During the next year and a half, the Finnish IT Center for Science (CSC) will be purchasing a new supercomputer in two phases. The first phase consists of Atos' air-cooled BullSequana X400 cluster which makes use of Intel's Cascade Lake Xeon processors along with Mellanox HDR InfiniBand for a theoretical performance of 2 petaflops. Meanwhile, system memory per node will range from 96 GB up to 1.5 TB with the entire system receiving a 4.9 PB Lustre parallel file system as well from DDN. Furthermore, a separate partition of phase one will be used for AI research and will feature 320 NVIDIA V100 NVLinked GPUs configured in 4-GPU nodes. It is expected that peak performance will reach 2.5 petaflops. Phase one will be brought online at some point in the summer of 2019.

Where things get interesting is in phase two, which is set for completion during the spring of 2020. Atos' will be building CSC a liquid-cooled HDR-connected BullSequana XH2000 supercomputer that will be configured with 200,000 AMD EPYC "Rome" CPU cores which for the mathematicians out there works out to 3,125 64 core AMD EPYC processors. Of course, all that x86 muscle will require a great deal of system memory, as such, each node will be equipped with 256 GB for good measure. Storage will consist of an 8 PB Lustre parallel file system that is to be provided by DDN. Overall phase two will increase computing capacity by 6.4 petaflops (peak). With deals like this already being signed it would appear AMD's next-generation EPYC processors are shaping up nicely considering Intel had this market cornered for nearly a decade.

AMD's Radeon Technology Group (RTG) continues to be in a state of flux, with another executive leaving the company. This time, RTG General Manager and Senior Vice President, Mike Rayfield (pictured on the left below) will be resigning by year's end, marking the second time in a calendar year the group has been devoid of a formal leader. Rayfield previously worked for NVIDIA as the General Manager of their Tegra business unit, and was Vice President and General Manager of Micron's mobile storage business unit prior to that. Having served in all these roles for quite some time, it is somewhat of a surprise to see him leaving AMD so quickly, having only just joined the company back in February 2018. With that said, AMD has said the reason for this departure is so that he can "spend more time with his family and pursue his personal passions".

David Wang (picture on the right below), the current Senior Vice President of Engineering at RTG, will be his interim successor. He already has a storied history at AMD, having worked on ATI/AMD graphics cards as a GPU engineer from the R300 to GCN 1.0 in a time period ranging from the years 2000 to 2012. Under AMD's dual leadership model which was implemented after Raja Koduri's resignation, both Wang and Rayfield have worked together leading the RTG group. Thanks to this, the transition of duties should be relatively smooth even though Wang's role is only temporary, and it will be interesting to see whom AMD picks as Rayfield's long-term successor.

AMD is putting final touches on its Ryzen 3000U series APUs for ultra-portable notebooks and 2-in-1 devices. Thai PC enthusiast Tum Apisak shared links to Geekbench scores of at least three SKUs, the Ryzen 3 3200U, the Ryzen 3300U, and the Ryzen 5 3500U. The Ryzen 3 3200U combines a 2-core/4-thread CPU component, while the Ryzen 3 3300U packs a 4-core/4-thread CPU, and the Ryzen 5 3500U a better equipped 4-core/8-thread CPU. While the 3200U's CPU is clocked high at 2.60 GHz, the 3300U and 3500U are both clocked at 2.10 GHz. The iGPU specs are still under the wraps as Geekbench only tested the single- and multi-threaded CPU performance. The 3200U scores 3428 points single-threaded owing to its higher nominal clocks, and around 6500 points multi-threaded. The 3300U scores 9686 points in multi-threaded owing to its additional cores (sans SMT). The 3500U increases the multi-threaded score to over 11280 points multi-threaded, on account of being quad-core with SMT.

There's no clarity on the underlying micro-architecture. While the source mentions the codename of these chips as Picasso, the silicon still appears to be 14 nm "Raven Ridge." Over generation, AMD only appears to have pushed its current parts lower down the product stack. For example, the Ryzen 3 3300U appears to share the same CPU configuration (albeit with 5% higher clock-speeds) as the Ryzen 5 2500U from the current-generation. The Ryzen 5 3500U, on the other hand, appears to have essentially the same (again, marginally speed-bumped) CPU as the Ryzen 7 2700U. HP is ready with notebook and 2-in-1 products based on all three chips, although they're unlikely to launch before year-end. Perhaps CES could be a nice launchpad.

