AMD unveils next-gene­ra­ti­on hard­ware and soft­ware road­maps, expan­ded pro­duct port­fo­lio addres­sing new mar­kets, and stra­te­gies to acce­le­ra­te data cen­ter growth and deli­ver per­va­si­ve AI leadership

SANTA CLARA, Calif., June 09, 2022 (GLOBE NEWSWIRE) — Today at its Finan­cial Ana­lyst Day, AMD (NASDAQ: AMD) out­lined its stra­tegy to deli­ver its next pha­se of growth dri­ven by the company’s expan­ded port­fo­lio of high-per­for­mance and adap­ti­ve com­pu­ting pro­ducts span­ning the data cen­ter, embedded, cli­ent, and gam­ing markets.

“From the cloud and PCs to com­mu­ni­ca­ti­ons and intel­li­gent end­points, AMD’s high-per­for­mance and adap­ti­ve com­pu­ting solu­ti­ons play an incre­asing­ly lar­ger role in sha­ping the capa­bi­li­ties of near­ly every ser­vice and pro­duct defi­ning the future of com­pu­ting today,” said Dr. Lisa Su, AMD chair and CEO. “The clo­se of our trans­for­ma­tio­nal acqui­si­ti­on of Xilinx and our expan­ded port­fo­lio of lea­der­ship com­pu­te engi­nes pro­vi­de AMD with signi­fi­cant oppor­tu­ni­ties to deli­ver con­tin­ued strong reve­nue growth with com­pel­ling share­hol­der returns as we cap­tu­re a lar­ger share of the diver­se $300 bil­li­on mar­ket for our high-per­for­mance and adap­ti­ve products.”

Tech­no­lo­gy and Pro­duct Port­fo­lio Updates
AMD announ­ced expan­ded mul­ti-gene­ra­tio­nal CPU core, gra­phics, and adap­ti­ve com­pu­ting archi­tec­tu­re road­maps inclu­ding new details on the:

• “Zen 4” CPU core expec­ted to power the world’s first high-per­for­mance 5nm x86 CPUs later this year. “Zen 4” is expec­ted to increase IPC 8%-10%¹ and deli­ver more than a 25% increase in per­for­mance-per-watt² and 35% over­all per­for­mance increase com­pared to “Zen 3” when run­ning desk­top appli­ca­ti­ons³.
• “Zen 5” CPU core plan­ned for 2024, which is built from the ground up to deli­ver per­for­mance and effi­ci­en­cy lea­der­ship across a broad ran­ge of workloads and fea­tures and includes opti­miza­ti­ons for AI and machi­ne learning.
• AMD RDNA™ 3 gam­ing archi­tec­tu­re that com­bi­nes a chip­let design, next gene­ra­ti­on AMD Infi­ni­ty Cache™ tech­no­lo­gy, lea­ding-edge 5nm manu­fac­tu­ring tech­no­lo­gy, and other enhance­ments to deli­ver more than 50% grea­ter per­for­mance-per-watt com­pared to the pri­or gene­ra­ti­on⁴.
• 4^th Gen Infi­ni­ty Archi­tec­tu­re that fur­ther extends AMD’s lea­der­ship modu­lar SoC design approach with a high-speed inter­con­nect, allo­wing seam­less inte­gra­ti­on of both AMD IP and 3^rd par­ty chip­lets to enable an enti­re­ly new class of high-per­for­mance and adap­ti­ve pro­ces­sors and pro­vi­ding a cus­tom-rea­dy hete­ro­ge­nous com­pu­ting platform.
• AMD CDNA™ 3 archi­tec­tu­re, which com­bi­nes 5nm chip­lets, 3D die stack­ing, 4^th gene­ra­ti­on Infi­ni­ty Archi­tec­tu­re, next-gene­ra­ti­on AMD Infi­ni­ty Cache™ tech­no­lo­gy, and HBM memo­ry in a sin­gle packa­ge with a uni­fied memo­ry pro­gramming model. The first AMD CDNA 3 archi­tec­tu­re-based pro­ducts are plan­ned for 2023 and are expec­ted to deli­ver more than 5X grea­ter per­for­mance-per-watt com­pared to AMD CDNA 2 archi­tec­tu­re on AI trai­ning workloads⁵.
• AMD XDNA, the foun­da­tio­nal archi­tec­tu­re IP from Xilinx that con­sists of key tech­no­lo­gies inclu­ding the FPGA fabric and AI Engi­ne (AIE). The FPGA fabric com­bi­nes an adap­ti­ve inter­con­nect with FPGA logic and local memo­ry, while the AIE pro­vi­des a dataf­low archi­tec­tu­re opti­mi­zed for high per­for­mance and ener­gy effi­ci­ent AI and signal pro­ces­sing appli­ca­ti­ons. AMD plans to inte­gra­te AMD XDNA IP across mul­ti­ple pro­ducts in the future, start­ing with AMD Ryzen™ pro­ces­sors plan­ned for 2023.

