MojoKid writes: AMD got the attention of PC performance enthusiasts everywhere with the recent launch of its Ryzen 7 series processors. The trio of 8-core chips competitively take on Intel's Core i7 series at the high-end of its product stack. However, with the extra attention AMD garnered, came significant scrutiny as well. With any entirely new platform architecture, there are bound to be a few performance anomalies — as was the case with the now infamous lower performance "1080p gaming" situation with Ryzen. In a recent status update, AMD noted they were already working with developers to help implement "simple changes" that can help a game engine's understanding of the AMD Zen core topology that would likely provide an additional performance uplift with Ryzen. Today, we have some early proof-positive of that, as Oxide Games, in concert with AMD, released a patch for its game title Ashes Of The Singularity. Ashes has been a "poster child" game engine of sorts for AMD Radeon graphics over the years (especially with respect to DX12) and it was one that ironically showed some of the worst variations in Ryzen CPU performance versus Intel. With this new patch that is now public for the game, however, AMD claims to have regained significant ground in benchmark results at all resolutions. In the 1080p benchmarks with poweful GPUs indeed a Ryzen 7 1800X shows an approximate 20% performance improvement with the latest version of the Ashes, closing the gap significantly versus Intel. This appears to be at least an early sign that AMD can indeed work with game and other app developers to tune for the Ryzen architecture and wring out additional performance.
MojoKid writes: Intel is readying its 10 nanometer Cannonlake processor architecture for release later this year and has begun touting advanced manufacturing process technology that it's built on. This comes on the heels of major competitive announcements, like Samsung at 10nm with Qualcomm, for example, which is already shipping Snapdragon 835 processors based on Samsung 10nm FinFET tech. The chip is rumored to make an appearance tomorrow at Samsung Unpacked in New York in the Galaxy S8 smartphone. Regardless, Intel stated at its Technology and Manufacturing Day presentation today that its 10nm technology is a full generation ahead of the "other guys." Intel also claims that it has the highest feature density at 10nm and is leveraging the use of it to better ensure that it stays well ahead of competitors like Samsung with regard to performance and costs (which are claimed to be 30 percent lower than its rivals). Intel was able to shrink its minimum gate pitch from 70nm to 54nm, going from the 14nm to 10nm process nodes. In addition, the minimum metal pitch has shrunk from 52mm to 36mm. As a result, Intel's logic transistor density at 10nm is 2.7x higher than its previous generation 14nm process, and claimed to be another 2x greater than its closest competitors. Intel also claims that further refinements at 10nm mean its upcoming Cannonlake processors will provide up to a 25 percent uplift in performance, while consuming 45 percent less power than 14nm counterparts (i.e. Skylake and Kaby Lake).
MojoKid writes: Intel took the wraps off its Optane Memory devices for client PCs today and the product looks just like current generation 80mm M.2 "gumstick" type solid state drives. However, Intel Optane Memory is based on the company's 3D Xpoint memory technology and is meant to be used as an accelerator for systems with relatively low-speed storage devices, like hard drives. Intel Optane Memory products and associated software are designed to cache the most frequently accessed bits of data on a compatible system, which can significantly increase performance and responsiveness of slower drives. The SSD can be paired to any standard hard drive or SATA drive for that matter, regardless of the capacity. The Optane memory is used as a high-speed repository of the most commonly accessed data blocks (not necessarily complete files). Usage patterns on the hard drive are monitored and the most frequently accessed bits of data are copied from the hard drive to the Optane drive. Because it's is used as a cache, it is not presented to the end-user as a separate volume. The first products in Intel's Optane Memory line-up will be M.2 type NVMe SSDs, with capacities of 16GB and 32GB. Note that Intel Optane Memory will work only on Windows 10 64-bit systems with Intel 7th Gen Kaby Lake-based processors and 200-series chipsets, or newer systems. 16GB and 32GB Intel Optane Memory modules will be available initially through retailers with MSRPs of $44 for the 16GB part and $77 for the 32GB model. There are already over 130 motherboards on the market and systems featuring the technology will be made available soon from all of the major players, including Dell, Lenovo, HP and others.