AMD today released their yearly driver update called Radeon Software Adrenalin 2019 Edition v18.2.2. For the past three years, the company has been releasing year-end feature-packed updates to Radeon Software. The 2017 version saw the introduction of Radeon Software "Adrenalin", before which we had Crimson/ReLive and Omega. The company also sped up its driver release cycle to match AAA game releases, and over the years, increased its reliability score and customer satisfaction. With Radeon Software Adrenalin, AMD is targeting three key consumer segments: Gamers, Enthusiasts, and Creators.

Intel's discrete graphics card for PC enthusiasts is real. Intel won't just address the pro-graphics and accelerated-compute markets, but also consumer graphics, challenging the duopoly of NVIDIA GeForce and AMD Radeon. Scheduled for 2020, the new Intel Xᵉ is a family of discrete GPUs targeting client-segment (consumer graphics) as well as enterprise (pro-graphics and compute).

As for performance, we speculate that the first Xᵉ products could span a vast lineup of ASICs starting single-digit TFLOP/s range for the client-segment GPU, looking purely at a nondescript performance-time graph presented by Intel. This graph depicts performance double linearly over time up to Gen9, and increase to Intel's own state 1 TFLOP/s for the Gen11 iGPU core in 2019 (a full four years following Gen9). There are a spectrum of GPUs going from the entry-level client-segment all the way up to mid-range and enthusiast segment (Intel finally used the E-word).

Intel today unveiled its first clean-slate CPU core micro-architecture since "Nehalem," codenamed "Sunny Cove." Over the past decade, the 9-odd generations of Core processors were based on incrementally refined descendants of "Nehalem," running all the way down to "Coffee Lake." Intel now wants a clean-slate core design, much like AMD "Zen" is a clean-slate compared to "Stars" or to a large extent even "Bulldozer." This allows Intel to introduce significant gains in IPC (single-thread performance) over the current generation. Intel's IPC growth curve over the past three micro-architectures has remained flat, and only grew single-digit percentages over the generations prior.

It's important to note here, that "Sunny Cove" is the codename for the core design. Intel's earlier codenaming was all-encompassing, covering not just cores, but also uncore, and entire dies. It's up to Intel's future chip-designers to design dies with many of these cores, a future-generation iGPU such as Gen11, and a next-generation uncore that probably integrates PCIe gen 4.0 and DDR5 memory. Intel details "Sunny Cove" as far as mentioning IPC gains, a new ISA (new instruction sets and hardware capabilities, including AVX-512), and improved scalability (ability to increase core-counts without running into latency problems).

Today, Intel is revealing major details about its upcoming CPU and graphics architectures to select audience. A big scoop VideoCardz landed is the company's next-generation Gen11 integrated graphics core, the first major upgrade to the company's 4-year old Gen9 architecture. According to them, a Gen11 (default GT2 trim we assume) graphics core should offer a compute throughput of 1 TFLOP/s, which is in the league of the Radeon Vega 8, with its 1.12 TFLOP/s throughput. The Vega 8 is part of AMD's Ryzen 3 2200G processor.

Raw compute power only paints half the picture, the iGPU reportedly also supports tile-based rendering. This is a highly publicized method of rendering that made its consumer debut with NVIDIA "Pascal." Also mentioned are redesigned FPU interfaces, support for half-precision FP16, 2x pixel/clock pipelines, display stream compression that lets it support 5K and 8K displays, and adaptive sync. Intel will debut its Gen11 iGPU with its upcoming Core "Ice Lake" processors that debut on the company's 10 nm silicon fabrication process.

Oh hey remember this news post from July last year? Damien Triolet's work history off-late has been one of many such recent stories. These tend to begin with AMD, and RTG in particular, getting a cash infusion and growing in 2016 and 2017 to where they hired some of the best engineers and marketing personnel from the industry- media or otherwise. This follows a more stagnant GPU division in 2017-2018, Intel deciding to dip their toes back into the discrete GPU market, and in turn.. persuading many to cross over to the blue side.

