News Posts matching "MCM"

Return to Keyword Browsing

AMD 3rd Generation Ryzen Probable SKUs, Specs, Pricing Leaked?

One of our readers tipped us off with a very plausible looking image that drops a motherlode of information about what AMD's 2nd generation Ryzen (aka Ryzen 3000 series) processor lineup could look like. This includes a vast selection of SKUs, their CPU and iGPU core configurations, clock-speeds, and OEM channel pricing. The list speaks of a reentry for 7th generation A-series "Excavator" as Duron X4 series, followed by Duron 300GE-series based on a highly cut down "Raven Ridge," Athlon 300GE 2-core/4-thread based on an implausible "Zen+ 12 nm" APU die, followed by quad-core Ryzen 3 3000 series processors with and without iGPUs, making up the company's entry-level product lineup.

The core counts seem to jump from 4-core straight to 8-core, with no 6-core in between, for the Ryzen 5 series. This is also where AMD's new IP, the 7 nm "Zen 2" architecture, begins. There appears to be a large APU die (or a 3-chip MCM) with an 8-core CPU and 20-CU iGPU, which makes up certain Ryzen 5 SKUs. These chips are either 8-core/8-thread or 8-core/16-thread. The Ryzen 7 series is made up of 12-core/24-thread processors that are devoid of iGPU. The new Ryzen 9 series extension caps off the lineup with 16-core/32-thread SKUs. And these are just socket AM4.

Samsung AMD's Second Foundry Partner for "Polaris 30"

AMD's "Polaris 30" silicon at the heart of Radeon RX 590 graphics card is the company's first 12 nm GPU. Unlike NVIDIA, which is exclusively sourcing its "Turing" family of GPUs from TSMC, the "Polaris 30" is coming from not one, but two sources. This, according to AMD in response to a question by TechPowerUp. The two foundries manufacturing "Polaris 30" are GlobalFoundries and Samsung. AMD did not provide us with visual cues on how to tell chips made from either foundries apart (such as serial numbering schemes). Packaging of dies sourced from both foundries is done in China, and the national-origin marking for the chip is on the package, rather than printed on the die.

GlobalFoundries' 12 nanometer FinFET node, called GloFo 12LP, shares a lot of similarities with Samsung's 11LPP, because both are "nodelets" that are derived from an original 14 nm FinFET process blueprint Samsung licensed to GloFo, deployed in its facility in upstate New York, where AMD's "Zen" processors are made. GloFo's 12 nanometer process is a refinement of its 14 nm node, in which 12 nm transistors are etched onto silicon using the same lithography meant for 14 nm. It doesn't improve transistor densities, but provides dividends in power, which explains why "Polaris 30" and "Pinnacle Ridge" have the same die sizes as "Polaris 20" and "Summit Ridge," respectively. This WikiChip article provides a good explanation of how GloFo 12LP is a nodelet.

Intel Could Upstage EPYC "Rome" Launch with "Cascade Lake" Before Year-end

Intel is reportedly working tirelessly to launch its "Cascade Lake" Xeon Scalable 48-core enterprise processor before year-end, according to a launch window timeline slide leaked by datacenter hardware provider QCT. The slide suggests a late-Q4 thru Q1-2019 launch timeline for the XCC (extreme core count) version of "Cascade Lake," which packs 48 CPU cores across two dies on an MCM. This launch is part of QCT's "early shipment program," which means select enterprise customers can obtain the hardware in pre-approved quantities. In other words, this is a limited launch, but one that's probably enough to upstage AMD's 7 nm EPYC "Rome" 64-core processor launch.

It's only by late-Q1 thru Q2-2019 that the Xeon "Cascade Lake" family would be substantially launched, including lower core-count variants that are still 2-die MCMs. This aligns to preempt or match AMD's 7 nm EPYC family rollout through 2019. "Cascade Lake" is probably Intel's final enterprise microarchitecture to be built on the 14 nm++ node, and consists of 2-die multi-chip modules that feature 48 cores, and a 12-channel memory interface (6-channel per die); with 88-lane PCIe from the CPU socket. The processor is capable of multi-socket configurations. It will also be Intel's launch platform for substantially launching its Optane Persistent Memory product series.