Expan­ded Data Cen­ter Solu­ti­ons Portfolio
AMD reve­a­led an expan­ded port­fo­lio of high-per­for­mance, next-gene­ra­ti­on CPUs, acce­le­ra­tors, data pro­ces­sing units (DPUs), and adap­ti­ve com­pu­ting pro­ducts opti­mi­zed for mul­ti­ple workloads, including:

• 4^th Gen AMD EPYC™ pro­ces­sors powered by “Zen 4” and “Zen 4c” cores. 
  • “Gen­oa” powered by “Zen 4”: On-track to launch in Q4 2022 as the hig­hest per­for­mance gene­ral pur­po­se ser­ver pro­ces­sor available, with the top-of-stack pro­duct deli­ve­ring grea­ter than 75% fas­ter enter­pri­se Java® per­for­mance com­pared to top-of-stack 3^rd Gen EPYC pro­ces­sors⁶.
  • “Ber­ga­mo” powered by “Zen 4c”: Expec­ted to be the hig­hest per­for­mance ser­ver pro­ces­sors for cloud nati­ve com­pu­ting, offe­ring more than dou­ble the con­tai­ner den­si­ty of 3^rd Gen EPYC pro­ces­sors at their launch plan­ned for the first half of 2023⁷.
  • “Genoa‑X” powered by “Zen 4”: An opti­mi­zed ver­si­on of 4^th Gen EPYC pro­ces­sors with AMD 3D V‑Cache™ tech­no­lo­gy to enable lea­der­ship per­for­mance in rela­tio­nal data­ba­se and tech­ni­cal com­pu­ting workloads⁸.
  • “Sie­na” powered by “Zen 4”: The first AMD EPYC pro­ces­sor opti­mi­zed for intel­li­gent edge and com­mu­ni­ca­ti­ons deploy­ments that requi­re hig­her com­pu­te den­si­ties in a cost and power opti­mi­zed platform.
• AMD Instinct™ MI300 acce­le­ra­tors, the world’s first data cen­ter APUs, expec­ted to deli­ver a grea­ter than 8x increase in AI trai­ning per­for­mance com­pared to the AMD Instinct MI200 acce­le­ra­tor⁹. MI300 acce­le­ra­tors levera­ge a ground­brea­king 3D chip­let design com­bi­ning AMD CDNA 3 GPU, “Zen 4” CPU, cache memo­ry, and HBM chip­lets that are desi­gned to pro­vi­de lea­der­ship memo­ry band­width and appli­ca­ti­on laten­cy for AI trai­ning and HPC workloads.
• AMD Pen­san­do DPUs that com­bi­ne a robust soft­ware stack with “zero trust secu­ri­ty” throug­hout and an indus­try-lea­ding packet pro­ces­sor to crea­te the world’s most intel­li­gent and per­for­mant DPU, which is alre­a­dy deploy­ed at sca­le across cloud and enter­pri­se customers.
• Alveo™ Smart­NICs deploy­ed by hypers­ca­le cus­to­mers to acce­le­ra­te cus­tom workloads and extend con­fi­den­ti­al com­pu­ting to the net­wor­king interface.

Acce­le­ra­ting Lea­der­ship in Per­va­si­ve AI
AMD is uni­que­ly posi­tio­ned with its broad pro­duct port­fo­lio and expe­ri­ence ser­ving diver­se embedded mar­kets to help cus­to­mers deve­lop and deploy appli­ca­ti­ons with mul­ti­ple forms of AI.

