{"id":122838,"date":"2025-04-02T12:55:26","date_gmt":"2025-04-02T10:55:26","guid":{"rendered":"https:\/\/prohoster.info\/?p=122838"},"modified":"2025-04-02T12:55:26","modified_gmt":"2025-04-02T10:55:26","slug":"novaya-statya-radeon-rx-9070-xt-protiv-geforce-rtx-5070-ti","status":"publish","type":"post","link":"https:\/\/prohoster.info\/et\/blog\/novosti-interneta\/novaya-statya-radeon-rx-9070-xt-protiv-geforce-rtx-5070-ti","title":{"rendered":"Uus artikkel: Radeon RX 9070 XT vs GeForce RTX 5070 Ti","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Blackwell seeria GPU omadused ja testid GeForce RTX 5080<\/a><\/noindex> kas on indicateerinud, et k\u00f5ik 50. seeria positsioonid, v\u00e4lja arvatud GeForce RTX 5090, on vaid veidi rohkem kui pehme uuendus vastavate eelmine p\u00f5lvkonna mudelid. Olukorda on halvendanud uute graafikakaartide puudus ja sellest tulenevalt taevani ulatuvad hinnad.<\/p>\n

Enam enamikul ostjatest tipptasemel seadmetest, mille \u00fcle monopoilselt valitseb NVIDIA, tekitab vaid teoreetilist huvi. Kuid juba j\u00e4rgmisel tasemel on \"roheline\" graafika tehnoloogiline stagnatsioon loonud olukorra, kus AMD v\u00f5iks suure sammu astuda, et tasa teha oma mahaj\u00e4\u00e4mus RDNA arhitektuuri murettekitavates aspektides, nagu ray tracing ja upscale. Kuid ennustus, et AMD taastab konkurendiga pariteedi (kuigi mitte lipulaeva sektoris), on meieni j\u00f5udnud juba pikka aega ning iga kord k\u00f5lab see \u00fcha v\u00e4hem veenvalt, mist\u00f5ttu Radeon RX 9000 graafikakaartide kuulutamine ei p\u00e4lvinud suurt t\u00e4helepanu NVIDIA elava tegevuse taustal. Kuid AMD on teinud tohutu t\u00f6\u00f6 vigade parandamiseks ja kui nad ei ole k\u00f5iki puudusi k\u00f5rvaldanud, siis v\u00e4hemalt kompenseerinud need eelnevate GPU-de peamised puudused.<\/p>\n

Radeon RX 9070 XT \u00fclevaade esitatakse meile ebatavalises 3DNews formaadis, kuna \"punane\" uuendus ja tema peamine konkurent \u2014 GeForce RTX 5070 Ti \u2014 j\u00f5udsid meieni samal ajal, ning \u00fche seadmest ei saa vaadata teise kontekstist eemal, sest Venemaal m\u00fc\u00fcakse m\u00f5lemat ligikaudu sama hinna eest.<\/p>\n\n\n\n
\"Uus <\/td>\n <\/td>\n \"Uus <\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

<\/center> \t\t\t\t <\/p>\n

⇡<\/a><\/noindex>#<\/a><\/noindex> Graafiline protsessor Navi 48 ja arhitektuur RDNA 4<\/strong><\/h2>\n

Et NVIDIA on muutunud k\u00f5rge j\u00f5udlusega GPU turul monopoliks ja k\u00e4itub vastavalt, ei ole \u00f6elnud vaid laisk. Harva osutatakse sellele, et k\u00f5ik ettev\u00f5tted, kes projekteerivad suuri protsessoreid, on sunnitud nende tootmiseks kasutama \u00fche allt\u00f6\u00f6v\u00f5tja teenuseid \u2014 Taiwanis asuvat TSMC-d. TSMC v\u00f5imekuse j\u00e4rele on kindlasti \u00fcks p\u00f5hjusi, miks AMD l\u00f5petas suurte tarbijate GPU-de arendamise, ning \"punaste\" j\u00e4rgmise p\u00f5lvkonna graafikaprotsessorite seeriast on hetkel saadaval vaid \u00fcks greip \u2014 Navi 48.<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Tootja<\/th>\nAMD<\/th>\n<\/tr>\n
Nimi<\/td>\nNavi 32<\/td>\nNavi 31<\/td>\nNavi 48<\/td>\n<\/tr>\n
Kus kasutatakse<\/td>\n\n

Radeon RX 7700 XT;<\/p>\n

Radeon RX 7800 XT<\/p>\n<\/td>\n

\n

Radeon RX 7900 GRE;<\/p>\n

Radeon RX 7900 XT;<\/p>\n

Radeon RX 7900 XTX<\/p>\n<\/td>\n

\n

Radeon RX 9070;<\/p>\n

Radeon RX 9070 XTX<\/p>\n<\/td>\n<\/tr>\n

Arhitektuur<\/td>\nRDNA 3<\/td>\nRDNA 3<\/td>\nRDNA 4<\/td>\n<\/tr>\n
Tootmisprotsess, nm<\/td>\nTSMC N5\/N6<\/td>\nTSMC N5\/N6<\/td>\nTSMC N4P<\/td>\n<\/tr>\n
Transistorite arv, miljard<\/td>\n28,1<\/td>\n57,7<\/td>\n53,9<\/td>\n<\/tr>\n
Kiibi pindala, mm2<\/td>\n346<\/td>\n522<\/td>\n357<\/td>\n<\/tr>\n
CU\/WGP\/SE arv<\/td>\n<\/tr>\n
Arvutus\u00fcksused (CU)<\/td>\n60<\/td>\n96<\/td>\n64<\/td>\n<\/tr>\n
T\u00f6\u00f6grupi protsessorid (WGP)<\/td>\n30<\/td>\n48<\/td>\n32<\/td>\n<\/tr>\n
Shaderi massiivid (SA)<\/td>\nEi<\/td>\nEi<\/td>\nEi<\/td>\n<\/tr>\n
Shaderi mootori seadmed (SE)<\/td>\n4?<\/td>\n6<\/td>\n4<\/td>\n<\/tr>\n
Arvutus\u00fcksuse konfiguratsioon<\/td>\n<\/tr>\n
Vektori ALU (FP32)<\/td>\n2 × 32<\/td>\n2 × 32<\/td>\n2 × 32<\/td>\n<\/tr>\n
Vektori ALU (FP32\/INT32)<\/td>\n2 × 32<\/td>\n2 × 32<\/td>\n2 × 32<\/td>\n<\/tr>\n
Skalaarne ALU<\/td>\n2<\/td>\n2<\/td>\n2<\/td>\n<\/tr>\n
Spetsiaalse otstarbega ALU (SFU)<\/td>\n2 × 8<\/td>\n2 × 8<\/td>\n2 × 8<\/td>\n<\/tr>\n
Kiirte j\u00e4lgimise plokid (Ray Accelerators)<\/td>\n1<\/td>\n1<\/td>\n1<\/td>\n<\/tr>\n
Tekstuuri paigutamise plokid (TMU)<\/td>\n4<\/td>\n4<\/td>\n4<\/td>\n<\/tr>\n
Vektori\/skalaarsed registrid, Kb<\/td>\n384\/10<\/td>\n384\/10<\/td>\n384\/16<\/td>\n<\/tr>\n
L0 vahem\u00e4lu suurus, Kb<\/td>\n32<\/td>\n32<\/td>\n32<\/td>\n<\/tr>\n
T\u00f6\u00f6grupi protsessori (WGP) konfiguratsioon<\/td>\n<\/tr>\n
Shared Memory maht, kB<\/td>\n128<\/td>\n128<\/td>\n128<\/td>\n<\/tr>\n
Juhiste vahem\u00e4lu maht, kB<\/td>\n32<\/td>\n32<\/td>\n32<\/td>\n<\/tr>\n
Skalaark\u00e4idud vahem\u00e4lu maht, kB<\/td>\n16<\/td>\n16<\/td>\n16<\/td>\n<\/tr>\n
GPU arvutusplokid<\/td>\n<\/tr>\n
Vektori ALU (FP32)<\/td>\n3 840 (7 680)<\/td>\n6 144 (12 288)<\/td>\n4 096 (8 192)<\/td>\n<\/tr>\n
Tekstuuri paigutamise plokid (TMU)<\/td>\n240<\/td>\n384<\/td>\n256<\/td>\n<\/tr>\n
Rasteriseerimise operatsioonide plokid (ROP)<\/td>\n96<\/td>\n192<\/td>\n128<\/td>\n<\/tr>\n
Kiirte j\u00e4lgimise plokid (Ray Accelerators)<\/td>\n60<\/td>\n96<\/td>\n64<\/td>\n<\/tr>\n
M\u00e4lu konfiguratsioon<\/td>\n<\/tr>\n
L1 vahem\u00e4lu maht, kB<\/td>\n2 048<\/td>\n3 072<\/td>\n2 048<\/td>\n<\/tr>\n
L2 vahem\u00e4lu maht, MB<\/td>\n4<\/td>\n6<\/td>\n8<\/td>\n<\/tr>\n
Infinity Cache maht, MB<\/td>\n64<\/td>\n96<\/td>\n64<\/td>\n<\/tr>\n
VRAM andmebus, bit<\/td>\n256<\/td>\n384<\/td>\n256<\/td>\n<\/tr>\n
VRAM kiibi t\u00fc\u00fcp<\/td>\nGDDR6<\/td>\nGDDR6<\/td>\nGDDR6<\/td>\n<\/tr>\n
<\/td>\n<\/tr>\n
PCI Express liides<\/td>\n4.0 x16<\/td>\n4.0 x16<\/td>\n5.0 x16<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Vana kiibi seast sarnaneb see k\u00f5ige enam teise taseme mudeliga, Navi 32, peamiste omaduste poolest, kuid sisaldab peaaegu sama palju transistorit kui Navi 31 (53,9 ja 57,7 miljardit vastavalt). Navi 48 loodud TSMC N4P tehnoloogia abil, mis kuulub samasse 5 nm fotolitograafilisse s\u00f5lme, mis on omandatud juba kaks aastat tagasi. Sellegipoolest katab Navi 48 vaid 68% Navi 31 pindalast, kuna AMD loobus chiplet disainist, sest MCD chipletid, mis sisaldavad kolmanda taseme vahem\u00e4lu ja m\u00e4lu kontrollerid, toodetakse karmima N6 tehnoloogia abil.<\/p>\n

