Graphics Cards - Page 128

ZOTAC has just revealed its new GeForce RTX 3090 Ti PGF OC graphics card, a new China-exclusive absolute behemoth of a GPU... clocking in at a huge 4-slot card.

The new ZOTAC GeForce RTX 3090 Ti PGF OC graphics card has the same GPU specs as the ZOTAC RTX 3090 Ti AMP Extreme, with a GPU boost clock of up to 1890MHz and default TDP of 450W. But, the AMP Extreme can enjoy a modified 495W of power with overclocking... we should expect the same with the RTX 3090 Ti PGF OC model.

But man... is the new ZOTAC RTX 3090 Ti PGF OC a big boi, with a huge 4-slot design that towers over the 3.5-slot design of the ZOTAC RTX 3090 Ti AMP Extreme GPU. The 4-slot design makes it the biggest custom GeForce RTX 3090 Ti on the planet so far... and they're all chunky beasts as it is.

NVIDIA is cozying up very tightly with TSMC leading into its next-gen Ada Lovelace GPU launch later this year, which will power the new GeForce RTX 40 series graphics cards... with new reports teasing TSMC will be the sole supplier of NVIDIA's GPUs in 2022.

We all know the issues that NVIDIA has been facing with its current Ampere GPU architecture, which is being fabbed by Samsung on their custom 8nm process. The new GA102 powering the GeForce RTX 3090 Ti is chewing upwards of 500W, with some custom models offering up to 1275W of power. This is all a big test for the next-gen GPUs, which are looking at using 600W+ of power.

The inefficient Samsung 8nm node can't compete with TSMC 7nm let alone TSMC's new 5N process node, which should see NVIDIA's new Ada Lovelace GPU call home in 2022 and beyond. NVIDIA isn't a stranger to TSMC, it's just been making most of its GeForce GPUs with South Korean rival Samsung, and it hasn't gone so well -- so the big switch back to TSMC is happening -- and a lot of money is exchanging hands, I'm sure.

AMD was gearing up for its big RDNA 2 refresh on April 20, but that has been delayed according to the latest rumors -- with a new RDNA 2 refresh launch of May 10 -- just over a month from now.

We can expect AMD to unleash a trio of new all-black Radeon RX 6000 series refresh GPUs, led by the new flagship Radeon RX 6950 XT graphics card. After that, we'll have the Radeon RX 6750 XT and the lower-end Radeon RX 6650 XT graphics cards.

AMD will be splashing out with faster GDDR6 memory clocked at 18Gbps, up from the 16Gbps GDDR6 on the Radeon RX 6000 series GPUs. The new flagship Radeon RX 6950 XT rocks a huge design just like the current Radeon RX 6900 XT has, but we should expect a big chunk of performance on top of the RX 6900 XT (like the new RTX 3090 Ti offers over the RTX 3090).

Intel's new desktop Arc Alchemist GPUs are only a few months away, and thanks to their recent presentation video of their Arc Control panel, desktop GPU details appeared.

Intel is rolling out a new GPU monitoring suite for its Arc Alchemist and Xe Iris desktop GPUs called Arc Control, with the recent presentation teaing the new "Performance Tab" for GPU monitoring and basic overclocking (OC) tools. In the quick sneak peak, Intel had a live performance monitoring panel showing the GPU power, GPU clock, and VRAM clock of the GPU inside of the PC.

We don't know which GPU that Intel was using in the machine, but it would've been an Arc Alchemist desktop GPU of some kind -- with a GPU boost clock of 2250MHz, GPU power (not the entire card, but the GPU only) at 175W and the GDDR6 memory clocks at 1093MHz. None of the new Arc Alchemist mobile GPUs have a power draw of 175W, with the Arc A770M maxing out at 150W. But remember, the 175W is GPU power draw alone... not the entire card... meaning we're looking at 200-225W+ full power for the whole card.

NVIDIA has only just launched its new flagship GeForce RTX 3090 Ti, with MSI unleashing its new GeForce RTX 3090 Ti SUPRIM X -- my review on the MSI RTX 3090 Ti SUPRIM X here -- and now MSI has announced its new 3dfx Voodoo5 7000 series GPU.

The new MSI Voodoo5 7000 series GPU is an "Extreme 2D/3D accelerator" that packs a huge 48GB of GDDR7 memory, something that Samsung only teased late last year. GDDR7 memory will reportedly arrive at a blistering 32Gbps bandwidth, which is up from the already ultra-fast 21Gbps on the latest GDDR6X memory on the GeForce RTX 3090 Ti.