According to Damien's LinkedIn and FaceBook profiles, he has started working for Intel from November 26, 2018 in a technical marketing position in their Gaming and Graphics division, a role analogous to his from his days at AMD. Presumably, he joins Raja Koduri and the many others who have followed this exact path of late, and everyone remains curious as to what the finished retail product will be. In the meantime, we here at TechPowerUp wish him the best again for his new venture. We had the pleasure of interacting with Damien on multiple occasions in the past, some as colleagues in the media giving hardware manufacturers a hard time, and others when he was hosting us as an AMD employee. His tenure at Hardware.fr has been inspiring to us, with excellent reviews that no doubt were what caught the eyes of AMD in the first place, and Intel will definitely gain from his presence.

AMD late November filed a trademark application with the USPTO for a new logo, for its second generation "Vega" graphics architecture, built around the 7 nm silicon fabrication process. The logo looks similar to the original Vega "V," with two bands marking out the Roman numeral II (2). This logo could appear on the product and marketing on a series of new Radeon Pro and Radeon Instinct (and possibly even gaming-grade Radeon RX?) graphics cards based on AMD's new "Vega 20" multi-chip module. This chip features a doubling in memory bandwidth thanks to its 4096-bit wide HBM2 interface, and the optical shrink of the GPU die to the 7 nm node could enable AMD to dial up engine clocks significantly.

Anyone looking for an update to their CPU that didn't quite jump on the Coffee Lake/Zen/Zen+ bandwagon is likely paying close attention to AMD's upcoming Zen 2 CPUs. The upcoming AMD processors will finally leave the company ahead of Intel in terms of manufacturing process for the first time in years, and will bring about AMD's new vision for HCC desktop processors in a chiplet design. With the release of Zen 2 set for 2019 (probably around Computex), and its launch being of such importance to AMD, it isn't that surprising that some promotions/teases are already popping up.

The tease in question was posted by an AMD-contracted Sales agency in South Korea, which launched a campaign inviting users to guess Cinebench scores for upcoming AMD processors: namely, the Ryzen 7 3700X and Ryzen 5 3600X - thus confirming the nomenclature for AMD's upcoming CPUs. The contest finishes on December 14th, and is basically asking users to take a gander on scores for unreleased CPUs - promising prizes of said CPUs when they launch.

No the title is not a typo, AMD has today released their Radeon Software Adrenalin 18.12.1.1 beta drivers. This release is essentially the same as the previous 18.12.1 beta drivers, except for the fact they add support for the Epic Games store. Other than that one inclusion nothing else seems to have changed with no performance improvements or fixes mentioned. The known issues AMD have listed remain the same as well, with some systems running multiple displays still potentially experiencing mouse lag when one display is enabled but turned off. Assassin's Creed: Odyssey is still likely to crash at certain locations on Windows 7 systems. Overall, if you plan to check out the Epic Games store you may want to grab these drivers otherwise you can likely just hit snooze and wait for the next release.DOWNLOAD: AMD Radeon Software Adrenalin 18.12.1.1 Beta

It's no news that AMD's Radeon RX 570 graphics card is carved out of the same "Polaris 20" silicon as the RX 580, by disabling 4 out of 36 GCN compute units. AMD kicked a controversy recently, when it launched a China-specific Radeon RX 580-branded SKU with the core-configuration of the cheaper RX 570, confusing Chinese consumers. It turns out that this RX 580 2,048 SP SKU is based on the same exact ASIC variant of the "Polaris 20" silicon as the RX 570, with the only difference being device ID.

We watch a lot of GamersNexus content. Our GPU Database curator noticed something interesting in their recent teardown of a Dataland Radeon RX 580 (2,048 SP) graphics card directly imported from China. The unique ASIC sub-variant code etched on the GPU's aluminium reinforcement brace matches that of the RX 570. AMD internally refers to the RX 570 as "Polaris 20 XL," and its ASIC code etched is supposed to be "215-0910052." For the RX 580, the real one, aka "Polaris 20 XTX," the code etched is "215-0910038." Thanks to GamersNexus' high-resolution filming, our curator was able to spot the ASIC code for "Polaris 20 XL" on the Dataland card's GPU. This confirms that AMD merely took an RX 570 and gave it a different device ID to create the RX 580 2,048 SP, leaving consumers to wade through the confusion.