\"Uus <\/p>\n

Uue graafikaprotsessori arhitektuur, RDNA 4, ei erine struktuurilt varasematest iteratsioonidest. Navi 48 sisaldab 64 Compute Unit\u2019it ehk 4 096 FP32 kaasasolevat shader ALU-d, mis on t\u00e4iesti t\u00fc\u00fcpiline valem teise klassi GPU-de jaoks \u2014 nagu ka 256-bitine videom\u00e4lu buss. Siiski, AMD vastandab j\u00f5udluse osas Navi 48 lipulaeva Navi 31-le, millel on 6 144 reaalaja shader ALU-d. Erinevuse peaks j\u00f5uliselt kompenseerima arvukad loogika t\u00e4iustused, mille m\u00f5istmiseks soovitame meenutada RDNA 3 arhitektuuri, mille kohta oleme detailset \u00fclevaadet esitanud. Radeon RX 7900 XTX<\/a><\/noindex>.<\/p>\n

V\u00f5rreldes \"roheline\" arhitektuuriga Blackwell<\/a><\/noindex> RDNA 4 on ulatuslikum ja s\u00fcgavam uuendus varasemaist arendustest. Kuid k\u00f5igi uuenduste kirjeldamine kokkuv\u00f5ttes kitseneb l\u00fchikese nimekirja t\u00f5ttu, et NVIDIA p\u00f6\u00f6rab traditsiooniliselt erilist t\u00e4helepanu keerukatele tark- v\u00f5i pooltarkade lahendustele, mis vajavad p\u00f5hjalikke kommentaare, samas kui AMD uuendused saavad tavaliselt alguse ja l\u00f5ppevad riistvaraga.<\/p>\n

\"Uus <\/p>\n

Fundamentaalsed muutused on toimunud Compute Unit\u2019i \u2014 jagamatut arvutust\u00fckki, mis on sarnane NVIDIA SM ja CPU tuumadega. Seega toetab k\u00e4skluste planeerija n\u00fc\u00fcd nn jagatud barj\u00e4\u00e4re, mis v\u00f5imaldavad k\u00e4sklusi t\u00e4itmiseks anda juba nende kehtivuse perioodil. Samuti on optimeeritud k\u00e4skluste eelselektsioon m\u00e4lust ning skalaarsed ALU-d, mida kasutatakse tingimuslike harude ja j\u00f5udude t\u00e4itmiseks, on p\u00e4ritud mobiilsest RDNA 3.5 arhitektuurist. Koos sellega suurendati skalaarkirjete mahtu: 10 KB-st 16 KB-ni.<\/p>\n

\"Uus <\/p>\n

\u00dcks RDNA 4 k\u00f5ige olulisemaid uuendusi on kahekordistunud WMMA matriitsinstruktsioonide t\u00e4itmise tempo, mida kasutatakse masin\u00f5ppe ja andmete t\u00f6\u00f6tlemise \u00fclesannetes n\u00e4rviv\u00f5rkude abil. Lisaks sisaldab RDNA 4 k\u00e4su komplekt k\u00e4ske strukturiseerimata ja harvaesinevate matriitside ning 8-bitiste ujuva punktiga andmete (FP8 ja BF8) arvutuste tegemiseks.<\/p>\n

\"Uus <\/p>\n

Siiski t\u00e4hendab 1 024 FLOP piiramise v\u00f5imekust, et Navi kiibid on j\u00f5udnud vaid kahe aasta taguse \"roheline\" arhitektuuri \u2014 Ampere tasemele. Lisaks, kui NVIDIA ja Intel kiibid kasutavad matriitoperatsioonide jaoks eraldi ALU massiive, j\u00e4\u00e4b RDNA 4 arhitektuuris see t\u00f6\u00f6 shaderite SIMD koormuse alla.<\/p>\n

Kuid \u00e4rgem unustagem, et reaalsetes rakendustes RDNA 3 graafikaprotsessorid ei j\u00f5ua kunagi t\u00e4ieliku koormuseni. Deklareeritud teraFLOPS FP32 \u00fcldotstarbelistes arvutustes on saavutatavad vaid tingimusel, et kasutatakse VOPD (Vector Operation Dual) formaati, mis v\u00f5imaldab kokku pakkida kaks s\u00f5ltumatut vektork\u00e4sku. Kuid see toob endaga kaasa palju piiranguid, seega pooled 32-bitised ALU-d Compute Unit\u2019i sees j\u00e4\u00e4vad enamasti t\u00f6\u00f6tuks. Segakoormuse (nt m\u00e4ngud koos upscale\u2019iga) puhul v\u00f5tavad matriitoperatsioonid loomulikult shaderitelt osa arvutusressursse, kuid need kadumised ei ole nii suured, kui esmapilgul tundub.<\/p>\n\n\n\n\n\n\n
<\/th>\nCompute Unit (AMD RDNA 3)<\/th>\nCompute Unit (AMD RDNA 4)<\/th>\nStreaming Multiprocessor (NVIDIA Ada Lovelace)<\/th>\nStreaming Multiprocessor (NVIDIA Blackwell)<\/th>\nXe-core (Intel Xe-HPG)<\/th>\nXe-core (Intel Xe2)<\/th>\n<\/tr>\n
T\u00e4itevplokid<\/td>\n\n

2 × SIMD32 (FP32\/INT32);<\/p>\n

2 × SIMD32 (FP32);<\/p>\n

2 × SIMD2 (FP64);<\/p>\n

2 × SIMD8 (SFU);<\/p>\n

2 × skalaarsed ALU<\/p>\n<\/td>\n

\n

2 × SIMD32 (FP32\/INT32);<\/p>\n

2 × SIMD32 (FP32);<\/p>\n

2 × SIMD2 (FP64);<\/p>\n

2 × SIMD8 (SFU);<\/p>\n

2 × skalaarsed ALU<\/p>\n<\/td>\n

\n

4 × SIMD16 (FP32\/INT32);<\/p>\n

4 × SIMD16 (FP32);<\/p>\n

2 × SISD? (FP64);<\/p>\n

4 × SIMD4 (SFU);<\/p>\n

4 × skalaarsed ALU;<\/p>\n

4 × tensorivad tuumad<\/p>\n<\/td>\n

\n

8 × SIMD16 (FP32\/INT32);<\/p>\n

2 × SISD? (FP64);<\/p>\n

4 × SIMD4 (SFU);<\/p>\n

4 × skalaarsed ALU;<\/p>\n

4 × tensorivad tuumad<\/p>\n<\/td>\n

\n

16 × SIMD8 (FP32);<\/p>\n

16 × SIMD8 (INT32);<\/p>\n

16 × SISD (FP64);<\/p>\n

16 × SIMD2 (SFU);<\/p>\n

16 × XMX<\/p>\n<\/td>\n

\n

8 × SIMD16 (FP32);<\/p>\n

8 × SIMD16 (INT32);<\/p>\n

8 × SIMD2 (FP64);<\/p>\n

8 × SIMD4 (SFU);<\/p>\n

8 × XMX<\/p>\n<\/td>\n<\/tr>\n

SIMD rida operatsioone taktis<\/td>\n\n

128 × FP32;<\/p>\n

64 × INT32;<\/p>\n

256 × FP16;<\/p>\n

4 × FP64;<\/p>\n

16 × transcedent-funktsioonid<\/p>\n<\/td>\n

\n

128 × FP32;<\/p>\n

64 × INT32;<\/p>\n

256 × FP16;<\/p>\n

4 × FP64;<\/p>\n

16 × transcedent-funktsioonid<\/p>\n<\/td>\n

\n

128 × FP32;<\/p>\n

64 × INT32;<\/p>\n

128 × FP16;<\/p>\n

2 × FP64;<\/p>\n

16 × transcedent-funktsioonid<\/p>\n<\/td>\n

\n

128 × FP32;<\/p>\n

128 × INT32;<\/p>\n

128 × FP16;<\/p>\n

2 × FP64;<\/p>\n

16 × transcedent-funktsioonid<\/p>\n<\/td>\n

\n

128 × FP32;<\/p>\n

128 × INT32;<\/p>\n

256 × FP16;<\/p>\n

16 × FP64;<\/p>\n

32 × transcedent-funktsioonid<\/p>\n<\/td>\n

\n

128 × FP32;<\/p>\n

128 × INT32;<\/p>\n

256 × FP16;<\/p>\n

16 × FP64;<\/p>\n

32 × transcedent-funktsioonid<\/p>\n<\/td>\n<\/tr>\n

Maatriksi operatsioonid, FLOP taktis (FP16)<\/td>\n512<\/td>\n1 024<\/td>\n2 048<\/td>\n2 048<\/td>\n2 048<\/td>\n2 048<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Kaadrite skaleerimine tehisintellekti abil ei ole enam teoreetiline situatsioon Radeon'i kiirendajatele. Uus, neljas versioon FSR toetub l\u00f5puks masin\u00f5ppele, sealhulgas ka kaadrite genereerimisele, ning kasutatavad mudelid on h\u00fcbriidne kombinatsioon konvolutsiooniv\u00f5rkudest ja transformeerijatest. Kahjuks, erinevalt varasematest versioonidest, mis t\u00f6\u00f6tasid praktiliselt igasuguste seadmete juures, toetab FSR 4 ainult RDNA 4 arhitektuuri graafikaprotsessoreid (seoses WMMA k\u00e4su komplekti laiendamisega FP8 andmete jaoks). Universaalset re\u017eiimi, nagu Intel'i XeSS-is, ei ole ette n\u00e4htud.<\/p>\n

\"Uus <\/p>\n

Teine RDNA arhitektuuri reformi suund on kiirteede j\u00e4lgimine. Ray Acceleratori (ja seega ka \u00fche Compute Unit'i) kohta on kiirteede kasti BVH ja kolmnurgaga ristumiste testimise kiirus kahekordistunud: 4 ja 1 takti kohta on n\u00fc\u00fcd 8 ja 2. Siiski, Intel'i Battlemage'i kiipide RT-plokk t\u00f6\u00f6tab 18 ja 2 kiirusel, samas kui Blackwell'i graafikaprotsessorid leiavad \u00fche taktiga neli kiirteede kolmnurgaga ristumist ja avaldamata arvu kasti BVH ristumisi. Lisaks, konkurentide puhul toimub testimine kahel viisil (boks BVH ja kolmnurk) samal ajal, aga AMD-l pole see endiselt v\u00f5imalik. Teatud m\u00e4\u00e4ral on paralleelsus v\u00f5imalik, kuna igas Ray Accelerator'is on n\u00fc\u00fcd kaks iseseisvat \"tester'it\", kuid mitte t\u00e4is kiirusel.<\/p>\n