The trans­for­ma­ti­ve acqui­si­ti­on of Xilinx pro­vi­des AMD with an unmat­ched set of hard­ware and soft­ware capa­bi­li­ties, inte­gra­ting the lea­der­ship Xilinx AI Engi­ne (AIE) across AMD Ryzen, AMD EPYC and Xilinx Ver­sal™ pro­ducts for small and mid-size AI models to com­ple­ment next-gene­ra­ti­on AMD Instinct acce­le­ra­tors and adap­ti­ve SoCs, enab­ling lea­der­ship per­for­mance on sca­le-out trai­ning and infe­rence workloads.

To uni­fy AI pro­gramming tools, AMD also announ­ced a mul­ti-gene­ra­ti­on Uni­fied AI Soft­ware road­map that will allow AI deve­lo­pers to pro­gram across its CPU, GPU, and Adap­ti­ve SoC pro­duct port­fo­lio from machi­ne lear­ning (ML) frame­works with the same set of tools and pre-opti­mi­zed models.

Expan­ding PC Leadership
AMD show­ca­sed its lea­der­ship in the glo­bal PC mar­ket, detail­ing how it con­ti­nues to deepen OEM part­ner­ships and dri­ve con­tin­ued growth across pre­mi­um, gam­ing, and com­mer­cial mar­kets, and pro­vi­ded a pre­view of its cli­ent road­map over the next seve­ral years, including:

• The “Phoe­nix Point” mobi­le pro­ces­sor plan­ned for 2023 will bring tog­e­ther the AMD “Zen 4” core archi­tec­tu­re with AMD RDNA 3 gra­phics archi­tec­tu­re and AIE, fol­lo­wed by the “Strix Point” pro­ces­sor plan­ned for 2024. “Phoe­nix Point” inno­va­tions include the AIE infe­rence acce­le­ra­tor, image signal pro­ces­sor, advan­ced dis­play for refresh and respon­se, AMD chip­let archi­tec­tu­re, and extre­me power management.
• The “Zen 4”-based Ryzen 7000 Series desk­top pro­ces­sors, which are expec­ted to deli­ver fas­ter clock speeds and grea­ter sin­gle and mul­ti-threa­ded per­for­mance com­pared to Ryzen 6000 pro­ces­sors¹⁰, will be fol­lo­wed by “Zen 5”-based “Gra­ni­te Ridge” processors.

Dri­ving Gra­phics Momentum
AMD announ­ced the latest deve­lo­p­ments desi­gned to con­ti­nue brin­ging world-class gra­phics solu­ti­ons to cus­to­mers world­wi­de, including:

• “Navi 3x” pro­ducts are expec­ted to launch later this year, built on the next-gene­ra­ti­on AMD RDNA 3 gam­ing architecture.
• More than 50 new gam­ing PC plat­forms are expec­ted to launch in 2022, ele­vat­ing gam­ing to new levels of per­for­mance and visu­al fide­li­ty by com­bi­ning AMD Rade­on™ RX Series gra­phics with AMD Ryzen processors.
• AMD expan­ded its lea­der­ship posi­ti­on in the gam­ing con­so­le space with the addi­ti­on of Valve’s Steam Deck™ gam­ing hand­held, powered by AMD “Zen 2” archi­tec­tu­re- based pro­ces­sors and AMD RDNA 2 archi­tec­tu­re-based graphics.
• New growth oppor­tu­ni­ties in 2022 and bey­ond, inclu­ding pro­vi­ding a ran­ge of gra­phics tech­no­lo­gies to acce­le­ra­te next-gene­ra­ti­on meta­ver­se appli­ca­ti­ons, ran­ging from 3D con­tent crea­ti­on bey­ond games and movies, to cloud gam­ing and inter­ac­ti­vi­ty within meta­ver­se environments.

