Graphics Cards - Page 198

NVIDIA just unveiled its next-generation Turing GPU architecture during SIGGRAPH 2018, with three new Quadro cards unveiled: Quadro RTX 5000, RTX 6000, and the flagship Quadro RTX 8000.

Well, just a couple of hours ago NVIDIA uploaded a video to their YouTube account dubbed 'Be For The Game'. The video shows off a bunch of high-quality shots of an unknown graphics card - could be a GTX 10 series card, or it could be a next-generation GeForce graphics card based on the just-announced Turing architecture. There are a few teases of '20' and '2080' in the video, with one particular part standing out.

At the end of the video, NVIDIA has edited it perfectly to display the date '20.AUG.2018' except that the '20' and '08' come up last... which could be a tease of '2080'. It could be some sly trick to tease us of the GeForce GTX 2080, or I could be looking into it too much. But for NVIDIA to rearrange those numbers: 2, 0, 8, 0 and have the other numbers still placed there - they're highlighting '2080' for a reason.

SIGGRAPH 2018 - NVIDIA has just kicked off its SIGGRAPH 2018 press conference with CEO Jensen Huang unveiling their new Turing GPU architecture and three new Quadro RTX graphics cards.

The new Turing GPU is the largest leap in GPU technology from NVIDIA since the CUDA GPU was unveiled back in 2006, with a huge 18.6 billion transistors on a 12nm process. There's also a Tensor Core inside with 125 TFLOPs of FP16 performance, and an RT Core with 10 Giga Rays/sec of performance. The Turing GPU in all of its glory uses 754mm2, compared to 471mm2 of Pascal (which had 11.8 billion transistors).

NVIDIA revealed three new Quadro RTX branded cards with the Quadro RTX 5000, Quadro RTX 6000, and Quadro RTX 8000. The flagship Quadro RTX 8000 features 4608 CUDA cores, an insane 48GB of GDDR6 memory and an equally as huge $10,000 price tag. The new Quadro RTX 6000 has its VRAM cut in half to 24GB with the rest of the specs at the same level for $6300 while the RTX 5000 has its CUDA cores cut down to 3072, and its GDDR6 down to 16GB with a price of $2300.

AMD has just announced a new graphics card at SIGGRAPH 2018, with the new Radeon Pro WX 8200 workstation graphics card set to ship to customers in September.

The new Radeon Pro WX 8200 is a cut down version of the flagship Radeon Pro WX 9100 except it's under half the price at $999 compared to $2200 of the WX 9100. How is the WX 8200 half the price of the WX 9100? Well, for starts we have 8GB of HBM2 compared to 16GB of HBM2 on the WX 9100, while the WX 8200 has also had its stream processors sliced down to 3584 compared to 4096 on the WX 9100.

AMD has used the same Vega 10 GPU on the same 14nm node, with 10.8 TFLOPs of single precision compute performance and 21.5 TFLOPs in half precision. Comparing this to the WX 9100 and its 12.3 TFLOPs of single precision compute performance, and 24.6 TFLOPs of half precision.

The weekend has exploded with rumors and fresh information on NVIDIA's next-gen graphics cards, with the weekend kicking off with news that NVIDIA had registered the "Turing" trademark, big enough news on its own but it gets way better.

This is where things get interesting: NVIDIA is reportedly preparing a new TITAN graphics card that is rumored to cost a huge $3000.

If these new rumors are true, NVIDIA is about to make one of the largest changes to GeForce branding in many many years. NVIDIA is reportedly going to use the 'RTX' branding on not only its next-gen TITAN (making it the TITAN RTX) but the new GeForce graphics cards will, if these rumors are true, arrive as the GeForce RTX 2080 and RTX 2070.

NVIDIA edges closer to the unveiling of their next-gen graphics cards and one of the craziest rumors of them all have hit: the company will reportedly use a dual-fan, open-air cooler on their purported GeForce GTX 1180/2080 graphics card. Don't forget to enter our GeForce GTX 1180/2080 giveaway! We're giving away not one, but TWO next-gen cards as soon as they launch.

BenchLife is reporting this rumor and if it turns out to be true, we could be looking at the first Founders Edition card with a dual-fan cooler, but it definitely wouldn't be the first NVIDIA graphics card to don a dual-fan cooler. NVIDIA previously used a dual-fan cooler on the high-end GeForce 7900GX2 back in the day, and have not used a dual-fan on a GeForce graphics card since.