\"Uus <\/p>\n

Lisaks on Ray Accelerator saanud v\u00f5imaluse t\u00f6\u00f6delda kiiri j\u00e4rjekorrata. Kiired, mis ei vaja aeglastele m\u00e4lutasemetele juurdep\u00e4\u00e4su, t\u00f6\u00f6deldakse esimesena ning testimise tulemused salvestatakse, et \u0161ederdiprogramm saaks need oodatud j\u00e4rjekorras.<\/p>\n

Ray tracing logika kasutab n\u00fc\u00fcd tihedamat BVH8 struktuuri BVH4 asemel (kaheksa asemel neli s\u00f5lme iga mahu hierarhia haru kohta), mis kiirendab protsessi ja v\u00e4hendab samal ajal videom\u00e4lu koormust. BVH suurus m\u00e4lus on veelgi v\u00e4henenud t\u00e4nu uuele primitiivide kokkusurumise algoritmile. L\u00f5puks pakub AMD v\u00f5imalust lahutada BVH kastide suuna absoluutsetest scene koordinatsioonidest. Kastide p\u00f6\u00f6ramine vastavalt objektide geomeetriale v\u00e4hendab otsingusammude arvu.<\/p>\n

\"Uus <\/p>\n

Kahjuks, hoolimata k\u00f5ikidest nende muudatustest, teostab RDNA 4 ikka BVH-d \u0161eidi ALU kaudu, mitte spetsialiseeritud loogika abil. Vastavalt Imagination Technologies'i pakutud tehnikate klassifikatsioonile on AMD lahendused kinni teisel tasemel, Intel'i graafikaprotsessorid kolmandal, samas kui NVIDIA seadmed liiguvad tasemelt 3 tasemele 4, t\u00e4nu SER funktsioonile \u2013 d\u00fcnaamiline k\u00e4suvoogude grupeerimine, et suurendada m\u00e4lule juurdep\u00e4\u00e4su koherentsemust ja seega v\u00e4hendada latentsust.<\/p>\n

\"Uus <\/p>\n

RDNA 4 arhitektuuris ei ole SER analooge, kuid n\u00fc\u00fcd on olemas Out of Order Memory funktsioon, mis aitab teisel viisil sama \u00fclesannet lahendada. Varasemad RDNA versioonid, t\u00f6\u00f6tades mitmete k\u00e4sugruppidega (AMD terminoloogias wavefront'idega), tagastasid andmed m\u00e4lust rangelt selles j\u00e4rjekorras, milles p\u00e4ringud tehti. Kui shader'i poolt n\u00f5utavad andmed on kiires vahem\u00e4lus, v\u00f5ib tagastus siiski viibida, kui teised shader'id on oma p\u00e4ringud ees teinud ja nende andmed asuvad suhteliselt aeglases VRAM-is. T\u00f5de on see, et see piirang tundub mitme\u00f5ppere\u017eiimi kontekstis t\u00e4iesti m\u00f5ttetu, kuid see ilmselt ilmus GPU korraldamise lihtsustamiseks ja j\u00e4\u00e4des mingiks ajaks (loe: rasteriseeritud m\u00e4ngudes) m\u00f5istlikuks kompromissiks. RDNA 4 v\u00f5imaldab k\u00e4sugruppidel andmeid tagastada, olenemata p\u00e4ringute j\u00e4rjekorrast.<\/p>\n

\"Uus <\/p>\n

Out of Order Memory on vaid \u00fcks muudatus RDNA 4 arhitektuuri m\u00e4lu jaoks. Varasemalt toimus registreerimine Compute Unit'ile shader'i programmide jaoks staatiliselt, arvestades maksimaalset tarbimist shader'i eluaja jooksul, mist\u00f5ttu teatud portsjon registrifailist j\u00e4i tegelikult t\u00fchjaks. RDNA 4 v\u00f5imaldab shader'itel registreid d\u00fcnaamiliselt k\u00fcsida ja vabastada. Sellega kasutatakse registrifaili mahtu efektiivsemalt.<\/p>\n

Huvitav, et Shader Engine \u2013 suur GPU struktuur, mis \u00fchendab CU, rasteriseerija, primitiivide ploki ja rasteriseerimise operatsioonide plokid \u2013 on RDNA 4-s kaotanud oma L1 vahem\u00e4lu. K\u00fcll aga sisaldab iga L2 vahem\u00e4lu segment kaks korda rohkem pankasid. Sellega suurenes mitte ainult selle maht, vaid ka l\u00e4bilaskev\u00f5ime, m\u00f5lemas suunas \u2013 Shader Engine ja Infinity Cache suunas (teisis\u00f5nu, L3). Infinity Cache maht j\u00e4i samaks nagu Navi 32-l \u2013 64 MB.<\/p>\n

Navi 48-l on 16 PCI Express 5. p\u00f5lvkonna bussirada, kuid m\u00e4ngulistes \u00fclesannetes ei suuda see loomulikult t\u00e4ielikult h\u00f5ivata nii kiiret liidest.<\/p>\n

\"Uus <\/p>\n

L\u00f5finally, AMD teatas, et j\u00e4rgmise p\u00f5lvkonna GPU-s sisalduv riistvara videokooder tagab parema kodeerimise kvaliteedi madala bitikiirusel, mis on voogedastuse jaoks populaarne, kuid mitte rohkem. See t\u00e4hendab, et YUV 4:2:2 v\u00e4rvisubdistributsioonitoe ei ole endiselt saadaval, samas kui NVIDIA ja Intel graafikaprotsessorid pakuvad seda.<\/p>\n

Mis puutub pildiv\u00e4ljundisse, siis Navi 48 kuva kontroller toetab uusimaid DisplayPort ja HDMI liidese versioone: vastavalt 2.1a ja 2.1b.<\/p>\n

⇡<\/a><\/noindex>#<\/a><\/noindex>Tehnilised spetsifikatsioonid, hinnad<\/strong><\/h2>\n

Radeon RX 9070 XT on varustatud t\u00e4iesti funktsionaalse Navi 48 kiibiga, millel on 64 Compute Unit'i, mis t\u00e4hendab 4 096 reaalset ALU standardse t\u00e4psusega. Raskusi on uue mudeli ametliku eelk\u00e4ija leidmisega \"punaste\" kiirendajate eelmistest p\u00f5lvkondadest, kuna mudelireegel ja GPU nimede printsiip on muutunud (Navi kiibid ei ole kunagi kaheksani j\u00f5udnud). Arvutusplokkide valemi ja m\u00e4lu stack'i omaduste poolest on Radeon RX 9070 XT sarnane Radeon RX 7800 XT-ga, kuid AMD positsioneerib seda Radeon RX 7900 XTX asendajana.<\/p>\n

Game Clock ja Boost Clock (sihtkella sagedus m\u00e4ngudes ja kella sageduse piir) graafikaprotsessoris on v\u00f5rreldes eelmise p\u00f5lvkonna lipulaeva omadustega t\u00f5eliselt kasvanud, ulatudes vastavalt 2,4 ja 2,97 GHz. Radeon RX 7900 XTX teoreetiline j\u00f5udlus, arvestades maksimumkella sagedust, on 26% k\u00f5rgem, kuid AMD kavatseb selle erinevuse kompenseerida RDNA 4 arhitektuuri t\u00e4iustustega.<\/p>\n

Radeon RX 9070 XT sai endale 16 GB GDDR6 m\u00e4lu, mille andmeedastuskiirus on 20 Gbit\/s 256-bitise bussiga. Kuuldavasti kaalunud AMD v\u00f5imalust suurendada VRAMi 32 GB-ni, otsustas siiski, et selline konfiguratsioon on tarbijalahenduste jaoks \u00fclem\u00e4\u00e4rane. Kahju, sest m\u00e4ngude vajadused kasvavad pidevalt, ja just videom\u00e4lu maht on olnud paljude eelk\u00e4ijate seadmise oluline konkurentsieelis.<\/p>\n