MSI's new Voodoo5 7000 series GPU with 48GB of GDDR7 memory underneath still rocks an old-school DVI connector, so that you can run 8K 120FPS on Ultra details with reverse 2dfx technology -- MSI and 3dfx's take on AI upscaling... turning 3D graphics to 2D graphics. You need to buy the monster Voodoo5 7000 series GPU with 48GB of GDDR7 to render 8K 120FPS, only so you can play it at 1024 x 768 @ 30FPS.

I'm sure Tom from Moore's Law is Dead will want to have the world exclusive on Moore Threads' new MTT S60 and S2000 GPUs.

Moore Threads announced the new GPUs today, on the same day as Intel detailing its new Arc A-Series Mobile GPUs that are rolling out into laptops over the coming months. The new MTT S60 and MTS2000 desktop GPUs from Moore Threads are made on the 12nm process node, using the MUSA (MT Unified System Architecture).

MUSA is the first GPU architecture out of China, and get this -- MUSA supports the DirectX Runtime -- something that's missing from every other Chinese GPU. MUSA not only supports DirectX, but also OpenCL, OpenGL, Vulkan -- and even NVIDIA's CUDA.

Intel announced its new Arc Alchemist Mobile GPUs today, with a duo of Arc GPUs now called the ACM-G10 and ACM-G11... with some more details on the ACM-G10 GPU now here.

The new Intel ACM-G10 GPU used to be called the DG2-512EU, with the chip rocking 21.7 billion transistors and measures in at 406mm2 in size. This means Intel's new ACM-G10 is larger not just NVIDIA's mid-range GA104 GPU but also AMD's new mid-range Navi 22 GPU.

NVIDIA's mid-range GA102 GPU has 17.4 billion transistors and 392mm2 die size, while AMD's mid-range Navi 22 GPU has 17.2 billion transistors and a size size of 336mm2. Intel is using the TSMC 6N process node for its new Arc GPUs, while AMD taps TSMC N7 and NVIDIA with Samsung 8N.

NVIDIA's new flagship GeForce RTX 3090 Ti is now official, with liquid cooling leader EK announcing a bunch of new Quantum Vector water blocks for the new RTX 3090 Ti GPU.

The new GeForce RTX 3090 Ti might have the same 24GB of GDDR6X memory, but they're Micron's latest 2Gbit modules which means there's no double-sided GDDR6X here, thus no need for active backplates. NVIDIA's first GeForce RTX 3090 had double-sided GDDR6X memory and saw temperatures soaring to 110C when stressed.

EK knows that there isn't really a need to have an active backplate for the GeForce RTX 3090 Ti -- as the RTX 3090 Ti has the same 24GB GDDR6X as the RTX 3090, but with new 2GB chips so there's no GDDR6X modules on the back of the card -- enthusiasts don't care, they want active backplates for the RTX 3090 Ti.

Intel has officially launches its new Arc A-Series Mobile GPUs with the lowest-end Arc 3 GPUs launching now, while Arc 5 and Arc 7 launch in the coming months... but what about XeSS?

Intel's new XeSS isn't making the launch, with the AI upscaling technology promised for "early summer" which better lines up with the higher-end desktop Arc Alchemist GPUs launching. AMD just launched its new FidelityFX Super Resolution 2.0 (FSR 2.0) while NVIDIA has DLSS 2.3 out (and should launch DLSS 3.0 in the coming months leading into Ada Lovelace).

There is something Intel did confirm: XeSS will support up to 20 games at launch, including big titles like Death Stranding: Director's Cut, Ghostwire Tokyo, Shadow of the Tomb Raider, and many more. We will have a little while to go yet, but things are beginning to get clearer as Intel rolls out its Arc GPUs to the masses.

Intel has just launched its first Arc A-Series GPUs for laptops, detailing its Arc 3, Arc 5, and Arc 7 families of mobile GPUs that will be powering laptops throughout 2022.

Intel has launched its Arc A-Series Mobile GPUs with the flagship Arc A730M + A770M, with the A730M featuring 24 Xe GPU cores while the A770M rocks 32 Xe GPU cores. The Arc A730M has GPU clocks of up to 1100MHz while the flagship Arc A770M has GPU clocks of up to 1650MHz.