Sure, plenty of AIB partners use dual-fan coolers but they're all custom cards so they can do whatever they want. I really can't see NVIDIA using a dual-fan cooler on a new GeForce GTX 1180/2080 but then again, that's only because the company hasn't used a dual-fan cooler in a very long time. The aesthetics of the GTX 10 series FE cards is stuck in my mind after looking at, and using them everyday for over two years now... a dual-fan cooler just sounds strange.

AMD has released new Radeon drivers that are optimized for new games like Monster Hunter World, We Happy View, Madden NFL 19, and World of Warcraft: Battle for Azeroth. You can download them right here.

The new Radeon Adrenalin 18.8.1 drivers offer up to 5% more performance on the Radeon RX Vega 64 for Monster Hunter World, and for Radeon RX 580 owners you can expect up to 6% additional performance. There are some bug fixes included with Fortnite crashing upon launch in Windows 7 solved, as well as some issues with multi-GPU systems crashing when changing resolutions in-game fixed.

Radeon Software Adrenalin Edition 18.8.1 Release Notes

TweakTown was one of the first in the world to host an NVIDIA GeForce GTX 1080 giveaway, before it launched back in May 2016, and here we are doing it again. It might not be called the GeForce GTX 1180, but we are giving away TWO of whatever next-gen flagship graphics card NVIDIA releases, whenever they comes out. We're celebrating the fact it has been so long between exciting GPU releases that we're not giving away one, but TWO of their next-gen cards! SLI giveaway, baby!

All of the rumors are pointing towards a release sometime this month so we're starting ours now building up to its release, and then hosting livestreams and giveaways on top of this GTX 1180 giveaway once the card has been made official by NVIDIA. We don't know if the flagship new card will be a Turing-based GeForce GTX 1180 or something else such as a GTX 1185/1190/1195 or something, but whatever it is unveiled as TweakTown will give away the best card NVIDIA announces in this world-first competition.

How To Win:

NVIDIA hasn't made anything official about its next-gen graphics cards yet, with our own sources saying it will arrive as the GeForce GTX 11 series but there are Chinese AIB partners that are already unveiling next-gen GeForce graphics cards.

Now we have Maxsun unveiling their new GeForce GTX 1180 iCraft at the recent ChinaJoy event, with the new card being part of Maxsun's high-end product line. The card is a thin dual-slot card with RGB strips along the front of the card and in the GeForce GTX branding on the top of the card.

Maxsun's GeForce GTX 1180 iCraft doesn't have a DVI port with one slot of the bracket dedicated to exhausting heat, while there's a single 8-pin PCIe power connector. I'm expecting the new cards to use a single 8-pin PCIe power connector thanks to the new 12nm node and what I'm sure NVIDIA will unveil as massive power efficiency through the refreshed Turing GPU architecture.

TSMC is one of the world's largest chip manufacturers, handling wafers of chips for the likes of NVIDIA and Apple has been hit with viruses on Friday night that forced multiple factories to be shut down in Taiwan.

We don't know which specific factories were hit with the viruses and shut down, and whether they were factories that are important to NVIDIA and its next-gen GeForce GTX 1180 and other graphics cards expected to be released imminently, or Apple's upcoming next-gen iPhone. August is a big month for manufacturing as we're leading into the holiday season and consumer devices are being made now, but you can't make those new expensive products without chips made by TSMC if you're one of their customers.

TSMC is now the world's largest independent semiconducer in the industry, with the company making NVIDIA's new 12nm GPUs for their upcoming GTX 11 series graphics cards, and the new 7nm chips for Apple's new iPhones. Other TMSC customers having chips made in Taiwan include AMD and Qualcomm. A small disruption like this in production can have very hefty snowball effects depending on which companies chips were being made that day.

We already know that AMD is heavily committed to the semi-custom market with Sony working hand-in-hand with RTG on the development of Navi for their next-gen PlayStation 5 console, but now it's China's time in the spotlight with a Ryzen/Vega SoC in development.

A new semi-custom system-on-chip (SoC) was announced today that packs a quad-core Ryzen CPU and Vega GPU with 24 stream processors, all packed with 8GB of GDDR5 memory. AMD is working with Zhongshan Subor on the new chip, something that will be powering a "new gaming PC and upcoming console" in China.

Inside, teh new SoC packs a 4C/8T Ryzen CPU at 3GHz, and 24 Vega CUs at 1.3GHz joined by the 8GB of GDDR5. We should expect Radeon RX 580 level performance, which is fine for 1080p and 1440p gaming at 60FPS depending on the game and detail levels used. The more interesting side here is that the Vega GPU technology is using GDDR5 memory and not HBM2 like AMD originally said when they announced the Vega GPU architecture.