AMD Zen 2 "Rome" MCM Pictured Up Close

Here is the clearest picture of AMD "Rome," codename for the company's next-generation EPYC socket SP3r2 processor, which is a multi-chip module of 9 chiplets (up from four). While first-generation EPYC MCMs (and Ryzen Threadripper) were essentially "4P-on-a-stick," the new "Rome" MCM takes the concept further, by introducing a new centralized uncore component called the I/O die. Up to eight 7 nm "Zen 2" CPU dies surround this large 14 nm die, and connect to it via substrate, using InfinityFabric, without needing a silicon interposer. Each CPU chiplet features 8 cores, and hence we have 64 cores in total.

The CPU dies themselves are significantly smaller than current-generation "Zeppelin" dies, although looking at their size, we're not sure if they're packing disabled integrated memory controllers or PCIe roots anymore. While the transition to 7 nm can be expected to significantly reduce die size, groups of two dies appear to be making up the die-area of a single "Zeppelin." It's possible that the CPU chiplets in "Rome" physically lack an integrated northbridge and southbridge, and only feature a broad InfinityFabric interface. The I/O die handles memory, PCIe, and southbridge functions, featuring an 8-channel DDR4 memory interface that's as monolithic as Intel's implementations, a PCI-Express gen 4.0 root-complex, and other I/O.

AMD Radeon Pro WX 8200 Detailed

AMD is giving final touches to the new Radeon Pro WX 8200 professional graphics card. Launched over a year following the Radeon Pro WX 9100, the WX 8200 has a leaner feature-set, while continuing to be based on AMD's current IP. The card still uses the 14 nm "Vega 10" MCM, but with a core-config akin to that of the RX Vega 56. You get 56 NGCUs working out to 3,584 stream processors, 224 TMUs, 64 ROPs, and a 4096-bit wide HBM2 memory interface, handling 16 GB of memory on this card. The air-cooled card also has a leaner connectivity load-out, with just four mini-DisplayPort 1.4 connectors (the WX 9100 comes with six). Its 8-pin + 6-pin connectors are positioned at the rear-end to be SSI workstation-friendly. The card surfaced on CompuBench database, in which its performance numbers are only slightly behind those of the WX 9100, but significantly better than the "Polaris 20" based WX 7100. AMD is expected to launch this card at SIGGRAPH 2018.

AMD 2nd Gen Ryzen Threadripper MCM Configuration Confirmed, More Details

Tech Day slides leaked to the web by Kitguru provide a confirmation of how AMD is wiring out the additional two dies on the 24-core and 32-core Ryzen Threadripper 2000-series MCMs on the socket TR4 platform. We had speculated that because the quad-channel DDR4 memory and PCIe interfaces are wired to two diagonally-opposite dies on AMD X399 chipset motherboards; in the interest of backwards compatibility, AMD could wire out memory and PCIe from just two out of four dies on the multi-chip module, and have the two additional dies seek memory and PCIe over the InfinityFabric interfaces.

The obvious trade-offs with this design choice is that latencies to the dies with indirect memory/PCIe access are higher, and that reflects heavily in AMD's own performance figures for comparing the 32-core 2990WX with the 16-core 2950X from the same generation. The 2990WX is "only" up to 64 percent faster than 2950X at Cinebench R15 nT, despite having double the number of cores. To its credit, the 2950X has higher clock-speeds (3.50 GHz nominal with 4.40 GHz boost) than the 2990WX (3.00 GHz nominal with 4.00 GHz boost). The presentation also puts out interesting bits of information such as AMD's own performance numbers showing 10-15 percent performance gains between the 2950X and the 1950X; and performance gains of the 2990WX over Intel Core i9-7980XE.

AMD Announces 2nd Generation Ryzen Threadripper 2000, up to 32 Cores/64 Threads!