Radeon RX 9070 XT viidatud energiatarbimine on 304 W ja soovitatud hind $599 (100 dollari v\u00f5rra k\u00f5rgem kui Radeon RX 7800 XT).<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Tootja<\/th>\nAMD<\/th>\nNVIDIA<\/th>\n<\/tr>\n
Mudel<\/td>\nRadeon RX 7900 XT<\/td>\nRadeon RX 7900 XTX<\/td>\nRadeon RX 9070<\/td>\nRadeon RX 9070 XT<\/td>\nGeForce RTX 4070 Ti SUPER<\/td>\nGeForce RTX 4080 SUPER<\/td>\nGeForce RTX 5070 Ti<\/td>\nGeForce RTX 5080<\/td>\n<\/tr>\n
Graafikaprotsessor<\/td>\n<\/tr>\n
Nimi<\/td>\nNavi 31 XT<\/td>\nNavi 31 XTX<\/td>\nNavi 48 XT<\/td>\nNavi 48 XTX<\/td>\nAD103<\/td>\nAD103<\/td>\nGB203<\/td>\nGB203<\/td>\n<\/tr>\n
Arhitektuur<\/td>\nRDNA 3<\/td>\nRDNA 3<\/td>\nRDNA 4<\/td>\nRDNA 4<\/td>\nAda Lovelace<\/td>\nAda Lovelace<\/td>\nBlackwell<\/td>\nBlackwell<\/td>\n<\/tr>\n
Tehnoloogia protsess<\/td>\nTSMC N5\/N6<\/td>\nTSMC N5\/N6<\/td>\nTSMC N4C<\/td>\nTSMC N4C<\/td>\nTSMC 4N<\/td>\nTSMC 4N<\/td>\nTSMC 4NP<\/td>\nTSMC 4NP<\/td>\n<\/tr>\n
Transistorite arv, miljard<\/td>\n57,7<\/td>\n57,7<\/td>\n53,9<\/td>\n53,9<\/td>\n45,9<\/td>\n45,9<\/td>\n45,6<\/td>\n45,6<\/td>\n<\/tr>\n
Taktsfrekvens (Base Clock\/Game Clock\/Boost Clock), MHz<\/td>\n1 500 \/2 025\/2 394<\/td>\n1 855 \/2 269\/2 499<\/td>\n1 330 \/2 070\/2 520<\/td>\n1 660 \/2 400\/2 970<\/td>\n2 310\/2 610<\/td>\n2 205\/2 550<\/td>\n2 295\/2 452<\/td>\n2 295\/2 617<\/td>\n<\/tr>\n
Shader ALU (FP32)<\/td>\n5 376 (10 752)<\/td>\n6 144 (12 228)<\/td>\n3 584 (7 168)<\/td>\n4 096 (8 192)<\/td>\n8 448<\/td>\n10 240<\/td>\n8 960<\/td>\n10 752<\/td>\n<\/tr>\n
Tekstuuri paigutamise plokid (TMU)<\/td>\n336<\/td>\n384<\/td>\n224<\/td>\n256<\/td>\n264<\/td>\n320<\/td>\n280<\/td>\n336<\/td>\n<\/tr>\n
Rasteriseerimise operatsioonide plokid (ROP)<\/td>\n192<\/td>\n192<\/td>\n128<\/td>\n128<\/td>\n112<\/td>\n112<\/td>\n96<\/td>\n112<\/td>\n<\/tr>\n
Tensor tuumad<\/td>\nEi<\/td>\nEi<\/td>\nEi<\/td>\nEi<\/td>\n264<\/td>\n320<\/td>\n280<\/td>\n336<\/td>\n<\/tr>\n
RT-tuums<\/td>\n84<\/td>\n96<\/td>\n112<\/td>\n128<\/td>\n66<\/td>\n80<\/td>\n70<\/td>\n84<\/td>\n<\/tr>\n
Viimane tasandi vahem\u00e4lu maht, MB<\/td>\n80<\/td>\n96<\/td>\n64<\/td>\n64<\/td>\n48<\/td>\n64<\/td>\n64<\/td>\n64<\/td>\n<\/tr>\n
Operatiivm\u00e4lu<\/td>\n<\/tr>\n
Bussilaius, bit<\/td>\n320<\/td>\n384<\/td>\n256<\/td>\n256<\/td>\n256<\/td>\n256<\/td>\n256<\/td>\n256<\/td>\n<\/tr>\n
K\u00fcpsise t\u00fc\u00fcp<\/td>\nGDDR6 SGRAM<\/td>\nGDDR6 SGRAM<\/td>\nGDDR6 SGRAM<\/td>\nGDDR6 SGRAM<\/td>\nGDDR6X SGRAM<\/td>\nGDDR6X SGRAM<\/td>\nGDDR7 SGRAM<\/td>\nGDDR7 SGRAM<\/td>\n<\/tr>\n
\u00dclekandev\u00f5ime kontaktiga, Gbit\/s<\/td>\n20<\/td>\n20<\/td>\n20<\/td>\n20<\/td>\n21<\/td>\n23<\/td>\n28<\/td>\n30<\/td>\n<\/tr>\n
Kokku \u00fclekandev\u00f5ime, GB\/s<\/td>\n800<\/td>\n960<\/td>\n640<\/td>\n640<\/td>\n672<\/td>\n736<\/td>\n896<\/td>\n960<\/td>\n<\/tr>\n
Maht, GB<\/td>\n20<\/td>\n24<\/td>\n16<\/td>\n16<\/td>\n16<\/td>\n16<\/td>\n16<\/td>\n16<\/td>\n<\/tr>\n
Tootlikkus<\/td>\n<\/tr>\n
Tipptasemel FP32 j\u00f5udlus, TFLOPS<\/td>\n25,7 (51,5)<\/td>\n30,7 (61,1)<\/td>\n18,1 (36,1)<\/td>\n24,3 (48,7)<\/td>\n44,1<\/td>\n52,2<\/td>\n43,9<\/td>\n56,3<\/td>\n<\/tr>\n
FP64\/FP32 j\u00f5udlus<\/td>\n1\/32 (1\/64)<\/td>\n1\/32 (1\/64)<\/td>\n1\/32 (1\/64)<\/td>\n1\/32 (1\/64)<\/td>\n1\/64<\/td>\n1\/64<\/td>\n1\/64<\/td>\n1\/64<\/td>\n<\/tr>\n
FP16\/FP32 j\u00f5udlus<\/td>\n2\/1<\/td>\n2\/1<\/td>\n2\/1<\/td>\n2\/1<\/td>\n1\/1<\/td>\n1\/1<\/td>\n1\/1<\/td>\n1\/1<\/td>\n<\/tr>\n
Muud<\/td>\n<\/tr>\n
PCI Express buss<\/td>\nPCI Express 4.0 x16<\/td>\nPCI Express 4.0 x16<\/td>\nPCI Express 5.0 x16<\/td>\nPCI Express 5.0 x16<\/td>\nPCI Express 4.0 x16<\/td>\nPCI Express 4.0 x16<\/td>\n5.0 x16<\/td>\n5.0 x16<\/td>\n<\/tr>\n
Kujutise v\u00e4ljundi liidesed<\/td>\nDisplayPort 2.1, HDMI 2.1a<\/td>\nDisplayPort 2.1, HDMI 2.1a<\/td>\nDisplayPort 2.1a, HDMI 2.1b<\/td>\nDisplayPort 2.1a, HDMI 2.1b<\/td>\nDisplayPort 1.4a, HDMI 2.1<\/td>\nDisplayPort 1.4a, HDMI 2.1<\/td>\nDisplayPort 2.1b, HDMI 2.1b<\/td>\nDisplayPort 2.1b, HDMI 2.1b<\/td>\n<\/tr>\n
TDP\/TBP, W<\/td>\n315<\/td>\n335<\/td>\n220<\/td>\n304<\/td>\n285<\/td>\n320<\/td>\n300<\/td>\n360<\/td>\n<\/tr>\n
Jaehind (USA), $<\/td>\n899 (soovituslik v\u00e4ljalaskekuup\u00e4eval)<\/td>\n999 (soovituslik v\u00e4ljalaskekuup\u00e4eval)<\/td>\n549 (soovituslik v\u00e4ljalaskekuup\u00e4eval)<\/td>\n599 (soovituslik v\u00e4ljalaskekuup\u00e4eval)<\/td>\n799 (soovituslik v\u00e4ljalaskekuup\u00e4eval)<\/td>\n999 (soovituslik v\u00e4ljalaskekuup\u00e4eval)<\/td>\n749 (soovituslik v\u00e4ljalaskekuup\u00e4eval)<\/td>\n999 (soovituslik v\u00e4ljalaskekuup\u00e4eval)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

\u0412 \u0441\u0432\u043e\u044e \u043e\u0447\u0435\u0440\u0435\u0434\u044c, GeForce RTX 5070 Ti \u043e\u0441\u043d\u043e\u0432\u0430\u043d \u043d\u0430 \u0442\u043e\u043c \u0436\u0435 \u0433\u0440\u0430\u0444\u0438\u0447\u0435\u0441\u043a\u043e\u043c \u043f\u0440\u043e\u0446\u0435\u0441\u0441\u043e\u0440\u0435 GB203, \u043a\u043e\u0442\u043e\u0440\u044b\u0439 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u0442\u0441\u044f \u0432 GeForce RTX 5080, \u043d\u043e \u043c\u043b\u0430\u0434\u0448\u0430\u044f \u043c\u043e\u0434\u0435\u043b\u044c \u043b\u0438\u0448\u0438\u043b\u0430\u0441\u044c 14 \u043f\u043e\u0442\u043e\u043a\u043e\u0432\u044b\u0445 \u043c\u0443\u043b\u044c\u0442\u0438\u043f\u0440\u043e\u0446\u0435\u0441\u0441\u043e\u0440\u043e\u0432, \u0442\u043e \u0435\u0441\u0442\u044c 1 792 FP32-\u0441\u043e\u0432\u043c\u0435\u0441\u0442\u0438\u043c\u044b\u0445 \u0448\u0435\u0439\u0434\u0435\u0440\u043d\u044b\u0445 ALU, \u0430 \u0438\u0433\u0440\u043e\u0432\u0430\u044f \u0442\u0430\u043a\u0442\u043e\u0432\u0430\u044f \u0447\u0430\u0441\u0442\u043e\u0442\u0430 \u0431\u044b\u043b\u0430 \u0441\u043d\u0438\u0436\u0435\u043d\u0430. \u0412 \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u0435 GeForce RTX 5080 \u043e\u0442\u0434\u0435\u043b\u044f\u044e\u0442 \u043e\u0442 RTX 5070 Ti 27 % \u0442\u0435\u043e\u0440\u0435\u0442\u0438\u0447\u0435\u0441\u043a\u043e\u0433\u043e \u0431\u044b\u0441\u0442\u0440\u043e\u0434\u0435\u0439\u0441\u0442\u0432\u0438\u044f, \u0430 \u0441\u0430\u043c GeForce RTX 5070 Ti \u043f\u043e \u044d\u0442\u043e\u043c\u0443 \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u0443 \u043d\u0435 \u043e\u0442\u043b\u0438\u0447\u0430\u0435\u0442\u0441\u044f \u043e\u0442 RTX 4070 Ti SUPER. \u041a\u0430\u043a \u0438 \u0432 \u0441\u043b\u0443\u0447\u0430\u0435 Radeon RX 9070 XT, \u043d\u0435\u0445\u0432\u0430\u0442\u043a\u0443 \u0441\u044b\u0440\u043e\u0439 \u0432\u044b\u0447\u0438\u0441\u043b\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0439 \u043c\u043e\u0449\u043d\u043e\u0441\u0442\u0438 \u043f\u0440\u0438\u0437\u0432\u0430\u043d\u0430 \u043a\u043e\u043c\u043f\u0435\u043d\u0441\u0438\u0440\u043e\u0432\u0430\u0442\u044c \u043d\u043e\u0432\u0430\u044f \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440\u0430 GPU, \u0438 \u0432 \u043f\u0435\u0440\u0432\u0443\u044e \u043e\u0447\u0435\u0440\u0435\u0434\u044c — \u0444\u0443\u043d\u043a\u0446\u0438\u044f \u0433\u0435\u043d\u0435\u0440\u0430\u0446\u0438\u0438 \u043c\u043d\u043e\u0436\u0435\u0441\u0442\u0432\u0435\u043d\u043d\u044b\u0445 \u043a\u0430\u0434\u0440\u043e\u0432 \u0441 \u043f\u043e\u043c\u043e\u0449\u044c\u044e DLSS (\u0432\u043f\u043b\u043e\u0442\u044c \u0434\u043e \u0442\u0440\u0435\u0445 \u0441\u0433\u0435\u043d\u0435\u0440\u0438\u0440\u043e\u0432\u0430\u043d\u043d\u044b\u0445 \u043a\u0430\u0434\u0440\u043e\u0432 \u0432 \u043f\u0440\u043e\u043c\u0435\u0436\u0443\u0442\u043a\u0435 \u043c\u0435\u0436\u0434\u0443 \u0434\u0432\u0443\u043c\u044f «\u043d\u0430\u0441\u0442\u043e\u044f\u0449\u0438\u043c\u0438»). <\/p>\n

GeForce RTX 5070 Ti on kavandatud 300 W tarbimisega ja MSRP $749 juures on see $50 odavam kui GeForce RTX 4070 Ti SUPER.<\/p>\n