MojoKid writes: Qualcomm is lifting the veil on performance benchmark numbers for its new Snapdragon 835 processor today (or Mobile Platform, as Qualcomm is referring to it now) and it's looking like a notable improvement over the Snapdragon 820 and 821. The Snapdragon 835 is expected to provide up to 11 hours of 4K video playback and also serve up hours of VR gaming on a single charge, along with performance increases of up to 25 percent in both CPU and graphics/gaming related workloads. The Snapdragon 835 SoC is built on 10nm FinFET technology, which results in significantly lower power consumption, though Qualcomm notes most of the power consumption gains were realized in the 8-core CPU block — an ARM big.Little design — for efficiency. Four larger, semi-custom Kyro 280 cores are clocked at 2.45GHz and the smaller, lower-power four cores are clocked up to 1.9GHz. Though proving battery life claims will have to wait for retail shipping smartphones that employ the new Snapdragon SoC to ship in market, benchmark numbers taken from Qualcomm prototype devices show impressive results. Actual gains north of 20 percent in gaming and graphics tests were observed, but more modest gains of 10 — 15 percent were realized in CPU-centric tests.
MojoKid writes: Intel unveiled its first SSD product that will leverage 3D Xpoint memory technology, the new Optane SSD DC P4800X. The Intel SSD DC P4800X resembles some of Intel's previous enterprise storage products, but this product is all new, from its controller to its 3D Xpoint storage media that was co-developed with Micron. The drive's sequential throughput isn't impressive versus other high-end, enterprise NVMe storage products, but the Intel Optane SSD DX P4800X shines at very low queue depths with high random 4kB IO throughput, where NAND flash-based storage products tend to falter. The drive's endurance is also exceptionally high, rated for 30 drive writes per day or 12.3 Petabytes Written. Intel provided some performance data comparing its SSD SC P3700 NAND drive to the Optane SSD DC P4800X in a few different scenarios. This test shows read IO latency with the drive under load and not only is the P4800X's read IO latency significantly lower, but it is very consistent regardless of load. With a 70/30 mixed read write workload, the Optane SSD DC P4800X also offers between 5 and 8x better performance versus standard NVMe drives. The 375GB Intel Optane SSD DC P4800X add-in-card will be priced at $1520, which is roughly three times the cost per gigabyte of Intel's high-end SSD DC P3700. In the short term, expect Intel Optane solid state drives to command a premium. As availability ramps, however, prices will likely come down.
MojoKid writes: Today, AMD unveiled additional details with respect to the entire Ryzen 5 processor line-up. Unlike the Ryzen 7 series, which consists entirely of 8-core / 16-thread processors, the Ryzen 5 family has two tiers consisting of 6-core / 12-thread and 4-core / 8-thread processors. The entry-level part is the Ryzen 5 1400, a 4-core, 8-thread CPU with base and turbo clocks of 3.2GHz and 3.4GHz, respectively. The Ryzen 5 1500X has the same quad-core configuration, but with base and turbo clocks of 3.5GHz and 3.7GHz and also has support for an extended XFR frequency range of up to 3.9GHz. The Ryzen 5 1600 is a 6-core / 12-thread processor, with 3.2GHz base and 3.6GHz boost clocks. And at the top of the stack is the Ryzen 5 1600X – which has a similar 6-core configuration – but cranks things up even further to 3.6GHz / 4.0GHz. With XFR, the absolute maximum frequency for all of the Ryzen 5 processors will be somewhat higher, but AMD hasn't disclosed specifics for all parts. AMD's Ryzen 5 processor line-up will work with the very same AM4 platform as the higher-end Ryzen 7. Ryzen 5 series processors will be launching officially on April 11, with prices starting at $169 for the Ryzen 5 1400. An additional $20 will get you a Ryzen 5 1500X, while the 6-core Ryzen 5 1600 and 1600X will sell for $219 and $249, respectively.