NVIDIA seems to be launching its new GeForce GTX 1180 during its GeForce Gaming Celebration at Gamescom 2018, with the purported PCB of the new GTX 1180 with either 8GB or 16GB of GDDR6 leaking.

This is the front of the new GTX 1180 PCB where we can see an interesting new SLI finger that is much bigger, some tighter VRAM placements which should be a new home for GDDR6. We also have a 6+8-pin PCIe power connector setup, which is an additional 6-pin PCIe power connector on top of the single 8-pin PCIe power connector that was required on the GeForce GTX 1080 Founders Edition.

Keen-eyed readers will notice that we probably won't have a DVI output on the new GeForce GTX 1180, but we should expect the new VirtualLink connector through USB Type-C. The SLI finger upgrade could be awesome, paving the way for NVLink for gamers compared to the aging (and virtually dead) SLI technology that has sh*t multi-GPU scaling. Could we expect 90-100% scaling with multiple GTX 1180s in NVLink/next-gen SLI? We're going to need it for those 4K 144Hz HDR G-Sync monitors and 65-inch 4K 120Hz HDR G-Sync BFGD TVs.

NVIDIA has just released their latest Game Ready 398.82 drivers, which are Game Ready for GeForce graphics card owners for World of Warcraft: Battle of Azeroth which launches on August 14, and Monster World Hunter on August 9.

World of Warcraft: Battle for Azeroth is right around the corner, with NVIDIA recommending the GeForce GTX 1060 for 1080p 60FPS, the GTX 1070 for 1440p 60FPS and the GTX 1080 for 4K 60FPS. NVIDIA has benchmarks the game, where they've said: "If you like World of Warcraft, you like stats, so here's some: at 1920×1080, our benchmarking concluded with an average framerate of 65.4 on the GeForce GTX 1060, when paired with an i5-6600K. A minimum framerate of 53 was the lowest we saw, and the frametime plot was near-perfect, indicating performance was smooth and stable".

NVIDIA continues: "At 2560×1440, the GeForce GTX 1070 gave us a 65.6 FPS average, with a minimum of 52, and equally-smooth frametimes And at 4K, it's recommended that players step up to an i7 for a perfect 60 frames per second on a GeForce GTX 1080, with the occasional dip to 51 FPS".

NVIDIA's various AIB partners will have their GeForce GTX 11 series graphics cards in consumers hands in the August/September timeframe, with Inno3D ahead of the pack when it comes to subtle viral marketing of their new cooling technology.

Inno3D held an event in China recently where they teased their next-gen iChill AIO liquid cooling technology which should be used on their next-gen GTX 11 series graphics cards. Inno3D's next-gen iChill AIO liquid cooler will be used on the company's flagship overclocked GTX 1180 and GTX 1170 graphics cards (or whatever they end up as, GTX 1185 or GTX 1195).

The new coolers will have the iChill logo on both the front and the back of the card, illuminated by beautiful LEDs. It will be a dual-slot cooler with dual 8-pin PCIe power connectors while the watercooling tubes feeding out the back above the power connectors.

NVIDIA has just announced its new GeForce Gaming Celebration event for August 20, which will take place at Gamescom 2018 in Cologne, Germany.

NVIDIA will be showcasing its latest graphics cards and 4K 144Hz HDR G-Sync gaming monitors and 65-inch 4K 120Hz HDR BFGD TVs, alongside a slew of games and other gaming-related goodies during Gamescom 2018. Until now, rumors suggested July 30 as the release date for the GeForce GTX 1180 graphics card with an August/September release window that AIB partners told me during Computex 2018... and now that seems the case.

The company has officially announced teh event on their website, saying that it will be "loaded with new, exclusive, hands-on demos of the hottest upcoming games, stage presentations from the world's biggest game developers, and some spectacular surprises". I wonder what those "spectacular surprises" could be, eh?

NVIDIA delayed its GeForce GTX 11 series launch earlier this year which is why we're in such a sad state of affairs with rumors and leaks, and then the mining boom happened right in the middle of all of this, as well as AMD completely failing with its Radeon business... all of which have created the perfect storm for NVIDIA.