\u0420\u0430\u0437\u043d\u0438\u0446\u0430 \u0432 $150 \u0440\u0435\u043a\u043e\u043c\u0435\u043d\u0434\u043e\u0432\u0430\u043d\u043d\u043e\u0439 \u0432\u0430\u043b\u044e\u0442\u043d\u043e\u0439 \u0441\u0442\u043e\u0438\u043c\u043e\u0441\u0442\u0438 \u043c\u0435\u0436\u0434\u0443 Radeon RX 9070 XT \u0438 GeForce RTX 5070 Ti \u043d\u0435 \u0437\u0430\u043c\u0435\u0442\u043d\u0430 \u043f\u043e \u0442\u0435\u043a\u0443\u0449\u0438\u043c \u0440\u043e\u0441\u0441\u0438\u0439\u0441\u043a\u0438\u043c \u0446\u0435\u043d\u0430\u043c: \u0441\u0430\u043c\u044b\u0435 \u0434\u043e\u0441\u0442\u0443\u043f\u043d\u044b\u0435 \u043f\u0430\u0440\u0442\u043d\u0435\u0440\u0441\u043a\u0438\u0435 \u043c\u043e\u0434\u0435\u043b\u0438 \u043c\u044b \u043d\u0430\u0448\u043b\u0438 \u0432 DNS \u0437\u0430 99 999 \u0438 104 999 \u0440\u0443\u0431. \u0441\u043e\u043e\u0442\u0432\u0435\u0442\u0441\u0442\u0432\u0435\u043d\u043d\u043e. \u0418\u043c\u0435\u043d\u043d\u043e \u0438\u0445 \u043d\u0430\u043c \u043f\u0440\u0435\u0434\u0441\u0442\u043e\u0438\u0442 \u0438\u0441\u043f\u044b\u0442\u0430\u0442\u044c \u0432 \u0431\u0435\u043d\u0447\u043c\u0430\u0440\u043a\u0430\u0445. Radeon RX 9070 XT \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u044f\u0435\u0442 \u0432\u0438\u0434\u0435\u043e\u043a\u0430\u0440\u0442\u0430 GIGABYTE GAMING OC \u0441 \u0437\u0430\u0432\u043e\u0434\u0441\u043a\u0438\u043c \u0440\u0430\u0437\u0433\u043e\u043d\u043e\u043c GPU \u0434\u043e 2,52\/3,06 \u0413\u0413\u0446, \u0430 GeForce RTX 5070 Ti — \u0443\u0441\u043a\u043e\u0440\u0438\u0442\u0435\u043b\u044c Palit GamingPro \u0441 \u0440\u0435\u0444\u0435\u0440\u0435\u043d\u0441\u043d\u044b\u043c\u0438 \u0441\u043f\u0435\u0446\u0438\u0444\u0438\u043a\u0430\u0446\u0438\u044f\u043c\u0438.<\/p>\n

⇡<\/a><\/noindex>#<\/a><\/noindex>GIGABYTE Radeon RX 9070 XT GAMING OC: konstruktsioon<\/strong><\/h2>\n

Radeon RX 9070 XT \u0432 \u0438\u0441\u043f\u043e\u043b\u043d\u0435\u043d\u0438\u0438 GIGABYTE \u0438\u043c\u0435\u0435\u0442 \u0434\u043e\u0432\u043e\u043b\u044c\u043d\u043e \u0441\u043a\u0440\u043e\u043c\u043d\u044b\u0435 \u0433\u0430\u0431\u0430\u0440\u0438\u0442\u044b \u0434\u043b\u044f \u0432\u0438\u0434\u0435\u043e\u043a\u0430\u0440\u0442\u044b \u0441 \u044d\u043d\u0435\u0440\u0433\u043e\u043f\u043e\u0442\u0440\u0435\u0431\u043b\u0435\u043d\u0438\u0435\u043c \u043d\u0435 \u043c\u0435\u043d\u044c\u0448\u0435 (\u0438, \u043a\u0430\u043a \u043f\u043e\u043a\u0430\u0436\u0443\u0442 \u0442\u0435\u0441\u0442\u044b, \u043d\u0430\u043c\u043d\u043e\u0433\u043e \u0431\u043e\u043b\u044c\u0448\u0435) 300 \u0412\u0442: 288 \u043c\u043c \u0432 \u0434\u043b\u0438\u043d\u0443 \u0438 \u0442\u0440\u0438 \u0441\u043b\u043e\u0442\u0430 \u0432 \u0448\u0438\u0440\u0438\u043d\u0443. \u0415\u0434\u0438\u043d\u0441\u0442\u0432\u0435\u043d\u043d\u044b\u043c \u044d\u043b\u0435\u043c\u0435\u043d\u0442\u043e\u043c \u0441\u0432\u0435\u0442\u043e\u0434\u0438\u043e\u0434\u043d\u043e\u0439 \u043f\u043e\u0434\u0441\u0432\u0435\u0442\u043a\u0438 \u044f\u0432\u043b\u044f\u0435\u0442\u0441\u044f \u043a\u043e\u0440\u043e\u0442\u043a\u0430\u044f \u043f\u043e\u043b\u043e\u0441\u0430 \u043d\u0430 \u0431\u043e\u043a\u0443. \u0428\u0438\u043b\u044c\u0434\u0438\u043a \u0441 \u043b\u043e\u0433\u043e\u0442\u0438\u043f\u043e\u043c \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044f — \u043f\u043e\u0434\u0432\u0438\u0436\u043d\u044b\u0439: \u0435\u0441\u043b\u0438 \u0441\u043c\u0435\u0441\u0442\u0438\u0442\u044c \u0435\u0433\u043e \u043d\u0430 LED-\u043f\u043e\u043b\u043e\u0441\u0443, \u043b\u043e\u0433\u043e\u0442\u0438\u043f \u043d\u0430\u0447\u0438\u043d\u0430\u0435\u0442\u0441\u044f \u0441\u0432\u0435\u0442\u0438\u0442\u044c\u0441\u044f.<\/p>\n

\"Uus <\/p>\n

Jahutuss\u00fcsteemi hooldab kolm ventilaatorit, mille labade diameeter on 87 mm. \u00d5hk voolab l\u00e4bi radiaatori v\u00e4ikese sektsiooni, t\u00e4nu tagapaneelil olevale l\u00f5ike avale.<\/p>\n

\"Uus <\/p>\n

\u0413\u0440\u0430\u0444\u0438\u0447\u0435\u0441\u043a\u0438\u0439 \u043f\u0440\u043e\u0446\u0435\u0441\u0441\u043e\u0440 \u0438 \u0447\u0438\u043f\u044b \u0432\u0438\u0434\u0435\u043e\u043f\u0430\u043c\u044f\u0442\u0438 \u043e\u0442\u0434\u0430\u044e\u0442 \u0442\u0435\u043f\u043b\u043e \u0438\u0441\u043f\u0430\u0440\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0439 \u043a\u0430\u043c\u0435\u0440\u0435, \u043d\u043e \u043f\u0435\u0440\u0432\u044b\u0439 — \u043d\u0430\u043f\u0440\u044f\u043c\u0443\u044e, \u0430 \u043f\u043e\u0441\u043b\u0435\u0434\u043d\u0438\u0435 — \u0447\u0435\u0440\u0435\u0437 \u041f-\u043e\u0431\u0440\u0430\u0437\u043d\u0443\u044e \u043c\u0435\u0442\u0430\u043b\u043b\u0438\u0447\u0435\u0441\u043a\u0443\u044e \u043f\u0440\u043e\u0441\u0442\u0430\u0432\u043a\u0443. \u041a \u043b\u0430\u043c\u0435\u043b\u044f\u043c \u0440\u0430\u0434\u0438\u0430\u0442\u043e\u0440\u0430 \u043f\u0440\u0438\u043a\u0440\u0435\u043f\u043b\u0435\u043d\u044b \u043e\u0442\u0434\u0435\u043b\u044c\u043d\u044b\u0435 \u043f\u043b\u0430\u0441\u0442\u0438\u043d\u044b, \u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u0441\u043b\u0443\u0436\u0430\u0442 \u0442\u0435\u043f\u043b\u043e\u0441\u044a\u0435\u043c\u043d\u0438\u043a\u0430\u043c\u0438 \u043a\u043e\u043c\u043f\u043e\u043d\u0435\u043d\u0442\u043e\u0432 VRM. \u0422\u0435\u043f\u043b\u043e\u0442\u0440\u0443\u0431\u043e\u043a \u0432\u0441\u0435\u0433\u043e \u0441\u0435\u043c\u044c, \u0438\u0445 \u0434\u0438\u0430\u043c\u0435\u0442\u0440 — 6 \u043c\u043c. \u0417\u0430\u043c\u0435\u0442\u0438\u043c, \u0447\u0442\u043e \u043e\u0431\u044b\u0447\u043d\u044b\u043c \u0442\u0435\u0440\u043c\u043e\u043f\u0440\u043e\u043a\u043b\u0430\u0434\u043a\u0430\u043c GIGABYTE \u043f\u0440\u0435\u0434\u043f\u043e\u0447\u043b\u0430 \u0436\u0438\u0434\u043a\u0438\u0435. \u0412\u043e\u0437\u043c\u043e\u0436\u043d\u043e, \u0438\u043c\u0435\u043d\u043d\u043e \u0432 \u044d\u0442\u043e\u043c \u043f\u0440\u0438\u0447\u0438\u043d\u0430 \u0442\u043e\u0433\u043e, \u043a\u0430\u043a \u043d\u0430\u0433\u0440\u0435\u0432\u0430\u044e\u0442\u0441\u044f \u043f\u043e\u0434 \u043d\u0430\u0433\u0440\u0443\u0437\u043a\u043e\u0439 \u043c\u0438\u043a\u0440\u043e\u0441\u0445\u0435\u043c\u044b VRAM (\u0441\u043c. \u044d\u043c\u043f\u0438\u0440\u0438\u0447\u0435\u0441\u043a\u0438\u0439 \u0440\u0430\u0437\u0434\u0435\u043b \u043e\u0431\u0437\u043e\u0440\u0430).<\/p>\n

\"Uus <\/p>\n

Tagaplaadil on suur roll tr\u00fckkplaadi soojuse eemaldamisel mitmete termopaigaldustega (siin juba tavalised, mitte vedelad).<\/p>\n

\"Uus <\/p>\n

⇡<\/a><\/noindex>#<\/a><\/noindex>GIGABYTE Radeon RX 9070 XT GAMING OC: tr\u00fckkplaat<\/strong><\/h2>\n