MSI late November released BIOS updates for its entire range of socket AM4 motherboards, mainly to include AGESA 1.0.0.6, which improves memory compatibility. The updates also pack various motherboard model-specific fixes. Apparently there has been a goof-up with this update (not in a bad way, though). The latest BIOS updates "accidentally" unlock the otherwise locked Athlon 200GE dual-core processor with integrated graphics.

No, you won't get more cores or iGPU stream processors, but the base-clock multiplier letting you easily overclock the processor is now unlocked. This adds value to the $55 processor, given that competing Celeron and Pentium Gold SKUs are overpriced in the market these days. In their overclocking adventures with the 200GE paired with a MSI B450 Gaming Pro Carbon AC motherboard, TechSpot observed that the chip (which is clocked at 3.2 GHz by default) wouldn't POST beyond 3.90 GHz, and that you need vCore at 1.4 V to sustain even 3.90 GHz. They found 3.80 GHz at 1.35 V to be 100% stable (an 18.75% overclock). The memory clocks are still restricted to DDR4-2666, although you can still play with timings.

AMD seems to be picking up steam over Intel's previous sky-high dominance of the desktop CPU market (Intel still dominates aplenty, really; but AMD has been clawing back market percentage monthly). The latest figures from the German retailer show shoppers taking advantage of AMD's newfound competitiveness in the CPU space, with increasing sales momentum starting on June 2018 up to a staggering 69% total AMD units sold against Intel's 31% during the month of November.

AMD is preparing Radeon Software Adrenalin 2019 Edition, keeping up with the tradition of big year-end software launched that are high on new features. VideoCardz got a whiff of three if its key features. One of them is Voice Control. You will soon be able to talk to your graphics card through voice-commands picked up from your microphone. Rolling out initially in English and Chinese, you will soon be able to control a lot of things while still in game and not having to take your hands off your keyboard/mouse. These include taking screenshots, telling the driver to record or stream your gameplay using ReLive, adjust display settings such as brightness, contrast, and gamma; etc. It should also take commands to change resolution, rotate displays, move workspaces between multiple displays, etc.

Enthusiasts can rejoice as WattMan could get an expanded feature-set, including one-click overclocking, similar to NVIDIA's "Turing" OC Scanner. This one-click OC feature cranks up not just your GPU's clock speeds, but also video memory. The drivers also support fully automated undervolting, which should make several current-generation Radeon graphics cards more energy-efficient. Lastly, AMD seems to be compensating for NVIDIA's VirtualLink implementation by introducing a more advanced interface - direct streaming to VR headsets. Think Steam Link for VR headsets.

Intel's 9th generation re-branding of its "Skylake-X Refresh" HEDT processors could have a rather short lifespan in the company's product-stack of just three quarters. Intel is planning to launch its next HEDT platform, codenamed "Glacier Falls," which succeeds the current "Basin Falls Refresh." "Glacier Falls" could launch around late-Q2/early-Q3, with unveils slated for Computex 2019 (June). These details are part of a slide leak from an internal presentation from motherboard manufacturer GIGABYTE.

The slide revealing "Glacier Falls" also mentions two new mainstream-desktop platform chipsets, the B365 Express and the H310C Express. There is no information on what sets these apart from the current B360 and H310, respectively. We predict their entry is necessitated by stronger CPU VRM requirements to support 9th generation Core 6-core and 8-core processor SKUs that are known to pull up to 140 Watts of power (unrelated to TDP, which is calculated on the basis of nominal clock speed and not Turbo Boost). There's also a faint possibility of Intel giving the B365 CPU overclocking support to compete with the value proposition of AMD's B450.