MojoKid writes: A couple of weeks ago, AMD launched its Ryzen processor line in conjunction with its AM4 platform. The feedback has been mostly positive, save for lower-than-expected gaming performance at 1080p. AMD noted most games are not optimized for AMD hardware currently, and that 1080p gaming performance will improve in time as developers take advantage of Ryzen's architecture. Some have attributed the problem has to do with a conflict between Ryzen and the Windows 10 thread scheduler. However, AMD claims there is no merit to that theory. "We have investigated reports alleging incorrect thread scheduling on the AMD Ryzen processor. Based on our findings, AMD believes that the Windows 10 thread scheduler is operating properly for 'Zen', and we do not presently believe there is an issue with the scheduler adversely utilizing the logical and physical configurations of the architecture," AMD said. AMD also addressed reports simultaneous multi-threading (SMT) is causing performance to dip in games. The company said it expects games to either benefit from SMT or see no change at all in performance. According to AMD, this is the case in several popular game titles. However, AMD isn't leaving things entirely in the hands of developers, noting: "We have already identified some simple changes that can improve a game's understanding of the "Zen" core/cache topology, and we intend to provide a status update to the community when they are ready." AMD also addressed several other areas of concern in their update and it does appear the company is getting a better handle on a few of the anomalies with Ryzen performance.
MojoKid writes: NVIDIA is officially launching its most powerful gaming graphics card today, the GeForce GTX 1080 Ti. It was announced last week at the Game Developers Conference and pre-orders began shortly thereafter. However, the cards will begin shipping today and NVIDIA has lifted the veil on performance reviews. Though its memory complement and a few blocks within the GPU are reduced versus the NVIDIA's previous top-end card, the Titan X, the GeForce GTX 1080 Ti makes up for its shortcomings with a combination of refinement and the brute force of higher memory clocks, based on new and improved Micron GDDR5X memory, faster core clocks and an improved cooler. For gamers, the good news is, the 1080 Ti retails for $699, versus $1200 for the Titan X, and it is in fact faster, for the most part. Throughout a battery of game tests and benchmarks, regardless of the resolution or settings used, the GeForce GTX 1080 Ti performed on par with or slightly faster than the NVIDIA Titan X and roughly 30 % – 35% better than a standard GeForce GTX 1080 Founder's Edition. Versus AMD's current flagship GPU, the Radeon R9 Fury X, there is no competition; the GeForce GTX 1080 Ti was nearly 2x faster than the Fury X in some cases.
MojoKid writes: AMD lifted the veil this morning on architecture detail and performance expectations of its next generation Zen-based server platform, codenamed Naples. Naples is an up to 32-core / 64-thread variant of Zen, targeted at enterprise and data center applications. The processors will feature eight-channel DDR4 memory controllers (with up to 16 DIMMs attached per CPU), with support for up to 4TB of memory, and 128 lanes of on-chip PCI Express connectivity. In a 2P (dual processors / dual socket) configuration, Naples offers up to 64 physical scores (128 threads), access to 32 DIMM slots and an aggregate 16 memory channels. Versus a 2P Intel Xeon E5-2699A V4 based server, the 2P Naples setup ends up with double the memory channels, a higher total memory capacity, more cores (20 more physical cores, 40 more threads), and 48 more available PCI Express lanes. AMD's performance comparisons at its tech day event pit a 2P Naples server, with 512GB of DDR4 RAM, up against a 2P Intel Xeon E4-2699A V4 configuration with 384GB of RAM. The Naples system had a higher memory capacity and that memory was clocked much higher too – 2400MHz vs. 1866MHz. The Naples system has more cores, and with SMT on, can ultimately process more threads as a result. The AMD Naples system also has double the memory channels, further improving peak memory bandwidth. In its demos, AMD used a seismic analysis workload, which involved multiple iterations of 3D wave equations. According to AMD, the test taxes the entire system, including CPU cores, memory and I/O. In this demo, the AMD server system completed equations roughly 2.5x faster than the dual-socket Intel Xeon server. Expected price points weren't given, but Naples processors and servers should be available in Q2 this year.