We will see 16GB of GDDR6 memory on the Intel Arc A770M on a 256-bit memory bus and 120-150W TDP, while the Arc A730M has 12GB of GDDR6 memory on a 192-bit memory bus and 80-120W TDP. Under that we have the Arc A550M with 16 Xe GPU cores and up to 900MHz GPU clocks, 8GB of GDDR6 on a 128-bit memory bus and 60-80W TDP.

Intel has just officially teased its Arc A-Series desktop GPU, with the official Intel Arc Alchemist Desktop Limited Edition graphics card teased by the company just now. Check it out:

The company teases its upcoming Intel Arc A-Series Limited Edition GPU, where inside we should see the full-spec ACM-G10 GPU with 32 Xe-Cores and 16GB of GDDR6 memory. For display connectivity we can see that Intel is going with 3 x DisplayPort and 1 x HDMI 2.1 connector -- at least we'd hope so.

We don't know how many PCIe power connectors Intel's new Arc A-Series Limited Edition graphics card will have, but I doubt it'll have one of those nice new 16-pin PCIe power connectors that NVIDIA's new flagship GeForce RTX 3090 Ti has... not that Intel's more mid-range Arc A-Series GPU would need that.

COLORFUL has just unleashed a trio of new custom GeForce RTX 3090 Ti graphics cards, with the new iGame GeForce RTX 3090 Ti Vulcan and iGame GeForce RTX 3090 Ti Neptune both capable of taking up to 510W of power.

The new flagship COLORFUL iGame GeForce RTX 3090 Ti Vulcan rocks an LCD display on top, with a 480 x 128 resolution and can show real-time monitoring of your GPU and GDDR6X temperatures, custom GIFs, and images. You can use the 2-in-1 cable to connect the display to the card's interface, uploading to a customized display.

NVIDIA's tweaked GA102 GPU is pushed up to 1860MHz boost GPU clock, while the "One-Key OC" mode (the physical OC button on the back of the card) will push the GPU boost up to 1920MHz. I'm sure it'll sit somewhere around 2000MHz or so, overclockable up to 2050-2100MHz. But, with all that OC means the power numbers drive up: 480W default, 510W with OC... all over the single 16-pin PCIe power connector (which sprawls out to 3 x 8-pin connectors, like the other custom RTX 3090 Ti cards).

NVIDIA's new GeForce RTX 3090 Ti is now official, with EVGA teasing its new GeForce RTX 3090 Ti KINGPIN Hybrid graphics card... wielding dual 16-pin PCIe power connectors, and an OLED display.

We saw EVGA's exotic GeForce RTX 3090 Ti KINGPIN Hybrid earlier this year when the company teased it in January, but now it's been shown off in more detail now that NVIDIA has released the RTX 3090 Ti. There's no pictures of the AIO cooler, but you can see a huge bulky AIO cooler is going to be with the card... I didn't like it on the ASUS ROG Strix LC GeForce RTX 3090 Ti OC Edition that I reviewed in the last 24 hours.

But what EVGA also does is include a slick OLED panel so that you can keep an eye on those GPU and GDDR6X temperatures, while the new GeForce RTX 3090 Ti KINGPIN Hybrid requires a truly ridiculous dual 16-pin PCIe power connector setup.

NVIDIA's new GeForce RTX 3090 Ti graphics card is only hours away, and so too are the custom AIB versions... including the EVGA and COLORFUL variants of the RTX 3090 Ti we have here.

The new EVGA GeForce RTX 3090 Ti FTW3 graphics card is a THICC boi, with a huge 3.5-slot -- yeah, beyond 3-slot -- design and massive new PCB that is ready for that new 16-pin power connector and 100W more power from the GA102-350 GPU.

I mean, it's a seriously bulky card that is comically large... the huge 3.5-slot design sees EVGA putting the 16-pin PCIe power connector on the end of the card. Man, I love that... the I/O bracket has also been refreshed, but still has the 3 x DisplayPort 1.4 + HDMI 2.1 connectors. All ready for 4K 120FPS and 8K 60FPS displays and TVs of now, and the future.

If you were expecting big upgrades from NVIDIA's new flagship GeForce RTX 3090 Ti, I wouldn't be holding my breath... because you're going to experience another 5-10% more performance.

Even with all of that additional grunt under its hood -- 2GHz+ GPU boost, faster and newer 21Gbps GDDR6X memory, and up to 100W of more power -- the new GeForce RTX 3090 is only 5-10% faster than the GeForce RTX 3090.

The new rumor is coming from CapFrameX, who tweeted out "some performance numbers" from the new RTX 3090 Ti, and the current flagship RTX 3090. He said that his "first impression" was that the RTX 3090 Ti was around 10% faster than the RTX 3090 at 4K, with up to 450W power and up to 2GHz GPU boost.