SAPPHIRE Technology introduces GearBox Thunderbolt 3 eGFX - a new expansion chassis for Mac Pros, Ultrabooks and Small Form Factor Computers. Connect a powerful graphics card to a Thunderbolt 3 ready device and experience greatly improved performance in professional software and games.

PATRIOT , a global leader in performance memory, SSDs, gaming peripherals, and flash storage solutions, has announced the release of their fastest kit ever of DDR4 memory today, VIPER STEEL SERIES DDR4 16GB (2x 8 GB) 4400 MHz. Built for the latest Intel and AMD platforms, the VIPER STEEL SERIES provides extreme performance and stability for the most demanding computer environments. The VIPER STEEL SERIES DDR4 memory kit brings Intel XMP 2.0 performance to new heights and is able to achieve up to 4400 MHz with a simple setting on supporting Intel's latest processors and motherboards.

The VIPER STEEL SERIES utilizes a custom designed high-performance aluminum heatshield for providing optimal heat dissipation when overclocking and foregoing high-performance tasks. These hand-tested memory modules feature the highest quality memory chips and are running performance timings of CL19-19-19-39 at 1.45 V. VIPER STEEL SERIES DDR4 4400 MHz is optimized for PC enthusiasts, overclockers, or those who want to push their system's performance to the limit through the use of extremely fast components.

AMD has released today the Radeon Software Adrenalin 18.12.1 beta drivers. Overall this release merely adds support for Just Cause 4, which is set to release on December 4, 2018. While it is good to see AMD improving support for games ahead of their release, this driver offers nothing else tangible, meaning no performance boosts in other titles, changes or fixes to speak of. That said, AMD has listed a few known Issues with these drivers, such as; mouse lag on multiple displays when they are enabled but one is powered off, and Assassin's Creed: Odyssey may crash at specific locations on Windows 7 systems. We have of course uploaded these latest drivers to our download section.DOWNLOAD: AMD Radeon Software Adrenalin 18.12.1 Beta

GeIL - Golden Emperor International Ltd. - one of the world's leading PC components and peripheral manufacturers today announced the new additions to their SUPER LUCE RGB SYNC Series DDR4 memory. The new modules are available in three capacities and speeds: 4133 MHz 16 GB (8 GB x2) CL19-19-19-39 1.40 V, 3600 MHz 16GB (8 GB x2) CL16-18-18-36 1.35 V, and 3200 MHz 16 GB (8 GB x2) CL14-14-14-34 1.35V

The 4133 MHz 16 GB (8 GB x2) CL19-19-19-39 1.40 V DDR4 kits can be the overclocker's new favorite with support for most ASUS, GIGABYTE, MSI and ASROCK Z390 and Z370 motherboards as well as giving enthusiasts the room to push their systems to the limit without exception. The 3600 MHz 16GB (8 GB x2) CL16-18-18-36 1.35 V DDR4 kits meets the memory needs for any gaming system using the latest Intel platforms and is an excellent solution for first time overclockers. The 3200 MHz 16 GB (8GBx2) CL14-14-14-34 1.35 V DDR4 kits are compatible with both Intel and AMD AM4 platforms and feature a low latency design to enhance performance and offer the most affordable upgrade option for users.

According to Jon Peddie Research, overall desktop GPU shipments have fallen by 16% YoY - a not unexpected turn of events considering the state of the crypto mining boom then and now (where it's virtually absent). The YoY change means that production volumes planned during the mining boom are now above and beyond the channel's capability to move them through user demand, hence the diminishing prices in graphics cards (aided by the dump of mining-bought graphics cards in the second-hand markets).

Overall, GPU shipments still increased by 10.64% compared to last quarter in the overall market, fueled mostly by Intel - AMD shipments increased 6.5% Nvidia increased 4.3% and Intel increased 13.1% compared to their own previous shipment quotas. Still, AMD's market share from last quarter decreased -0.6%, Intel's increased 1.5%, and Nvidia's market share decreased -0.97%. JPR also cites the US's additional tax on China imported goods, as well as descending stock market values as reasons for consumers (both singular and business) to be holding off on purchases.