MojoKid writes: AMD has finally lifted the veil on independent reviews of its new Ryzen series of desktop processors that bring the company's CPU architecture back more on competitive footing versus its rival, Intel's Core series. The initial family of Ryzen processors consists of three 8-core chips, the Ryzen 7 1800X at 3.6GHz with boost to 4.1GHz, the Ryzen 7 1700X at 3.4Ghz with boost to 3.8GHz and the Ryzen 7 1700 at 3GHz with boost to 3.7GHz. Each has support for 2 threads per core, for a total of 16 threads with 16MB of L3 cache on-board, 512K of L2 and TDPs that range from 65 watts for the Ryzen 7 1700 at the low-end, on up to 95 watts for the 1700X and 1800X. In comparison to AMD's long-standing A-series APUs and FX-series processors, the new architecture is significantly more efficient and performant than any of AMD's previous desktop processor offerings. AMD designed the Zen microarchitecture at the heart of Ryzen with performance, throughput, and efficiency in mind. Initially, AMD had reported a 40% target for IPC (instructions per clock) improvement with Zen but actually realized about a 52% lift in overall performance. In the general compute workloads, rendering, and clock-for-clock comparisons, the Ryzen 7 1800X either outperformed or gives Intel's much more expensive Core i7-6900K a run for its money. The lower clock speeds of the Ryzen 7 1700X and 1700 obviously resulted in performance a notch behind the flagship 1800X, but those processors also performed quite well. Ryzen was especially strong in heavily threaded workloads like 3D rendering and Ray Tracing, but even in less strenuous tests like PCMark, the Ryzen 7 series competed favorably. It's not all good news, though. With some older code, audio encoding, lower-res gaming, and platform level tests, Ryzen trailed Intel – sometimes by a wide margin. There's obviously still optimization work that needs to be done – from both AMD and software developers.
MojoKid writes: NVIDIA just lifted the veil on its latest monster graphics card for gamers — the long-rumored GeForce GTX 1080 Ti — at an event this evening in San Francisco during the Game Developers Conference (GDC). The card will sit at the top of NVIDIA's GeForce offering with the Titan X and GeForce GTX 1080 in NVIDIA's Pascal-powered product stack, promising significant performance gains over the GTX 1080 and faster than Titan X performance, for a much lower price of $699. The 12 billion transistor NVIDIA GP102 on the card has 3,584 CUDA cores, which is actually the same as NVIDIA's Titan X. However, the GeForce GTX 1080 Ti will have fewer ROP units at 88, versus 96 in the Titan X. The 1080 Ti will also, however, come equipped with 11GB of premium GDDR5X memory from Micron clocked at 11,000 MHz for an effective 11Gbps data rate. Peak compute throughput of the GeForce GTX 1080 Ti is slightly higher than the Titan X due to the Ti's higher Boost clock. Memory bandwidth over its narrower 352-bit GDDR5 memory interface is 484GB/s, which is also slightly higher than a Titan X as well. NVIDIA also noted that peak overclocks on the core should hit 2GHz or higher with minimal coaxing. As a result, the GeForce GTX 1080 Ti will be faster than the Titan X out of the box, faster still when overclocked.
MojoKid writes: LG unveiled the new G6 smartphone today, going completely back to the drawing board versus its predecessor, the not so well-received G5. In its place is a very compact aluminum unibody design and a large 5.7-inch QHD+ display with a 2880X1440 resolution. That display is the main focal point of the G6, and it has a rather unorthodox 18:9 screen ratio, which LG says allows that smartphone to better fit in your hand. LG also notes that the aspect ratio is being adopted as a universal format from the likes of film studios and content providers like Netflix. Its thin bezel also gives the LG G6 an 80 percent screen to body ratio. The handset is powered by a Qualcomm Snapdragon 821 processor along with 4GB of RAM, 32GB of internal storage and a microSD slot which can accommodate up to an additional 2TB of storage. LG also outfitted the G6 with dual 13MP rear cameras: a wide angle (F2.4 / 125) shooter and a standard camera (F1.8 / 71) with optical image stabilization. The LG G6 launches next month and will be available in Ice Platinum, Mystic White, Astro Black color options. Pricing is TBD.
