Schlagwort: Samsung

Samsung Electronics Develops Second-Generation SmartSSD Computational Storage Drive With Upgraded Processing Functionality

Cuts pro­ces­sing time by over 50%, ener­gy con­sump­ti­on by up to 70% and CPU uti­liz­a­ti­on by up to 97% com­pa­red to con­ven­tio­nal SSD drives

Sam­sung Elec­tro­nics, the world lea­der in advan­ced memo­ry tech­no­lo­gy, today announ­ced that it has suc­cess­ful­ly deve­lo­ped a second genera­ti­on of its pionee­ring SmartSSD.

The new pro­prie­ta­ry com­pu­ta­tio­nal sto­rage incor­po­ra­tes data pro­ces­sing func­tio­n­a­li­ty wit­hin a high-per­for­mance SSD. Unli­ke exis­ting SSDs, Samsung’s SmartS­SD can pro­cess data direct­ly, ther­eby mini­mi­zing data trans­fers bet­ween the CPU, GPU and RAM. This tech­no­lo­gy can avoid the bot­t­len­ecks that often occur when moving data bet­ween sto­rage devices and CPUs, resul­ting in mar­ked­ly impro­ved sys­tem per­for­mance and much hig­her ener­gy effi­ci­en­cy. (…) Wei­ter­le­sen »

Samsung Introduces Game Changing Exynos 2200 Processor With Xclipse GPU Powered By AMD RDNA 2 Architecture

The new pre­mi­um mobi­le pro­ces­sor comes with hard­ware-acce­le­ra­ted ray tra­cing and sta­te of the art Arm-based pro­ces­sing technology

Sam­sung Elec­tro­nics, a world lea­der in advan­ced semi­con­duc­tor tech­no­lo­gy, today announ­ced its new pre­mi­um mobi­le pro­ces­sor, the Exy­nos 2200. The Exy­nos 2200 is a fresh­ly desi­gned mobi­le pro­ces­sor with a power­ful AMD RDNA 2 archi­tec­tu­re based Sam­sung Xclip­se gra­phics pro­ces­sing unit (GPU). With the most cut­ting-edge Arm®-based CPU cores avail­ab­le in the mar­ket today and an upgraded neural pro­ces­sing unit (NPU), the Exy­nos 2200 will enab­le the ulti­ma­te mobi­le pho­ne gaming expe­ri­ence, as well as enhan­cing the over­all expe­ri­ence in social media apps and photography.

Built on the most advan­ced 4‑nanometer (nm) EUV (extre­me ultra­vio­let litho­gra­phy) pro­cess, and com­bi­ned with cut­ting-edge mobi­le, GPU and NPU tech­no­lo­gy, Sam­sung has craf­ted the Exy­nos 2200 to pro­vi­de the finest expe­ri­ence for smart­pho­ne users. With the Xclip­se, our new mobi­le GPU built with RDNA 2 gra­phics tech­no­lo­gy from the indus­try lea­der AMD, the Exy­nos 2200 will rede­fi­ne mobi­le gaming expe­ri­ence, aided by enhan­ced gra­phics and AI per­for­mance,” said Yon­gin Park, Pre­si­dent of Sys­tem LSI Busi­ness at Sam­sung Elec­tro­nics. “As well as brin­ging the best mobi­le expe­ri­ence to the users, Sam­sung will con­ti­nue its efforts to lead the jour­ney in logic chip inno­va­ti­on.” (…) Wei­ter­le­sen »

Samsung Develops Industry’s First LPDDR5X DRAM

Samsung’s LPDDR5X DRAM will provide over 1.3x faster processing speeds and consumer nearly 20% less power than the previous LPDDR5 solution

The LPDDR5X solution will broaden the use of high-performance, low-power memory beyond smartphones to AI and edge applications

 

Sam­sung Elec­tro­nics Co., Ltd., the world lea­der in advan­ced memo­ry tech­no­lo­gy, today announ­ced that it has deve­lo­ped the industry’s first 14-nano­me­ter (nm) based 16-giga­bit (Gb) Low Power Dou­ble Data Rate 5X (LPDDR5X) DRAM, desi­gned to dri­ve fur­ther growth throughout the high-speed data ser­vice app­li­ca­ti­ons inclu­ding 5G, arti­fi­cial intel­li­gence (AI) and the metaverse.
(…) Wei­ter­le­sen »

Samsung Develops Industry’s First HKMG-Based DDR5 Memory; Ideal for Bandwidth-Intensive Advanced Computing Applications

512GB capa­ci­ty DDR5 modu­le made pos­si­ble by an 8‑layer TSV structure
HKMG mate­ri­al redu­ces power by 13 per­cent while doub­ling the speed of DDR4