New Finan­cial Report­ing Segments
Start­ing with second quar­ter 2022 results, AMD is updating its finan­cial report­ing seg­ments to ali­gn with its stra­te­gic end markets:

• Data Cen­ter: Inclu­ding ser­ver CPUs, data cen­ter GPUs, and the por­ti­ons of Xilinx reve­nue rela­ted to the data cen­ter business
• Embedded: Inclu­ding the Xilinx embedded busi­ness plus the AMD embedded business
• Cli­ent: Inclu­ding the tra­di­tio­nal desk­top and note­book PC business
• Gam­ing: Inclu­ding the dis­crete gra­phics gam­ing busi­ness and the semi-cus­tom game con­so­le business.

Tog­e­ther We Advance
To com­ple­ment the evo­lu­ti­on of its stra­te­gic end mar­kets and lea­der­ship pro­duct port­fo­lio, AMD also show­ca­sed a new evo­lu­ti­on of its brand. The new brand plat­form, “tog­e­ther we advance_” demons­tra­tes how, tog­e­ther with its part­ners, cus­to­mers, and employees, AMD is advan­cing inno­va­ti­on to crea­te solu­ti­ons to the world’s toug­hest chal­lenges. The new brand cam­paign is the lar­gest in AMD histo­ry and sees the AMD arrow mark beco­ming more pro­mi­nent in all com­mu­ni­ca­ti­ons assets, illus­t­ra­ting how per­va­si­ve AMD tech­no­lo­gy is by powe­ring lar­ge, diver­se, and gro­wing markets.

Sup­port­ing Resources

• Watch the event replay
• Access event pre­sen­ta­ti­ons
• Read more about AMD brand updates
• Fol­low AMD on Twit­ter

About AMD
For more than 50 years AMD has dri­ven inno­va­ti­on in high-per­for­mance com­pu­ting, gra­phics, and visua­liza­ti­on tech­no­lo­gies. AMD employees are focu­sed on buil­ding lea­der­ship high-per­for­mance and adap­ti­ve pro­ducts that push the boun­da­ries of what is pos­si­ble. Bil­li­ons of peo­p­le, lea­ding For­tu­ne 500 busi­nesses, and cut­ting-edge sci­en­ti­fic rese­arch insti­tu­ti­ons around the world rely on AMD tech­no­lo­gy dai­ly to impro­ve how they live, work, and play. For more infor­ma­ti­on about how AMD is enab­ling today and inspi­ring tomor­row, visit the AMD (NASDAQ: AMD) web­site, blog, Lin­ke­dIn, and Twit­ter pages. 