A SiSoft SANDRA results database entry for a 2P AMD "Rome" EPYC machine sheds light on the lower cache hierarchy. Each 64-core EPYC "Rome" processor is made up of eight 7 nm 8-core "Zen 2" CPU chiplets, which converge at a 14 nm I/O controller die, which handles memory and PCIe connectivity of the processor. The result mentions cache hierarchy, with 512 KB dedicated L2 cache per core, and "16 x 16 MB L3." Like CPU-Z, SANDRA has the ability to see L3 cache by arrangement. For the Ryzen 7 2700X, it reads the L3 cache as "2 x 8 MB L3," corresponding to the per-CCX L3 cache amount of 8 MB.

For each 64-core "Rome" processor, there are a total of 8 chiplets. With SANDRA detecting "16 x 16 MB L3" for 64-core "Rome," it becomes highly likely that each of the 8-core chiplets features two 16 MB L3 cache slices, and that its 8 cores are split into two quad-core CCX units with 16 MB L3 cache, each. This doubling in L3 cache per CCX could help the processors cushion data transfers between the chiplet and the I/O die better. This becomes particularly important since the I/O die controls memory with its monolithic 8-channel DDR4 memory controller.

AMD's Ryzen Notebook lineup seems to be very important to company, at least when going by how often it gets mentioned in the AMD financial analyst calls. That's why it's even more surprising that the driver situation for these products has been nothing but terrible. Some Ryzen Raven Ridge based notebooks haven't seen a single driver update since their release over a year ago, which is much worse than on any other notebook platform.

Users complained about this on Reddit, and AMD responded through an official account that the issue is that "drivers are typically tailored for specific OEM platforms", and that "releasing generic APU graphics drivers across all AMD Ryzen mobile processor-based mobile systems could result in less-than-ideal user experiences". AMD also made it clear that they will be working with OEMs to increase the release frequency of Ryzen Mobile graphics drivers, targeting two releases per-year in 2019.To me this explanation sounds like bs.

It has been confirmed by Seasonic that their FOCUS PLUS power supplies are experiencing potential conflict with a select number of graphics cards. In regards to NVIDIA, the compatibility issue is currently limited to the ASUS GeForce GTX 970 STRIX. In limited instances, this particular graphics card can encounter a black screen under heavy load. Testing by Seasonic has determined that the issue is caused by higher than normal ripple when the GPU is heavily stressed. They also determined that solving the problem simply required using modified PCIe cables that feature enhanced shielding.

When it comes to AMD GPUs things are a bit more problematic. Both of AMD's Vega 56 and Vega 64 series of graphics cards with serial numbers before January 2018, can experience system shutdowns when paired with a FOCUS PLUS power supply. The issue is caused by higher than normal peak current being emitted by these graphics cards when under heavy load. This results in the power supply's internal protection safety being triggered. Considering Seasonic's popularity the fact these power supplies are encountering problems is likely going to be a hit to their brand image. At least they are stepping up to the plate having admitted to the problem while also investigating the root causes. They are also encouraging anyone that is encountering issues to contact them for assistance.

Sony's announcement of the Playstation team skipping E3 2019 took everyone by surprise aside from a few on Reddit who had paid attention to a thread created the day before. Reddit user RuthenicCookie seemed to know a lot more about Sony's plans for their popular game console for the next few years, as well as game titles supporting this current console generation and the next. Amidst a lot of the tasty rumor bits that should interest console gamers, something more relevant to us directly is the mention of the Playstation 5 to continue using AMD for processing power.

This is a logical move to just about everyone familiar with the industry, and Sony needed to up the CPU horsepower in particular to compete with the XBOX One X and offer a true 4K/60 FPS solution for gaming without framerate drops galore. As such, said redditor shared information saying that the current plans involve an 8-core Ryzen-based processor and an estimated console price point of $500. Sony may well share a teaser about the console next year, with retail availability expected in the holiday season 2020 (two years from now, thus). As such, developer kits are likely already ready meaning the specs are finalized as well. This may mean we will see either the first or second gen Ryzen APUs, and not Ryzen 2 as many may have hoped. No word yet on what Microsoft is cooking in their side of the kitchen, but incremental console updates means we may see a Ryzen 2-powered console sooner than later as well.