwell" GPUs. The company is expected to start things off with the two enthusiast-segment SKUs, the RTX 5090 and RTX 5080, in January, followed by the RTX 5070-series in February, and rounded off nicely with the RTX 5060-series in March. This would mean hundreds of individual new graphics card SKUs from NVIDIA's board partners, which are reportedly busy winding up the final inventory deliveries of their RTX 40-series "Ada" products, and transferring this production capacity to the RTX 50-series. So, when the RTX 50-series GPU models do come out across the quarter, there's plenty of inventory to go around. Board Channels reports that on NVIDIA's end, production of nearly every AD100-series silicon has ended, except the AD107, which will continue selling for entry-mainstream GeForce RTX 40-series SKUs. The AD106 production line has stopped, as has the AD103, AD104, and AD102.