\u041d\u0435\u043a\u043e\u0442\u043e\u0440\u044b\u0435 \u0440\u0430\u0437\u043d\u043e\u0432\u0438\u0434\u043d\u043e\u0441\u0442\u0438 Radeon RX 9070 XT \u043f\u043e\u043b\u0443\u0447\u0430\u044e\u0442 \u043f\u0438\u0442\u0430\u043d\u0438\u0435 \u0447\u0435\u0440\u0435\u0437 \u0440\u0430\u0437\u044a\u0435\u043c 12V-2×6, \u043d\u043e GIGABYTE \u043f\u0440\u0435\u0434\u043f\u043e\u0447\u043b\u0430 \u0442\u0440\u0438 \u0443\u043d\u0438\u0432\u0435\u0440\u0441\u0430\u043b\u044c\u043d\u044b\u0445 \u0432\u043e\u0441\u044c\u043c\u0438\u043a\u043e\u043d\u0442\u0430\u043a\u0442\u043d\u044b\u0445 \u0440\u0430\u0437\u044a\u0435\u043c\u0430. \u0412 \u0441\u0432\u0435\u0442\u0435 \u0432\u044b\u0441\u043e\u043a\u043e\u0433\u043e \u044d\u043d\u0435\u0440\u0433\u043e\u043f\u043e\u0442\u0440\u0435\u0431\u043b\u0435\u043d\u0438\u044f \u0432\u0438\u0434\u0435\u043e\u043a\u0430\u0440\u0442\u044b \u0438 \u043d\u0435\u043f\u0440\u0435\u043a\u0440\u0430\u0449\u0430\u044e\u0449\u0438\u0445\u0441\u044f \u0436\u0430\u043b\u043e\u0431 \u043d\u0430 \u043d\u043e\u0432\u044b\u0439 \u043a\u043e\u043d\u043d\u0435\u043a\u0442\u043e\u0440 \u044d\u0442\u043e \u0431\u043e\u043b\u0435\u0435 \u043d\u0430\u0434\u0435\u0436\u043d\u043e\u0435, \u0445\u043e\u0442\u044c \u0438 \u043d\u0435 \u0441\u0442\u043e\u043b\u044c \u0443\u0434\u043e\u0431\u043d\u043e\u0435 \u0438 \u044d\u0441\u0442\u0435\u0442\u0438\u0447\u043d\u043e\u0435 \u0440\u0435\u0448\u0435\u043d\u0438\u0435. \u0418\u0437 17 \u0444\u0430\u0437 \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043f\u0438\u0442\u0430\u043d\u0438\u044f 14, \u043f\u043e\u0434 \u0443\u043f\u0440\u0430\u0432\u043b\u0435\u043d\u0438\u0435\u043c \u0428\u0418\u041c-\u043a\u043e\u043d\u0442\u0440\u043e\u043b\u043b\u0435\u0440\u0430 Monolithic Power Systems MP2868A, \u0432\u044b\u0434\u0435\u043b\u0435\u043d\u044b \u0433\u0440\u0430\u0444\u0438\u0447\u0435\u0441\u043a\u043e\u043c\u0443 \u043f\u0440\u043e\u0446\u0435\u0441\u0441\u043e\u0440\u0443, \u0435\u0449\u0435 \u0442\u0440\u0438 — \u0432\u0438\u0434\u0435\u043e\u043f\u0430\u043c\u044f\u0442\u0438. \u041e\u0431\u0430 VRM \u0443\u043a\u043e\u043c\u043f\u043b\u0435\u043a\u0442\u043e\u0432\u0430\u043d\u044b \u0441\u0438\u043b\u043e\u0432\u044b\u043c\u0438 \u043a\u0430\u0441\u043a\u0430\u0434\u0430\u043c\u0438 MP87993 \u0441 \u043f\u0440\u0435\u0434\u043f\u043e\u043b\u043e\u0436\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u043c \u043d\u043e\u043c\u0438\u043d\u0430\u043b\u043e\u043c \u0442\u043e\u043a\u0430 90A.<\/p>\n

VRAM on kokku pandud SK hynixi kiipidest, mille t\u00e4histusel H56G42AS8DX-014 on kooditud ribalaius 20 Gbit\/s.<\/p>\n

\u041f\u043e\u043c\u0438\u043c\u043e \u043a\u043e\u043f\u0438\u0438 BIOS, \u0430\u043a\u0442\u0438\u0432\u043d\u043e\u0439 \u043f\u043e \u0443\u043c\u043e\u043b\u0447\u0430\u043d\u0438\u044e, \u0432\u0438\u0434\u0435\u043e\u043a\u0430\u0440\u0442\u0430 \u0438\u043c\u0435\u0435\u0442 \u0440\u0435\u0437\u0435\u0440\u0432\u043d\u0443\u044e «\u0442\u0438\u0445\u0443\u044e» \u043f\u0440\u043e\u0448\u0438\u0432\u043a\u0443.<\/p>\n

\"Uus \"Uus <\/p>\n

⇡<\/a><\/noindex>#<\/a><\/noindex>Palit GeForce RTX 5070 Ti GamingPro: konstruktsioon<\/strong><\/h2>\n

Palit GeForce RTX 5070 Ti GamingPro \u0437\u0430\u043d\u0438\u043c\u0430\u0435\u0442 331,9 \u043c\u043c \u0432 \u0434\u043b\u0438\u043d\u0443, \u043d\u043e \u0432\u0441\u0435\u0433\u043e \u0442\u0440\u0438 \u0441\u043b\u043e\u0442\u0430 \u0440\u0430\u0441\u0448\u0438\u0440\u0435\u043d\u0438\u044f. \u041f\u043e\u0434 \u043b\u043e\u0433\u043e\u0442\u0438\u043f\u043e\u043c GamingPro \u0438 \u0441\u043e\u0441\u0435\u0434\u043d\u0438\u043c \u0441\u0435\u0433\u043c\u0435\u043d\u0442\u043e\u043c \u0444\u0440\u043e\u043d\u0442\u0430\u043b\u044c\u043d\u043e\u0439 \u043f\u0430\u043d\u0435\u043b\u0438 \u043d\u0430\u0445\u043e\u0434\u044f\u0442\u0441\u044f \u0441\u0432\u0435\u0442\u043e\u0434\u0438\u043e\u0434\u044b RGB-\u043f\u043e\u0434\u0441\u0432\u0435\u0442\u043a\u0438. \u0420\u0430\u0431\u043e\u0442\u0443 LED \u043c\u043e\u0436\u043d\u043e \u0441\u0438\u043d\u0445\u0440\u043e\u043d\u0438\u0437\u0438\u0440\u043e\u0432\u0430\u0442\u044c \u0441 \u043c\u0430\u0442\u0435\u0440\u0438\u043d\u0441\u043a\u043e\u0439 \u043f\u043b\u0430\u0442\u043e\u0439 \u0447\u0435\u0440\u0435\u0437 \u0441\u0442\u0430\u043d\u0434\u0430\u0440\u0442\u043d\u044b\u0439 ARGB-\u043a\u043e\u043d\u043d\u0435\u043a\u0442\u043e\u0440, \u043a\u043e\u0442\u043e\u0440\u044b\u0439 \u0440\u0430\u0441\u043f\u043e\u043b\u043e\u0436\u0435\u043d \u0440\u044f\u0434\u043e\u043c \u0441\u043e \u0432\u0445\u043e\u0434\u043e\u043c \u043f\u0438\u0442\u0430\u043d\u0438\u044f 12V-2×6.<\/p>\n

\"Uus <\/p>\n

Korpuse k\u00fclgpinnad moodustab valatud alumiiniumraam, millel on ventilatsiooniavad pikitelgedel. PCB tagak\u00fcljel asuvas kaitseplaadis on tuttav v\u00f5rgu struktuur jahuti l\u00e4bi puhumiseks.<\/p>\n

\"Uus <\/p>\n

Jahutaja alusel asub suure pindalaga aurukamber, mis katab GPU kristalli ja selle \u00fcmber asuvate videom\u00e4lu kiipide. VRM komponentide jaoks on ette n\u00e4htud eraldi plaatjad soojuse eemaldajad. Radiaatori lamellid on t\u00f5mmatud kokku kaheksa 6 mm l\u00e4bim\u00f5\u00f5duga soojusr\u00f6riga.<\/p>\n

\"Uus <\/p>\n

Metalltagune plaat t\u00e4idab ainult kaitse- ja dekoratiivset funktsiooni, kuna selle ja PCB vahel ei ole \u00fchtegi soojusmatti.<\/p>\n

\"Uus <\/p>\n

\u041a \u0432\u0438\u0434\u0435\u043e\u043a\u0430\u0440\u0442\u0435 \u043f\u0440\u0438\u043b\u0430\u0433\u0430\u0435\u0442\u0441\u044f \u043f\u0435\u0440\u0435\u0445\u043e\u0434\u043d\u0438\u043a \u0441 \u0434\u0432\u0443\u0445 \u0432\u043e\u0441\u044c\u043c\u0438\u043a\u043e\u043d\u0442\u0430\u043a\u0442\u043d\u044b\u0445 \u0440\u0430\u0437\u044a\u0435\u043c\u043e\u0432 \u043f\u0438\u0442\u0430\u043d\u0438\u044f \u043d\u0430 12V-2×6 \u0438 \u043d\u0435\u0431\u043e\u043b\u044c\u0448\u043e\u0439 \u0442\u043a\u0430\u043d\u0435\u0432\u044b\u0439 \u043a\u043e\u0432\u0440\u0438\u043a \u0434\u043b\u044f \u043c\u044b\u0448\u0438.<\/p>\n

\"Uus <\/p>\n

⇡<\/a><\/noindex>#<\/a><\/noindex>Palit GeForce RTX 5070 Ti GamingPro: tr\u00fckkplaat<\/strong><\/h2>\n