AMD announced its second-generation Ryzen Threadripper high-end desktop (HEDT) processor series, succeeding its lean and successful first-generation that disrupted much of Intel's Core X HEDT series, forcing Intel to open up new high-core-count (HCC) market segments beyond its traditional $1000 price-point. AMD's 16-core $999 1950X proved competitive with even Intel's 12-core and 14-core SKUs priced well above the $1200-mark; and now AMD looks to beat Intel at its game, with the introduction of new 24-core and 32-core SKUs at prices that are sure to spell trouble for Intel's Core X HCC lineup. The lineup is partially open to pre-orders, with two SKUs launching within August (including the 32-core one), and two others in October.

At the heart of AMD's second-generation Ryzen Threadripper is the new 12 nm "Pinnacle Ridge" die, which made its debut with the 2nd Generation Ryzen AM4 family. This die proved to introduce 3-5 percent IPC improvements in single-threaded tasks, and multi-threaded improvements with an improved Precision Boost II algorithm, which boosted frequencies of each of 8 cores on-die. The Threadripper is still a multi-chip module, with 2 to 4 of these dies, depending on the SKU. There are four of these - the 12-core/24-thread Threadripper 2920X, the 16-core/32-thread Threadripper 2950X; the 24-core/48-thread Threadripper 2970WX, and the flagship 32-core/64-thread Threadripper 2990WX.

ASUS Gives Away Cooling Kits for its Socket TR4 Motherboards

To cope better with AMD 2nd generation Ryzen Threadripper processors, which come with up to four active dies on the MCM (multi-chip module), ASUS is giving away free Cooling Kits to owners of its socket TR4 motherboards, such as the ROG Zenith Extreme, ROG Strix X399-E Gaming, and Prime X399-A. The kits include a fan bracket that lets you strap a 40 mm fan onto your CPU VRM heatsink, and a so-called "SoC heatsink," designed to cool the SoC power phase MOSFETs (which now have to cope with the load of four SoC dies). The kit for the Zenith Extreme also includes a 10 mm-thick 40 mm fan, which plugs into one of your 4-pin PWM headers. The kit will neither be included with current or upcoming inventories of unsold X399 motherboards by ASUS. Customers who need it will have to contact their local ASUS office, which will verify the purchase of their ASUS X399 motherboard, and ship their kit for free.

On The Coming Chiplet Revolution and AMD's MCM Promise

With Moore's Law being pronounced as within its death throes, historic monolithic die designs are becoming increasingly expensive to manufacture. It's no secret that both AMD and NVIDIA have been exploring an MCM (Multi-Chip-Module) approach towards diverting from monolithic die designs over to a much more manageable, "chiplet" design. Essentially, AMD has achieved this in different ways with its Zen line of CPUs (two CPU modules of four cores each linked via the company's Infinity Fabric interconnect), and their own R9 and Vega graphics cards, which take another approach in packaging memory and the graphics processing die in the same silicon base - an interposer.

Intel Could Ditch AMD dGPU Die on Future Core G-series MCMs with "Arctic Sound"

Intel did the impossible in 2017, by collaborating with rival AMD after decades, on a product. The new Core i7-8000G series processors are multi-chip modules that combine quad-core "Kaby Lake" CPU dies with discrete AMD Radeon Vega GPU dies that have their own dedicated HBM2 stacks. With performance-segment notebooks and sleek AIO desktops building momentum for such products, Intel sees a future in building its own discrete GPUs, at least dies that can replace the AMD Radeon IP from its Core G-series processors.

With former AMD Graphics head Raja Koduri switching to Intel amidst rumors of the company investing in discrete GPUs of its own, details emerge of the company's future "Arctic Sound" and "Jupiter Sound" graphics IP, which point to the possibility of them being discrete GPU dies based on the Gen 12 and Gen 13 graphics architectures, respectively. According to Ashraf Eassa, a technology stock commentator with "The Motley Fool," both "Arctic Sound" and "Jupiter Sound" are discrete GPU dies that connect with Intel processor dies over EMIB, the company's proprietary high-density interconnect for multi-chip modules. It could be a long wait leading up to the two, since the company is still monetizing its Gen 9.5 architecture on 8th generation Core processors.