MojoKid writes: AMD CEO, Dr. Lisu Su took to the stage at AMD's Ryzen tech day yesterday and opened the event with official speeds, feeds, pricing, and benchmark scores for the company's upcoming Ryzen series processors. AMD's goal with Ryzen, which is based on its Zen microarchitecture, was a 40% IPC (instructions per clock) uplift. As it turns out, AMD was actually able to increase IPC by approximately 52% with the final shipping product, sometimes more depending on workload type. Dr. Su also showed the first die shot of an 8-core Ryzen processor, disclosing that it consists of approximately 4.8 billion transistors. AMD's flagship Ryzen 7 1800X 8-core/16 thread CPU will have a base clock speed of 3.6GHz, a boost clock of 4.0GHz, and a 95 watt TDP. AMD claims the Ryzen 7 1800X will be the fastest 8-core desktop processor on the market when it arrives. The next member of the line-up is the Ryzen 7 1700X with a base clock of 3.4GHz and a boost clock of 3.8GHz, also with 8 cores and a 95 watt TDP. Finally, the Ryzen 7 1700 – sans X – is also an 8-core / 16-thread CPU, but it has lower 3.0GHz base and 3.7GHz boost clocks, along with a lower 65 watt TDP. AMD took the opportunity to demo the Ryzen 7 1800X and it was approximately 9% faster than the Core i7-6900K running Cinebench R15's multi-threaded test, at about half the cost. And in another comparison, Dr. Su put the 8-core Ryzen 7 1700 up against the quad-core Core i7-7700K, converting a 4K 60 FPS video down to 1080P and the Ryzen CPU outpaces the Core i7 by 10 full seconds. Pricing for the three initial Ryzen 7 series processors will undercut competing Intel processors significantly. AMD's Ryzen 7 1800X will arrive at $499, Ryzen 7 1700X at $399, and Ryzen 7 1700 at $329. Based on current street prices, Ryzen will be between 20% — 50% lower priced but AMD is claiming performance that's better than Intel at those price points.
MojoKid writes: Yet another AMD Ryzen leak is making the rounds, one that details an extensive lineup of 17 processors. It's a continuation of a previous leak supposedly outing AMD's top-to-bottom retail launch lineup, only now with individual part numbers and TDP ratings for every SKU. The leaked chart lists all 17 Ryzen SKUs, a dozen of which sport 65 TDP ratings with the remaining five listed having a 95W TDP. Eight of the Ryzen chips are quad-core parts, four are six-core CPUs, and five are eight-core processors. AMD will allegedly bundle an updated Wraith cooler codenamed HS81 with its Black Edition Ryzen processors that have a 95W TDP. They include the Ryzen 7 1800X, Ryzen 7 1700X, and Ryzen 5 1600X. Also, new information on one of AMD's top-end chips, the Ryzen 7 1700X, has surfaced as well, claiming a $389 price tag and performance on par with Intel's Core i7-6900K Broadwell-E 8-core chip that retails for over $1K.
MojoKid writes: Western Digital has announced its latest innovation in the solid state storage market with the pilot production of the world's first 512 Gigabit, three-bits-per-cell, 64-layer 3D NAND (BiCS 3D) chip — that sure is a mouthful. This pilot run is being ushered in with its technology partner, Toshiba, at the pair's massive Yokkaichi, Japan fabrication facility. By using vertical stacking of 64 layers, Western Digital is able to achieve a much larger storage density for its NAND in a smaller footprint. Stacking also reduces production costs and is more reliable than planar NAND solutions. According to Western Digital, it will begin mass production of its new 512Gb NAND during the second half of 2017. It was also reported late last month that Western Digital is very interested in purchasing a 20 percent stake in Toshiba's NAND flash business.