Sam­sung Elec­tro­nics, the world lea­der in advan­ced memo­ry tech­no­lo­gy, today announ­ced that it has expan­ded its DDR5 DRAM memo­ry port­fo­lio with the industry’s first 512GB DDR5 modu­le based on High‑K Metal Gate (HKMG) pro­cess tech­no­lo­gy. Deli­vering more than twice the per­for­mance of DDR4 at up to 7,200 mega­bits per second (Mbps), the new DDR5 will be capa­ble of orches­tra­ting the most extre­me com­pu­te-hungry, high-band­width workloads in super­com­pu­ting, arti­fi­cial intel­li­gence (AI) and machi­ne lear­ning (ML), as well as data ana­ly­tics app­li­ca­ti­ons. (…) Wei­ter­le­sen »

Samsung Develops Industry’s First High Bandwidth Memory with AI Processing Power

Korea on Febru­a­ry 17, 2021

The new archi­tec­tu­re will deli­ver over twice the sys­tem performance
and redu­ce ener­gy con­sump­ti­on by more than 70%

Sam­sung Elec­tro­nics, the world lea­der in advan­ced memo­ry tech­no­lo­gy, today announ­ced that it has deve­lo­ped the industry’s first High Band­width Memo­ry (HBM) inte­gra­ted with arti­fi­cial intel­li­gence (AI) pro­ces­sing power — the HBM-PIM. The new pro­ces­sing-in-memo­ry (PIM) archi­tec­tu­re brings power­ful AI com­pu­ting capa­bi­li­ties insi­de high-per­for­mance memo­ry, to acce­le­ra­te lar­ge-sca­le pro­ces­sing in data cen­ters, high per­for­mance com­pu­ting (HPC) sys­tems and AI-enab­led mobi­le app­li­ca­ti­ons. (…) Wei­ter­le­sen »

Samsung angeblich mit Yield-Problemen bei 8‑nm — wechselt Nvidia zurück zu TSMC?

Nach­dem Nvi­dia vor­ges­tern Lie­fer­eng­päs­se bei der GeFor­ce RTX 3080 und 3090 bis hin­ein ins nächs­te Jahr auf Grund einer enor­men Nach­fra­ge bestä­tig­te (Tom’s Hard­ware), führt ein Bericht der chi­ne­si­schen Digi­Ti­mes nun an, dass Sam­sung bei der 8‑nm-Fer­ti­gung angeb­lich Pro­ble­me mit der Aus­beu­te an funk­ti­ons­fä­hi­gen Chips (Yield) hat und man bei Nvi­dia nun wohl über einen Wech­sel zu TSMC nach­den­ken würde.
(…) Wei­ter­le­sen »

Samsung Announces Availability of its Silicon-Proven 3D IC Technology for High-Performance Applications

Sam­sung ‘X‑Cube’ enab­les indus­try-first 3D SRAM-logic working sili­con at 7nm and bey­ond. Band­width and den­si­ty can be sca­led to suit diver­se design requi­re­ments in emer­ging app­li­ca­ti­ons. Korea on August 13, 2020 — Sam­sung Elec­tro­nics, a world lea­der in advan­ced semi­con­duc­tor tech­no­lo­gy, today announ­ced the immedia­te avai­la­bi­li­ty of its sili­­con-pro­­­ven 3D IC pack­a­ging tech­no­lo­gy, eXten­­ded-Cube (X‑Cube), (…) Wei­ter­le­sen »

Globalfoundries vor Expansion und Beteiligung durch Samsung?

Über Glo­bal­found­ries wur­de bei uns schon län­ger nicht mehr berich­tet, nach­dem AMD immer mehr Chips beim Kon­kur­ren­ten TSMC fer­ti­gen lässt. Die ehe­ma­li­gen Fabri­ken von AMD und IBM und fer­ti­gen heu­te vor allem Chips für AI und Auto­mo­ti­ve-Kun­den. Nach dem 14-nm-Pro­zess wur­de die feins­te Fer­ti­gungs­stu­fe mit 12 nm schon von Sam­sung lizen­siert. Die geplan­te Fer­ti­gung mit einer Struk­tur­brei­te von 7 nm gestri­chen. Damit ver­lor Glo­bal­found­ries vor allem Kun­den in der Pro­zes­sor­fer­ti­gung an den Kon­kur­ren­ten TSMC. (…) Wei­ter­le­sen »

Samsung Electronics Begins Mass Production at New EUV Manufacturing Line

Korea on Febru­a­ry 20, 2020   Samsung’s EUV capa­ci­ty under 7nm will trip­le by end of 2020 To start ship­ping first 7 and 6nm-based mobi­le chips from V1 line in 1Q Sam­sung Elec­tro­nics, a world lea­der in advan­ced semi­con­duc­tor tech­no­lo­gy, today announ­ced that its new cut­­ting-edge semi­con­duc­tor fab­ri­ca­ti­on line in Hwa­se­ong, Korea, has begun mass pro­duc­tion.   The faci­li­ty, (…) Wei­ter­le­sen »