Cau­tio­na­ry Statement
This press release con­ta­ins for­ward-loo­king state­ments con­cer­ning Advan­ced Micro Devices, Inc. (AMD) such as the fea­tures, func­tion­a­li­ty, per­for­mance, avai­la­bi­li­ty, timing and expec­ted bene­fits of AMD pro­ducts; AMD’s next pha­se of growth; AMD’s signi­fi­cant oppor­tu­ni­ty to deli­ver strong growth and share­hol­der returns over the next five years; expec­ted bene­fits of AMD’s acqui­si­ti­on of Pen­san­do Sys­tems; and AMD’s new growth oppor­tu­ni­ties in 2022 and bey­ond, which are made pur­su­ant to the Safe Har­bor pro­vi­si­ons of the Pri­va­te Secu­ri­ties Liti­ga­ti­on Reform Act of 1995. For­ward-loo­king state­ments are com­mon­ly iden­ti­fied by words such as “would,” “may,” “expects,” “belie­ves,” “plans,” “intends,” “pro­jects” and other terms with simi­lar mea­ning. Inves­tors are cau­tio­ned that the for­ward-loo­king state­ments in this press release are based on cur­rent beliefs, assump­ti­ons and expec­ta­ti­ons, speak only as of the date of this press release and invol­ve risks and uncer­tain­ties that could cau­se actu­al results to dif­fer mate­ri­al­ly from cur­rent expec­ta­ti­ons. Such state­ments are sub­ject to cer­tain known and unknown risks and uncer­tain­ties, many of which are dif­fi­cult to pre­dict and gene­ral­ly bey­ond AMD’s con­trol, that could cau­se actu­al results and other future events to dif­fer mate­ri­al­ly from tho­se expres­sed in, or impli­ed or pro­jec­ted by, the for­ward-loo­king infor­ma­ti­on and state­ments. Mate­ri­al fac­tors that could cau­se actu­al results to dif­fer mate­ri­al­ly from cur­rent expec­ta­ti­ons include, wit­hout limi­ta­ti­on, the fol­lo­wing: Intel Corporation’s domi­nan­ce of the micro­pro­ces­sor mar­ket and its aggres­si­ve busi­ness prac­ti­ces; glo­bal eco­no­mic uncer­tain­ty; loss of a signi­fi­cant cus­to­mer; impact of the COVID-19 pan­de­mic on AMD’s busi­ness, finan­cial con­di­ti­on and results of ope­ra­ti­ons; com­pe­ti­ti­ve mar­kets in which AMD’s pro­ducts are sold; mar­ket con­di­ti­ons of the indus­tries in which AMD pro­ducts are sold; cycli­cal natu­re of the semi­con­duc­tor indus­try; quar­ter­ly and sea­so­nal sales pat­terns; AMD’s abili­ty to ade­qua­te­ly pro­tect its tech­no­lo­gy or other intellec­tu­al pro­per­ty; unfa­vorable cur­ren­cy exch­an­ge rate fluc­tua­tions; abili­ty of third par­ty manu­fac­tu­r­ers to manu­fac­tu­re AMD’s pro­ducts on a time­ly basis in suf­fi­ci­ent quan­ti­ties and using com­pe­ti­ti­ve tech­no­lo­gies; avai­la­bi­li­ty of essen­ti­al equip­ment, mate­ri­als, sub­stra­tes or manu­fac­tu­ring pro­ces­ses; abili­ty to achie­ve expec­ted manu­fac­tu­ring yields for AMD’s pro­ducts; AMD’s abili­ty to intro­du­ce pro­ducts on a time­ly basis with expec­ted fea­tures and