Erinevalt videokaardist Palit GameRock<\/a><\/noindex>, \u043a\u043e\u0442\u043e\u0440\u0443\u044e \u043c\u044b \u0440\u0430\u0441\u0441\u043c\u043e\u0442\u0440\u0435\u043b\u0438 \u0432 \u043d\u0435\u0434\u0430\u0432\u043d\u0435\u043c \u043e\u0431\u0437\u043e\u0440\u0435 GeForce RTX 5080, \u0441\u043b\u0435\u0434\u0443\u044e\u0449\u0430\u044f \u043f\u043e \u0441\u0442\u0430\u0440\u0448\u0438\u043d\u0441\u0442\u0432\u0443 \u043c\u043e\u0434\u0435\u043b\u044c \u0441\u0435\u0440\u0438\u0438 \u0441\u043e\u0431\u0440\u0430\u043d\u0430 \u043d\u0430 \u043f\u0440\u043e\u0441\u0442\u043e\u0440\u043d\u043e\u0439 PCB, \u043d\u043e \u0441\u0438\u0441\u0442\u0435\u043c\u0430 \u043f\u0438\u0442\u0430\u043d\u0438\u044f \u0443 \u043d\u0435\u0435 \u043f\u043e\u043f\u0440\u043e\u0449\u0435. \u0413\u0440\u0430\u0444\u0438\u0447\u0435\u0441\u043a\u043e\u043c\u0443 \u043f\u0440\u043e\u0446\u0435\u0441\u0441\u043e\u0440\u0443 \u043f\u0440\u0438\u043d\u0430\u0434\u043b\u0435\u0436\u0438\u0442 15-\u0444\u0430\u0437\u043d\u044b\u0439 VRM \u043f\u043e\u0434 \u0443\u043f\u0440\u0430\u0432\u043b\u0435\u043d\u0438\u0435\u043c \u0428\u0418\u041c-\u043a\u043e\u043d\u0442\u0440\u043e\u043b\u043b\u0435\u0440\u0430 Monolithic Power Systems MP29816. \u041f\u0438\u0442\u0430\u043d\u0438\u0435 \u043c\u0438\u043a\u0440\u043e\u0441\u0445\u0435\u043c \u043f\u0430\u043c\u044f\u0442\u0438 GDDR7 — \u0442\u0440\u0435\u0445\u0444\u0430\u0437\u043d\u043e\u0435, \u043d\u0430 \u043e\u0441\u043d\u043e\u0432\u0435 \u043a\u043e\u043d\u0442\u0440\u043e\u043b\u043b\u0435\u0440\u0430 MP2988. \u041e\u0431\u0430 \u0440\u0435\u0433\u0443\u043b\u044f\u0442\u043e\u0440\u0430 \u0443\u043a\u043e\u043c\u043f\u043b\u0435\u043a\u0442\u043e\u0432\u0430\u043d\u044b \u0442\u0440\u0430\u043d\u0437\u0438\u0441\u0442\u043e\u0440\u043d\u044b\u043c\u0438 \u0441\u0431\u043e\u0440\u043a\u0430\u043c\u0438 MP87993 \u0441 \u043f\u0440\u0435\u0434\u043f\u043e\u043b\u043e\u0436\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u043c \u043d\u043e\u043c\u0438\u043d\u0430\u043b\u043e\u043c \u0442\u043e\u043a\u0430 90 \u0410.<\/p>\n

Videom\u00e4lu kiipide Samsung K4VAF325ZC-SC28 m\u00e4rgistus n\u00e4itab l\u00e4bilaskev\u00f5imet 28 Gbit\/s.<\/p>\n

\u041a\u0430\u043a \u0438 Radeon RX 9070 XT \u0432 \u0438\u0441\u043f\u043e\u043b\u043d\u0435\u043d\u0438\u0438 GIGABYTE, \u0443\u0441\u043a\u043e\u0440\u0438\u0442\u0435\u043b\u044c Palit \u0438\u043c\u0435\u0435\u0442 \u0434\u0432\u0430 \u0432\u0430\u0440\u0438\u0430\u043d\u0442\u0430 BIOS — «\u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u0439» \u0438 «\u0442\u0438\u0445\u0438\u0439».<\/p>\n

\"Uus \"Uus <\/p>\n

⇡<\/a><\/noindex>#<\/a><\/noindex>Testimine, testimise metoodika<\/strong><\/h2>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Teststand<\/th>\n<\/tr>\n
CPU<\/td>\nAMD Ryzen 9 7950X3D (PBO +150 MHz, CU -20)<\/td>\n<\/tr>\n
Emaplaat<\/td>\nASUS ROG Crosshair X670E Hero<\/td>\n<\/tr>\n
Operatiivm\u00e4lu<\/td>\nG.Skill Trident Z5 Neo RGB (F5-6000J3040G32GX2-TZ5NR), 2 × 32 \u0413\u0431\u0430\u0439\u0442 (6200 \u041c\u0422\/\u0441, CL30)<\/td>\n<\/tr>\n
ROM<\/td>\nSolidigm P44 Pro, 2 TB<\/td>\n<\/tr>\n
Toiteplokk<\/td>\nCorsair AX1600i, 1600 W<\/td>\n<\/tr>\n
CPU jahutuss\u00fcsteem<\/td>\n\u041a\u0430\u0441\u0442\u043e\u043c\u043d\u0430\u044f \u0421\u0416\u041e (EK-Quantum Velocity² DDC 4.2 PWM D-RGB + EK-Quantum Surface X280M)<\/td>\n<\/tr>\n
Korpus<\/td>\nAvatud seista<\/td>\n<\/tr>\n
Operatsioonis\u00fcsteem<\/td>\nWindows 11 Pro<\/td>\n<\/tr>\n
GPU AMD tarkvara<\/td>\n<\/tr>\n
K\u00f5ik graafikakaardid<\/td>\nAMD Software Adrenalin Edition 25.2.1\/25.3.1<\/td>\n<\/tr>\n
GPU NVIDIA tarkvara<\/td>\n<\/tr>\n
K\u00f5ik graafikakaardid<\/td>\nNVIDIA GeForce Game Ready Driver 532.60\/532.70<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
M\u00e4ngud ilma kiirteekonnata<\/th>\n<\/tr>\n
M\u00e4ng<\/td>\nAPI<\/td>\nTestimise meetod<\/td>\nGraafika seaded<\/td>\n<\/tr>\n
Alan Wake 2<\/td>\nDirectX 12<\/td>\nOCAT, asukoht Bright Falls<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n
Black Myth: Wukong<\/td>\nDirectX 12<\/td>\nSisseehitatud benhmark<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n
Cyberpunk 2077<\/td>\nDirectX 12<\/td>\nSisseehitatud benhmark<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n
F1 24<\/td>\nDirectX 12<\/td>\nSisseehitatud benhmark, Monaco rada (vihm)<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n
Hogwarts Legacy<\/td>\nDirectX 12<\/td>\nOCAT, s\u00f5it k\u00e4ru kaudu Hogwarts'i teele<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n
Horizon Zero Dawn Remastered<\/td>\nDirectX 12<\/td>\nSisseehitatud benhmark<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n
Metro Exodus<\/td>\nDirectX 12<\/td>\nSisseehitatud benhmark<\/td>\nMaks. graafika kvaliteet; Varjude m\u00e4\u00e4r: 100%<\/td>\n<\/tr>\n
Red Dead Redemption 2<\/td>\nVulkan<\/td>\nSisseehitatud benhmark<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n
Returnal<\/td>\nDirectX 12<\/td>\nSisseehitatud benhmark<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n
Total War: WARHAMMER III<\/td>\nDirectX 11<\/td>\nSisseehitatud benhmark (Mirrors of Madness Benchmark)<\/td>\nMaks. graafika kvaliteet<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Ray tracing m\u00e4ngud<\/th>\n<\/tr>\n
M\u00e4ng<\/td>\nAPI<\/td>\nTestimise meetod<\/td>\nGraafika seaded<\/td>\nRaamide skaaleerimine<\/td>\n<\/tr>\n
AMD<\/td>\nIntel<\/td>\nNVIDIA<\/td>\n<\/tr>\n
Alan Wake 2<\/td>\nDirectX 12<\/td>\nOCAT, asukoht Bright Falls<\/td>\nMaks. graafika kvaliteet, k\u00f5rge ray tracing kvaliteet<\/td>\nFSR Tasakaalus<\/td>\nFSR Tasakaalus<\/td>\nDLSS Tasakaalus + Ray Reconstruction (+ Raami genereerimine)<\/td>\n<\/tr>\n
Black Myth: Wukong<\/td>\nSisseehitatud benhmark<\/td>\nMaks. graafika ja ray tracing kvaliteet (Path Tracing)<\/td>\nFSR Tasakaalus (+ Raami genereerimine)<\/td>\nXeSS Tasakaalus\/FSR Tasakaalus + Raami genereerimine<\/td>\nDLSS Tasakaalus (+ Raami genereerimine)<\/td>\n<\/tr>\n
Cyberpunk 2077<\/td>\nSisseehitatud benhmark (OCAT raamide genereerimiseks)<\/td>\nMaks. graafika ja ray tracing kvaliteet (Path Tracing)<\/td>\nFSR Tasakaalus (+ Raami genereerimine)<\/td>\nXeSS Tasakaalus\/FSR Tasakaalus + Raami genereerimine<\/td>\nDLSS Tasakaalus (Transformer Model) + Ray Reconstruction (+ Raami genereerimine)<\/td>\n<\/tr>\n
F1 24<\/td>\nSisseehitatud benhmark, Monaco rada (vihm)<\/td>\nMaks. graafika ja ray tracing kvaliteet<\/td>\nFSR Tasakaalus (+ Raami genereerimine)<\/td>\nXeSS Tasakaalus\/FSR Tasakaalus + Raami genereerimine<\/td>\nDLSS Tasakaalus (+ Raami genereerimine)<\/td>\n<\/tr>\n
Hogwarts Legacy<\/td>\nOCAT, s\u00f5it k\u00e4ru kaudu Hogwarts'i teele<\/td>\nMaks. graafika ja ray tracing kvaliteet<\/td>\nFSR Tasakaalus<\/td>\nXeSS Tasakaalus<\/td>\nDLSS Tasakaalus + Ray Reconstruction (+ Raami genereerimine)<\/td>\n<\/tr>\n
Indiana Jones and the Great Circle<\/td>\nOCAT, asukoht Sukhothai<\/td>\nMaks. graafika kvaliteet, k\u00f5rge ray tracing kvaliteet (Path Tracing); Ray Traced Shadows: Ainult p\u00e4ikesevalgus<\/td>\nFSR Tasakaalus (+ Raami genereerimine)<\/td>\nXeSS Tasakaalus\/FSR Tasakaalus + Raami genereerimine<\/td>\nDLSS Tasakaalus (Transformer Model) + Ray Reconstruction (+ Raami genereerimine)<\/td>\n<\/tr>\n
Metro Exodus Enhanced Edition<\/td>\nSisseehitatud benhmark<\/td>\nMaks. graafika ja ray tracing kvaliteet<\/td>\nN\/D<\/td>\nN\/D<\/td>\nDLSS Tasakaalus<\/td>\n<\/tr>\n
Returnal<\/td>\nSisseehitatud benhmark (OCAT raamide genereerimiseks)<\/td>\nMaks. graafika ja ray tracing kvaliteet<\/td>\nFSR Tasakaalus (+ Raami genereerimine)<\/td>\nXeSS Tasakaalus\/FSR Tasakaalus + Raami genereerimine<\/td>\nDLSS Tasakaalus (+ Raami genereerimine)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Enamikus m\u00e4ngudes kuvatakse keskmisi ja minimaalseid (n\u00e4itame 1. protsentiiljaotust) kaadrisagedusi individuaalsete kaadrite renderdamise aegade massiivist v\u00f5i kiirelt framerate'ist, mis on saadud sisseehitatud testimise abil. Erandiks on m\u00e4ngud, millel ei ole sisseehitatud testimist, ning testid, kus kasutatakse kaadrite genereerimist: nendes olukordades kasutame vahekaadri intervallide salvestamiseks programmi OCAT.<\/p>\n\n\n\n\n\n\n\n\n
T\u00f6\u00f6programmid<\/th>\n<\/tr>\n
Rakendus<\/td>\nBenchmark<\/td>\nSeaded<\/td>\n<\/tr>\n
Adobe Premiere Pro 25.x<\/td>\nPugetBench for Premiere Pro 1.1.1 (testide koostis lingil)<\/a><\/noindex><\/td>\nStandard (4K)<\/td>\n<\/tr>\n
Blender 4.x<\/td>\nDemo Agent 327 Barbershop Blenderi veebilehelt<\/td>\nCycles renderdaja<\/td>\n<\/tr>\n
Blackmagic Design DaVinci Resolve Studio 19.x<\/td>\nPugetBench for DaVinci Resolve 1.1.1 (testide koostis lingil)<\/a><\/noindex><\/td>\nStandard (4K); H.264\/HEVC kodeerimisre\u017eiim: Automaatne<\/td>\n<\/tr>\n
CAD rakendused<\/td>\nSPECviewperf 2020 v3.1<\/td>\n\u0420\u0430\u0437\u0440\u0435\u0448\u0435\u043d\u0438\u0435 \u044d\u043a\u0440\u0430\u043d\u0430: 3840 × 2160<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Video dekodeerimine (ffmpeg 5.x)<\/th>\n<\/tr>\n
Formaat<\/td>\nEraldusv\u00f5ime<\/td>\nKodeerimisvalikud<\/td>\nAPI<\/td>\n<\/tr>\n
H.264 (YUV 4:2:0, 8 bit\/kanal)<\/td>\n1920 × 1080<\/td>\nHigh Profile, L4.1<\/td>\nD3D11VA<\/td>\n<\/tr>\n
3840 × 2160<\/td>\nHigh Profile, L5.1<\/td>\n<\/tr>\n
HEVC (YUV 4:2:0, 8 bit\/kanal)<\/td>\n1920 × 1080<\/td>\nMain Profile, L4.0<\/td>\n<\/tr>\n
3840 × 2160<\/td>\nMain Profile, L5.0<\/td>\n<\/tr>\n
7680 × 4320<\/td>\nMain Profile, L6.0<\/td>\n<\/tr>\n
VP9 (YUV 4:2:0, 8 bit\/kanal)<\/td>\n1920 × 1080<\/td>\nN\/D<\/td>\n<\/tr>\n
3840 × 2160<\/td>\n<\/tr>\n
7680 × 4320<\/td>\n<\/tr>\n
AV1 (YUV 4:2:0, 8 bit\/kanal)<\/td>\n1920 × 1080<\/td>\nMain Profile, L4.0<\/td>\n<\/tr>\n
3840 × 2160<\/td>\nMain Profile, L5.0<\/td>\n<\/tr>\n
7680 × 4320<\/td>\nMain Profile, L6.0<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Video kodeerimine (ffmpeg 5.x)<\/th>\n<\/tr>\n
Formaat<\/td>\nEraldusv\u00f5ime<\/td>\nKodeerimisvalikud<\/td>\nAPI<\/td>\n<\/tr>\n
AMD<\/td>\nIntel<\/td>\nNVIDIA<\/td>\nAMD<\/td>\nIntel<\/td>\nNVIDIA<\/td>\n<\/tr>\n
H.264 (YUV 4:2:0, 8 bit\/kanal)<\/td>\n1920 × 1080<\/td>\n-c:v h264_amf -quality speed -coder cabac -refs 1 -b:v 3M<\/td>\n-c:v h264_qsv -preset veryfast -profile:v main -level 4.1 -b:v 3M<\/td>\n-c:v h264_nvenc -preset fast -coder cabac -refs 1 -b:v 3M<\/td>\nAMF<\/td>\noneVPL<\/td>\nNVENC<\/td>\n<\/tr>\n
3840 × 2160<\/td>\n-c:v h264_amf -quality speed -coder cabac -refs 1 -b:v 7.5M<\/td>\n-c:v h264_qsv -preset veryfast -profile:v main -level 5.1 -b:v 7.5M<\/td>\n-c:v h264_nvenc -preset fast -coder cabac -refs 1 -b:v 7.5M<\/td>\n<\/tr>\n
HEVC (YUV 4:2:0, 8 bit\/kanal)<\/td>\n1920 × 1080<\/td>\n-c:v hevc_amf -quality speed -b:v 3M<\/td>\n-c:v hevc_qsv -preset veryfast -tier main -b:v 3M<\/td>\n-c:v hevc_nvenc -preset fast -b:v 3M<\/td>\n<\/tr>\n
3840 × 2160<\/td>\n-c:v hevc_amf -quality speed -b:v 7.5M<\/td>\n-c:v hevc_qsv -preset veryfast -tier main -b:v 7.5M<\/td>\n-c:v hevc_nvenc -preset fast -b:v 7.5M<\/td>\n<\/tr>\n
7680 × 4320<\/td>\n-c:v hevc_amf -quality speed -b:v 20M<\/td>\n-c:v hevc_qsv -preset veryfast -tier main -b:v 20M<\/td>\n-c:v hevc_nvenc -preset fast -b:v 20M<\/td>\n<\/tr>\n
AV1 (YUV 4:2:0, 8 bit\/kanal)<\/td>\n1920 × 1080<\/td>\n-c:v hevc_amf -quality speed -b:v 3M<\/td>\n-c:v av1_qsv -preset veryfast -profile main -b:v 3M<\/td>\n-c:v hevc_nvenc -preset fast -b:v 3M<\/td>\n<\/tr>\n
3840 × 2160<\/td>\n-c:v hevc_amf -quality speed -b:v 7.5M<\/td>\n-c:v av1_qsv -preset veryfast -profile main -b:v 7.5M<\/td>\n-c:v hevc_nvenc -preset fast -b:v 7.5M<\/td>\n<\/tr>\n
7680 × 4320<\/td>\n-c:v hevc_amf -quality speed -b:v 20M<\/td>\n-c:v av1_qsv -preset veryfast -profile main -b:v 20M<\/td>\n-c:v hevc_nvenc -preset fast -b:v 20M<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