Intel Core i7-8809G "Kaby Lake + Vega" MCM Specs Leaked Again, Indicate Dual IGP

Intel revealed specifications of its upcoming "Kaby Lake + AMD Vega" multi-chip module, the Core i7-8809G, on its website. A number of these specs were already sniffed out by Futuremark SystemInfo, but the website sheds light on a key feature - dual integrated graphics. The specs sheet confirms that the chip combines a 4-core/8-thread "Kaby Lake" CPU die with an AMD Radeon RX Vega M GH graphics die. The CPU is clocked at 3.10 GHz, and SystemInfo (from the older story) confirmed that its Turbo Boost frequency is up to 3.90 GHz. The L3 cache amount is maxed out a 8 MB. The reference memory clock is set at dual-channel DDR4-2400. What's more, the CPU component features an unlocked base-clock multiplier.

Things get interesting with the way Intel describes its integrated graphics solution. It mentions both the star-attraction, the AMD Radeon RX Vega M GH, and the Intel HD Graphics 630 located on the "Kaby Lake" CPU die. This indicates that Intel could deploy a mixed multi-GPU solution that's transparent to software, balancing graphics loads between the HD 630 and RX Vega M GH, depending on the load and thermal conditions. Speaking of which, Intel has rated the TDP of the MCM at 100W, with a rider stating "target package TDP," since there's no scientifically-correct way of measuring TDP on a multi-chip module. Intel could build performance-segment NUCs with this chip, in addition to selling them to mini-PC manufacturers.

AMD Expands EPYC Availability, Introduces ROCm 1.7 With Tensor Flow Support

AMD has been steadily increasing output and availability of their latest take on the server market with their EPYC CPUs. These are 32-core, 64-thread monsters that excel in delivering a better feature set in 1P configuration than even some of Intel's 2P setups, and reception for these AMD processors has been pretty warm as a result. The usage of an MCM design to create a 4-way cluster of small 8-core processor packages has allowed AMD to improve yields with minimum retooling and changes to its manufacturing lines, which in turn, has increased yields and profits for a company that sorely needed a a breakout product.

Intel NUC Based on Intel+Vega MCM Leaked

The first product based on Intel's ambitious "Kaby Lake-G" multi-chip module, which combines a quad-core "Kaby Lake-H" die with a graphics die based on AMD "Vega" architecture, will be a NUC (next unit of computing), and likely the spiritual successor to Intel's "Skull Canyon" NUC. The first picture of the motherboard of this NUC was leaked to the web, revealing a board that's only slightly smaller than the mini-ITX form-factor.

The board draws power from an external power brick, and appears to feature two distinct VRM areas for the CPU and GPU components of the "Kaby Lake-G" MCM SoC. The board feature two DDR4 SO-DIMM slots which are populated with dual-channel memory, and an M.2 NVMe slot, holding an SSD. There are two additional SATA 6 Gb/s ports, besides a plethora of other connectivity options.

Intel, AMD MCM Core i7 Design Specs, Benchmarks Leaked

Following today's surprise announcement of an Intel-AMD collaboration (of which NVIDIA seems to be the only company left in a somewhat more fragile position), there have already been a number of benchmark leaks for the new Intel + AMD devices. While Intel's original announcement was cryptic enough - to be expected, given the nature of the product and the ETA before its arrival to market - some details are already pouring out into the world wide web.

The new Intel products are expected to carry the "Kaby Lake G" codename, where the G goes hand in hand with the much increased graphics power of these solutions compared to other less exotic ones - meaning, not packing AMD Radeon graphics. For now, the known product names point to one Intel Core i7-8705G and Intel Core i7-8809G. Board names for these are 694E:C0 and 694C:C0, respectively.
Return to Keyword Browsing