Samsung to Advance High Performance Computing Systems with Launch of Industry’s First 3rd-generation (16GB) HBM2E

New HBM2E stacks eight 16Gb DRAM dies to achie­ve 16GB packa­ge capa­ci­ty and ensu­res a sta­ble data trans­fer speed at 3.2Gbps Sam­sung Elec­tro­nics, the world lea­der in advan­ced memo­ry tech­no­lo­gy, today announ­ced the mar­ket launch of ‘Flash­bolt’, its third-gene­r­a­­ti­on High Band­width Memo­ry 2E (HBM2E). The new 16-giga­­by­te (GB) HBM2E is uni­que­ly sui­ted to maxi­mi­ze high (…) Wei­ter­le­sen »

Samsung’s PM1733 SSD and High Density DIMMs Support AMD EPYC™ 7002 Series Processor

USA on August 9, 2019   Sam­sung Elec­tro­nics has taken its lea­ders­hip posi­ti­on in the memo­ry mar­ket a step fur­ther today by announ­cing sup­port of the Sam­sung PM1733 PCIe Gen4 Solid Sta­te Dri­ve (SSD) and high den­si­ty RDIMM1 and LRDIMM2 dyna­mic ran­dom access memo­ry (DRAM) for the AMD EPYC™ 7002 Genera­ti­on Pro­ces­sors. AMD laun­ched the 2nd Gen AMD EPYC™ pro­ces­sor (…) Wei­ter­le­sen »

Samsung will in zwei Jahren Produkte auf Basis von SoCs mit integrierter AMD RDNA-Grafik anbieten

Nach­dem Sam­sung und AMD bereits im Juni eine stra­te­gi­sche Part­ner­schaft ver­kün­det hat­ten, in der ver­ein­bart wur­de, dass Sam­sung AMDs Rade­on-DNA-Archi­tek­tur zur Ver­wen­dung in Mobil­ge­rä­ten lizen­siert, gab Sam­sung in der letz­ten Wochen im Rah­men der Prä­sen­ta­ti­on der Zah­len des zwei­ten Quar­tals einen Zeit­rah­men für das Erschei­nen ers­ter Pro­duk­te an. (…) Wei­ter­le­sen »

AMD and Samsung Announce Strategic Partnership in Ultra Low Power, High Performance Graphics Technologies

Sam­sung to inte­gra­te cus­tom AMD Rade­on™ gra­phics IP into future SoCs for mobi­le app­li­ca­ti­ons SEOUL, South Korea & SANTA CLARA, Calif.  06/03/2019 AMD (NASDAQ: AMD) and Sam­sung Elec­tro­nics Co., Ltd. today announ­ced a mul­­ti-year stra­te­gic part­ners­hip in ultra low power, high per­for­mance mobi­le gra­phics IP based on AMD Rade­on gra­phics tech­no­lo­gies. As part of the part­ners­hip, Sam­sung will licen­se (…) Wei­ter­le­sen »

Intel: Keine Ryzen-Konkurrenz in 10 nm vor 2022

Intels 10-nm-Desas­ter wird immer mas­si­ver. Eigent­lich soll­ten Intels Fabs in dem Fer­ti­gungs­pro­zess schon seit 2015 flei­ßig Pro­zes­so­ren aus­spu­cken, tat­säch­lich fer­ti­gen sie seit letz­tem Jahr ers­te Volu­men­ein­hei­ten, die aber mit defek­ter Gra­fik­ein­heit auf den asia­ti­schen Markt gebracht wur­den. AMD wird ver­mut­lich Mit­te die­ses Jah­res ers­te Pro­zes­so­ren in 7 nm Struk­tur­brei­te auf den Markt brin­gen, wobei TSMCs 7 nm wohl ver­gleich­bar fein wie Intels 10 nm sind. Damit hät­te AMD einen mas­si­ven Tech­no­lo­gie-Vor­sprung für die kom­men­den Jah­re. (…) Wei­ter­le­sen »

Samsung Electronics Introduces New High Bandwidth Memory Technology Tailored to Data Centers, Graphic Applications, and AI

SAN JOSE, Calif.–Samsung Elec­tro­nics Co., Ltd., the world lea­der in advan­ced semi­con­duc­tor tech­no­lo­gy, today announ­ced its new High Band­width Memo­ry (HBM2E) pro­duct at NVIDIA’s GPU Tech­no­lo­gy Con­fe­rence (GTC) to deli­ver the hig­hest DRAM per­for­mance levels for use in next-gene­r­a­­ti­on super­com­pu­ters, gra­phics sys­tems, and arti­fi­cial intel­li­gence (AI). The new solu­ti­on, Flash­bolt™, is the industry’s first HBM2E (…) Wei­ter­le­sen »