\u041c\u043e\u0449\u043d\u043e\u0441\u0442\u044c \u0432\u0438\u0434\u0435\u043e\u043a\u0430\u0440\u0442 \u0440\u0435\u0433\u0438\u0441\u0442\u0440\u0438\u0440\u0443\u0435\u0442\u0441\u044f \u043e\u0442\u0434\u0435\u043b\u044c\u043d\u043e \u043e\u0442 CPU \u0438 \u043f\u0440\u043e\u0447\u0438\u0445 \u043a\u043e\u043c\u043f\u043e\u043d\u0435\u043d\u0442\u043e\u0432 \u041f\u041a \u0441 \u043f\u043e\u043c\u043e\u0449\u044c\u044e \u0443\u0441\u0442\u0440\u043e\u0439\u0441\u0442\u0432\u0430 NVIDIA PCAT. \u0412 \u043a\u0430\u0447\u0435\u0441\u0442\u0432\u0435 \u043d\u0430\u0433\u0440\u0443\u0437\u043a\u0438 \u0434\u043b\u044f \u0442\u0435\u0441\u0442\u043e\u0432 \u043c\u043e\u0449\u043d\u043e\u0441\u0442\u0438 \u0438 \u0443\u0440\u043e\u0432\u043d\u044f \u0448\u0443\u043c\u0430 \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u0442\u0441\u044f \u0438\u0433\u0440\u0430 Cyberpunk 2077 \u043f\u0440\u0438 \u0440\u0430\u0437\u0440\u0435\u0448\u0435\u043d\u0438\u0438 3840 × 2160 \u0438 \u043c\u0430\u043a\u0441\u0438\u043c\u0430\u043b\u044c\u043d\u044b\u0445 \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u0430\u0445 \u043a\u0430\u0447\u0435\u0441\u0442\u0432\u0430 \u0433\u0440\u0430\u0444\u0438\u043a\u0438 (\u0431\u0435\u0437 \u0442\u0440\u0430\u0441\u0441\u0438\u0440\u043e\u0432\u043a\u0438 \u043b\u0443\u0447\u0435\u0439), \u0430 \u0442\u0430\u043a\u0436\u0435 \u0441\u0442\u0440\u0435\u0441\u0441-\u0442\u0435\u0441\u0442 FurMark \u0441 \u043d\u0430\u0438\u0431\u043e\u043b\u0435\u0435 \u0430\u0433\u0440\u0435\u0441\u0441\u0438\u0432\u043d\u044b\u043c\u0438 \u043d\u0430\u0441\u0442\u0440\u043e\u0439\u043a\u0430\u043c\u0438 (\u0440\u0430\u0437\u0440\u0435\u0448\u0435\u043d\u0438\u0435 3840 × 2160, MSAA 8x). \u0417\u0430\u043c\u0435\u0440\u044b \u0432\u0441\u0435\u0445 \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u043e\u0432 \u0432\u044b\u043f\u043e\u043b\u043d\u044f\u044e\u0442\u0441\u044f \u043f\u043e\u0441\u043b\u0435 \u043f\u0440\u043e\u0433\u0440\u0435\u0432\u0430 \u0432\u0438\u0434\u0435\u043e\u043a\u0430\u0440\u0442\u044b, \u043a\u043e\u0433\u0434\u0430 \u0442\u0435\u043c\u043f\u0435\u0440\u0430\u0442\u0443\u0440\u0430 GPU \u0438 \u0442\u0430\u043a\u0442\u043e\u0432\u044b\u0435 \u0447\u0430\u0441\u0442\u043e\u0442\u044b \u0441\u0442\u0430\u0431\u0438\u043b\u0438\u0437\u0438\u0440\u0443\u044e\u0442\u0441\u044f.<\/p>\n

⇡<\/a><\/noindex>#<\/a><\/noindex>Testimise osalised<\/strong><\/h2>\n

Toime testimises osalesid j\u00e4rgmised graafikakaardid:<\/p>\n