Now we have new information that VideoCardz is reporting from YouTuber 'Gamer Meld' who reportedly received an email from an NVIDIA AIB partner that provided some release dates for the GTX 11 series. The email said the GeForce GTX 1180 would be released on August 30, the GTX 1170 and GTX 1180+ on September 30, and the mid-range GTX 1160 on October 30.

So with this new information in-hand, let's break it down.

As we get closer to the purported release window of NVIDIA's next-gen GeForce GTX 11 series, the more juicy the rumors get - leading into today and the purported GeForce GTX 1170. As always, this is a huge rumor and I doubt it's even close to real, but let's take a look at it anyway.

The new purported GeForce GTX 1170 has been benchmarked with an Intel Core i5-8600K processor in 3DMark, beating out NVIDIA's best consumer GeForce graphics card, the GTX 1080 Ti. Inside, the GTX 1170 reportedly runs a 256-bit memory bus with 16GB of GDDR6 RAM and a GPU clock speed of 2.5GHz, which is huge.

NVIDIA used a brick wall with GPU clock speeds with Pascal with even the most enthusiast-focused GTX 1080/1080 Ti hitting a maximum GPU clock of 2.2GHz, unless it was heavily modded to take more power. When that happened, the walls to 2.5GHz were risen, and the path to 2.7-2.8GHz were opened.

Apple announced a bunch of updated MacBook Pros with some additional GPU horsepower if needed, with the announcement of a team up with Blackmagic on a new external GPU.

The new external GPU in question is based on the Radeon Pro 580 with 8GB of VRAM, which Apple says is good for up to 2.8x faster graphics performance on the refreshed 15-inch MBP, while it should be a whopping 8x faster than the GPU inside of the new 13-inch MBP.

The new Apple/Blackmagic external GPU mash up packs 2 x Thunderbolt 3 ports, 4 x USB 3.0 ports, and 1 x HDMI 2.0 port which means it doubles as a docking station, too. It's also the first external GPU solution to support Thunderbolt 3 displays, such as the 5K-capable LG display that Apple itself recommends for the MacBook Pro.

AMD will be unleashing its refreshed Vega 20 GPU in the coming weeks, made on the industry-leading 7nm node and it'll feature some great improvements over the current Vega 10 GPU made on 14nm.

Vega 20 isn't being aimed at gamers and rather the HPC/automotive/AI markets with its huge 32GB of HBM2 and PCIe 4.0 standard. AMD is looking to aim Vega 20 at NVIDIA's current flagship Tesla V100 solution, which was recently bumped up to 32GB of HBM2 at GTC 2018 earlier this year.

The new details on Vega 20 see it coming in at just 360mm2 compared to Vega 10 at 510mm2, a huge 70% reduction in total die size thanks to the fresh 7nm node. This is where a fork in the road happens: AMD can choose power savings with the die size savings, or they can ramp clock speeds up.

AMD has lost a few people to Intel in the last year, and now they're fighting back by securing an ex-Intel executive who served as the Vice President, Core and Visual Computing Group, Chief Engineer, VTT and Director of Hardware and Co-Director of Architecture, VPG at Intel. Martin Ashton had a huge title at AMD.

Ashton was with Intel for just 21 months, and before joining Chipzilla hs worked in the UK with Imagination Technologies. Imagination Technologies, in case you didn't know, made the PowerVR GPUs that went over to Apple, with Ashton moving to Intel before that happened. And now as Intel is slipping in more ways than one, Ashton has jumped ship to AMD.

AMD reached out to me to let me know that Martin will be joining the Radeon Technologies Group and will "drive architecture and execution of Radeon graphics IP that will extend across our discrete graphics, integrated graphics, and semi-custom products".

NVIDIA is expected to unveil their next-gen GeForce GTX 11 series graphics card later this month, with a release in August/September as per our previous exclusive reports, but what about gaming notebooks?

Well, it looks like the GeForce GTX 1160 could fly in and save the day during the holidays according to LaptopMedia who are reporting the GTX 1160 6GB mobile GPU will be used inside of the updated Lenovo Legion Y530 notebook, a notebook that currently packs the GTX 1050 and GTX 1050 Ti depending on which model you buy.

We have no idea what to expect from the new GeForce GTX 1160 in mobile form, but we should expect NVIDIA to use tap that new GDDR6 technology which will give the mid-range GPU a nice bump in any laptop it powers. We could expect a rather large 336GB/sec from 6GB of GDDR6 on a 192-bit memory bus, which would be a large performance jump that would bring it up to the 320GB/sec that the GTX 1080 pumps out with its GDDR5X.