NVIDIA is now hours away from launching its new flagship GeForce RTX 3090 Ti graphics card, with NVIDIA's in-house GeForce RTX 3090 Ti Founders Edition to become the first graphics card with a nifty new 16-pin PCIe power connector.

NVIDIA's upcoming GeForce RTX 3090 Ti Founders Edition will have a PCIe Gen5 16-pin power connector, replacing the 12-pin Molex MicroFit 3.0 plug that was on the RTX 30 series Founders Edition GPUs. The new 16-pin PCIe power connector can drive up to 450W on the RTX 3090 Ti, while the 12-pin PCIe power connector could drive up to 350W on the RTX 3090.

But on some of the custom GeForce RTX 3090 Ti graphic card design, we can expect up to 480W of power... or up to 516W with some tweaking. The upcoming ASUS ROG Strix LC GeForce RTX 3090 Ti (liquid-cooled with a dual-fan AIO cooler) has up to a 480W TDP, and can be tweaked for up to 516W. At this point, you're reaching the maximum power -- 525W -- of the PCIe 5.0 power connector and PCIe slot.

AMD is gearing up to its next-gen RDNA 3 GPU architecture, and during GDC 2022 we've got game developer Arkake Studios -- the team behind DEATHLOOP -- presenting at GDC 2022 with a talk called "A Guided Tour of Blackreed: Rendering Technologies in Deathloop".

In this chat, Arkane talked about the graphics engine advancements in DEATHLOOP, with the end of it addressing AMD's next-gen FSR 2.0 upscaling technology.AMD Developer Technology Engineer Lou Kramer talked about the comparisons between the edges of the wooden pattern being better-reconstructed in FSR 2.0 (Quality Mode, shifting from 1440p to 4K) that it evended up looking even better than native 4K... with TAA... and sharpening enabled.

But when it came to DEATHLOOP, we saw Kramer talking about the input sources of FSR 2.0 into the game -- including color buffer, depth buffer, and motion vectors -- DEATHLOOP also provides exposure textures for input, but FSR 2.0 can also compute one of those on its own.

AMD's new Radeon Adrenalin 22.3.2 drivers have been released, with support for Ghostwire: Tokyo and Tiny Tina's Wonderlands.

If you've got a Radeon RX 6900 XT, Radeon RX 6800 XT, or Radeon RX 6700 XT graphics card and were wanting to play Tiny Tina's Wonderlands, then you will enjoy up to 23% more performance, up to 21% more performance, and up to 20% more performance at 4K on those GPUs, respectively.

There's the usual bug fixes, with some issues squashed in Cyberpunk 2077 on the Radeon RX 570, issues with Stranger of Paradise: Final Fantasy Origin on the RX 6900 XT, and more. You can read more about that below, and download the new AMD Radeon Adrenalin 22.3.2 drivers right here (539MB download).

NVIDIA just unveiled its next-gen Hopper GPU architecture, AMD just teased its next-gen FSR 2.0 upscaling technology, and now Intel has detailed its upcoming XeSS upscaling tech at the Game Developers Conference (GDC 2022).

Intel had its software engineers discuss its new XeSS technology, saying that upscaling and anti-aliasing should be fixed as a single issue. Both older games and previous-gen GPUs would see performance improvements and quicker implementation with spatial upscalers.

On the other side, games that have ray tracing and other high-end graphics technologies that are only supported with the very latest GPU architectures, will obviously see bigger improvements in performance and IQ through temporal upscalers. As for GPU support, Intel plans to support as many GPUs as it can with XeSS -- supporting all GPUs that support Shader Model 6.4+ and DP4a instructions -- so most GPUs should be fine.

We are only a few days out from NVIDIA's new flagship GeForce RTX 3090 Ti graphics card launching, and with it will be a beefy new PCIe 5.0-ready 16-pin power connector.

But don't worry: you'll get a 16-pin to 3 x 8-pin PCIe power adapter, so you can plug your existing PSU cables into a turbocharged GeForce RTX 3090 Ti. But there aren't any PSUs on the market with high-end 16-pin PCIe power connectors just yet, which is why there will be a power adapter included.

The power adapter will provide 450W of power on its own, with the PCIe interface providing another 75W we have 525W total. This means we should expect custom variants of the GeForce RTX 3090 Ti to have a maximum TDP of 500W, up from the 400W on the custom RTX 3090 cards.