per­for­mance levels; AMD’s abili­ty to gene­ra­te reve­nue from its semi-cus­tom SoC pro­ducts; poten­ti­al secu­ri­ty vul­nerabi­li­ties; poten­ti­al secu­ri­ty inci­dents inclu­ding IT outa­ges, data loss, data brea­ches and cyber-attacks; uncer­tain­ties invol­ving the orde­ring and ship­ment of AMD’s pro­ducts; AMD’s reli­ance on third-par­ty intellec­tu­al pro­per­ty to design and intro­du­ce new pro­ducts in a time­ly man­ner; AMD’s reli­ance on third-par­ty com­pa­nies for design, manu­fac­tu­re and sup­p­ly of mother­boards, soft­ware and other com­pu­ter plat­form com­pon­ents; AMD’s reli­ance on Micro­soft and other soft­ware ven­dors’ sup­port to design and deve­lop soft­ware to run on AMD’s pro­ducts; AMD’s reli­ance on third-par­ty dis­tri­bu­tors and add-in-board part­ners; impact of modi­fi­ca­ti­on or inter­rup­ti­on of AMD’s inter­nal busi­ness pro­ces­ses and infor­ma­ti­on sys­tems; com­pa­ti­bi­li­ty of AMD’s pro­ducts with some or all indus­try-stan­dard soft­ware and hard­ware; cos­ts rela­ted to defec­ti­ve pro­ducts; effi­ci­en­cy of AMD’s sup­p­ly chain; AMD’s abili­ty to rely on third par­ty sup­p­ly-chain logi­stics func­tions; AMD’s abili­ty to effec­tively con­trol sales of its pro­ducts on the gray mar­ket; impact of govern­ment actions and regu­la­ti­ons such as export admi­nis­tra­ti­on regu­la­ti­ons, tariffs and trade pro­tec­tion mea­su­res; AMD’s abili­ty to rea­li­ze its defer­red tax assets; poten­ti­al tax lia­bi­li­ties; cur­rent and future claims and liti­ga­ti­on; impact of envi­ron­men­tal laws, con­flict mine­rals-rela­ted pro­vi­si­ons and other laws or regu­la­ti­ons; impact of acqui­si­ti­ons, joint ven­tures and/or invest­ments on AMD’s busi­ness, and abili­ty to inte­gra­te acqui­red busi­nesses, such as Xilinx; impact of any impair­ment of the com­bi­ned company’s assets on the com­bi­ned company’s finan­cial posi­ti­on and results of ope­ra­ti­on; rest­ric­tions impo­sed by agree­ments gover­ning AMD’s notes, the gua­ran­tees of Xilinx’s notes and the revol­ving cre­dit faci­li­ty; AMD’s indeb­ted­ness; AMD’s abili­ty to gene­ra­te suf­fi­ci­ent cash to meet its working capi­tal requi­re­ments or gene­ra­te suf­fi­ci­ent reve­nue and ope­ra­ting cash flow to make all of its plan­ned R&D or stra­te­gic invest­ments; poli­ti­cal, legal, eco­no­mic risks and natu­ral dis­as­ters; future impairm­ents of good­will and tech­no­lo­gy licen­se purcha­ses; AMD’s abili­ty to attract and retain qua­li­fied per­son­nel; AMD’s stock pri­ce vola­ti­li­ty; and world­wi­de poli­ti­cal con­di­ti­ons. Inves­tors are urged to review in detail the risks and uncer­tain­ties in AMD’s Secu­ri­ties and Exch­an­ge Com­mis­si­on filings, inclu­ding but not limi­t­ed to AMD’s most recent reports on Forms 10‑K and 10‑Q.

©2022 Advan­ced Micro Devices, Inc. All rights reser­ved. AMD, the AMD Arrow logo, AMD CDNA, EPYC, Infi­ni­ty Cache, Rade­on, RDNA, Ryzen and com­bi­na­ti­ons the­reof, are trade­marks of Advan­ced Micro Devices, Inc. Other names are for infor­ma­tio­nal pur­po­ses only and may be trade­marks of their respec­ti­ve owners.

¹ Z4-001: IPC uplift based on the avera­ge of estimated/published 2017 SPE­C­int^® and 2017 SPECfp^® scores and inter­nal estimates/testing on Cine­bench R23 1T and Geek­bench 5 1T.for “Zen4” and “Zen 3” processors
² Z4-003: Test­ing as of May 31, 2022, by AMD Per­for­mance Labs. Power mea­su­red at CPU socket only (Watts), CPU per­for­mance (“points”) mea­su­red with Cine­bench R23 nT. AMD Ryzen 9 5950X Sys­tem: AMD Refe­rence X570 Mother­board, 2x8 DDR4-3200. AMD Ryzen 7000 Series: AMD Refe­rence X670 Mother­board, Ryzen 7000 Series 16-core pre-pro­duc­tion pro­ces­sor sam­ple, 2x16GB DDR5-5200. All sys­tems con­fi­gu­red with Rade­on™ RX 6950XT GPU (dri­ver: 22.10 Prime), Win­dows 11 Build 22000.593, Sam­sung 980 Pro 1TB SSD, Ase­tek 280MM liquid coo­ler. Results may vary when final pro­ducts are released in market.
³ Z4-004: Test­ing as of May 5, 2022, by AMD Per­for­mance Labs. Sin­gle-thread per­for­mance eva­lua­ted with Cine­bench R23 1T. AMD Ryzen 9 5950X Sys­tem: ASUS ROG Cross­hair VIII Hero X570, 2x8 DDR4-3600C16. AMD Ryzen 7000 Series: AMD Refe­rence X670 Mother­board, Ryzen 7000 Series 16-core pre-pro­duc­tion pro­ces­sor sam­ple, 2x16GB DDR5-6000CL30. All sys­tems con­fi­gu­red with Rade­on™ RX 6950XT GPU (dri­ver: 22.10 Prime), Win­dows 11 Build 22000.593, Sam­sung 980 Pro 1TB SSD, Ase­tek 280MM liquid coo­ler. Results may vary when final pro­ducts are released in market.
⁴ Based on preli­mi­na­ry inter­nal engi­nee­ring esti­ma­tes. Actu­al results sub­ject to change
⁵ MI300-004: Mea­su­re­ments by AMD Per­for­mance Labs June 4, 2022. MI250X (560W) FP16 (306.4 esti­ma­ted deli­ver­ed TFLOPS based on 80% of peak theo­re­ti­cal floa­ting-point per­for­mance). MI300 FP8 per­for­mance based on preli­mi­na­ry esti­ma­tes and expec­ta­ti­ons. MI300 TDP power based on preli­mi­na­ry pro­jec­tions. Final per­for­mance may vary
⁶ SP5-005: Ser­ver-side Java mul­tiJVM workload demo com­pa­ri­son based on AMD mea­su­red test­ing as of 6/2/2022. Con­fi­gu­ra­ti­ons: 2x 96-core AMD 4th Gen EPYC (pre-pro­duc­tion sili­con) on a refe­rence sys­tem ver­sus 2x 64-core EPYC 7763 on a refe­rence sys­tem. Java ver­si­on JDK18.To fit in demo time per­mit­ted, the full ser­ver-side Java run that typi­cal­ly takes about 20 minu­tes was redu­ced to a short-pre­set time whe­re both sys­tems were run­ning at 99% CPU uti­liza­ti­on, eli­mi­na­ting the “warm up” peri­od data. OEM published scores will vary based on sys­tem con­fi­gu­ra­ti­on and use of pro­duc­tion silicon.
⁷ SP5-006: 128-core 4th Gen EPYC CPUs com­pared to a 64-core 3rd Gen EPYC 7763 for 2x the con­tai­ner density
⁸ “Tech­ni­cal Com­pu­ting” or “Tech­ni­cal Com­pu­ting Workloads” as defi­ned by AMD can include: elec­tro­nic design auto­ma­ti­on, com­pu­ta­tio­nal flu­id dyna­mics, fini­te ele­ment ana­ly­sis, seis­mic tomo­gra­phy, wea­ther fore­cas­ting, quan­tum mecha­nics, cli­ma­te rese­arch, mole­cu­lar mode­ling, or simi­lar workloads. GD-204
⁹ MI300-03 — Mea­su­re­ments by AMD Per­for­mance Labs June 4, 2022 on cur­rent spe­ci­fi­ca­ti­on and/or esti­ma­ti­on for deli­ver­ed FP8 floa­ting point per­for­mance with struc­tu­re spar­si­ty sup­port­ed for AMD Instinct™ MI300 vs. MI250X FP16 (306.4 esti­ma­ted deli­ver­ed TFLOPS based on 80% of peak theo­re­ti­cal floa­ting-point per­for­mance). MI300 per­for­mance based on preli­mi­na­ry esti­ma­tes and expec­ta­ti­ons. Final per­for­mance may vary. MI300-03
¹⁰ Z4-005: Test­ing as of May 5, 2022, by AMD Per­for­mance Labs. Sin­gle-thread per­for­mance eva­lua­ted with Cine­bench R23 1T. AMD Ryzen 9 5950X Sys­tem: ASUS ROG Cross­hair VIII Hero X570, 2x8 DDR4-3600C16. AMD Ryzen 7000 Series: AMD Refe­rence X670 Mother­board, Ryzen 7000 Series 16-core pre-pro­duc­tion pro­ces­sor sam­ple, 2x16GB DDR5-6000CL30. All sys­tems con­fi­gu­red with Rade­on™ RX 6950XT GPU (dri­ver: 22.10 Prime), Win­dows 11 Build 22000.593, Sam­sung 980 Pro 1TB SSD, Ase­tek 280MM liquid cooler.
Test­ing as of May 31, 2022 by AMD per­for­mance labs. Mul­ti-thread per­for­mance eva­lua­ted with Cine­bench R23 1T. AMD Ryzen 9 5950X Sys­tem: AMD Refe­rence X570 Mother­board, 2x8 DDR4-3200. AMD Ryzen 7000 Series: AMD Refe­rence X670 Mother­board, Ryzen 7000 Series 16-core pre-pro­duc­tion pro­ces­sor sam­ple, 2x16GB DDR5-5200. All sys­tems con­fi­gu­red with Rade­on™ RX 6950XT GPU (dri­ver: 22.10 Prime), Win­dows 11 Build 22000.593, Sam­sung 980 Pro 1TB SSD, Ase­tek 280MM liquid cooler
Results may vary when final pro­ducts are released in market.