{"id":157401,"date":"2025-12-25T17:36:11","date_gmt":"2025-12-25T17:36:11","guid":{"rendered":"https:\/\/kamucalisani.net\/?p=157401"},"modified":"2025-12-25T17:36:11","modified_gmt":"2025-12-25T17:36:11","slug":"samsung-kendi-gpularini-gelistirmeyi-planliyor-amd-ile-isbirligi-sona-erebilir","status":"publish","type":"post","link":"https:\/\/kamucalisani.net\/index.php\/2025\/12\/25\/samsung-kendi-gpularini-gelistirmeyi-planliyor-amd-ile-isbirligi-sona-erebilir\/","title":{"rendered":"Samsung kendi GPU’lar\u0131n\u0131 geli\u015ftirmeyi planl\u0131yor: AMD ile i\u015fbirli\u011fi sona erebilir"},"content":{"rendered":"

<\/span> Samsung, 2021 y\u0131l\u0131nda Exynos i\u015flemcilerde kullan\u0131lacak GPU\u2019lar i\u00e7in AMD ile ortakl\u0131k kurdu\u011funu duyurmu\u015ftu. Bu i\u015f birli\u011fiyle geli\u015ftirilen GPU’lar Samsung’un \u00fcst seviye i\u015flemcilerinde kendine yer buldu. Ancak daha \u00f6nce de g\u00fcndeme geldi\u011fi gibi, Samsung\u2019un yak\u0131n gelecekte AMD ile yollar\u0131n\u0131 ay\u0131rarak tamamen kendi geli\u015ftirdi\u011fi bir GPU\u2019ya ge\u00e7ebilece\u011fi <\/strong>konu\u015fuluyor. <\/figure>\n

Exynos 2800, Samsung\u2019un kendi GPU\u2019sunu kullanan ilk i\u015flemci olabilir<\/b><\/p>\n

G\u00fcney Kore kaynakl\u0131 bir rapora g\u00f6re Samsung, AMD\u2019nin RDNA mimarisini kullanmay\u0131 b\u0131rakmay\u0131 ve tamamen \u015firket i\u00e7inde geli\u015ftirilen bir GPU mimarisine ge\u00e7meyi planl\u0131yor. Bu yeni GPU teknolojisinin Exynos 2800<\/strong> i\u015flemcisiyle birlikte tan\u0131t\u0131laca\u011f\u0131 belirtiliyor. Exynos 2800\u2019\u00fcn ise 2028 y\u0131l\u0131nda<\/strong> \u00e7\u0131kmas\u0131 beklenen Galaxy S28<\/strong> serisinde yer almas\u0131 bekleniyor. Samsung, hem i\u015flemciyi hem de GPU\u2019yu kendi b\u00fcnyesinde geli\u015ftirerek Galaxy telefon ve tabletlerde performans optimizasyonunu daha ileri bir seviyeye ta\u015f\u0131yabilir.<\/p>\n

Exynos 2200, Exynos 2400 ve Exynos 2500 modellerinde AMD taraf\u0131ndan geli\u015ftirilen GPU\u2019lar kullan\u0131lm\u0131\u015ft\u0131. Exynos 2600’\u00fcn AMD’nin RDNA mimarisini kullanan ancak Samsung’un System LSI \u015firketi taraf\u0131ndan geli\u015ftirilen ilk Exynos \u00e7ipi oldu\u011fu bildiriliyor.\u00a0Exynos 2800 ile birlikte ise Samsung\u2019un tamamen kendi mimarisine dayanan bir GPU\u2019ya ge\u00e7mesi bekleniyor.<\/p>\n

Yapay zekada GPU’nun \u00f6neminin artmas\u0131 kararda rol oynuyor<\/b><\/p>\n

Samsung\u2019un mobil GPU\u2019yu tamamen kendi b\u00fcnyesine alma y\u00f6n\u00fcnde ad\u0131m atmas\u0131n\u0131n temel nedeni, yapay zeka \u00e7a\u011f\u0131nda GPU\u2019nun \u00f6nemi giderek artmas\u0131<\/strong>. GPU\u2019lar, paralel i\u015flem avantajlar\u0131 sayesinde ekranda g\u00f6r\u00fclen hemen her \u015feyi \u00e7izen ve i\u015flemci i\u00e7erisinde video oynatma, g\u00f6r\u00fcnt\u00fc birle\u015ftirme, g\u00f6rsel \u00fcretimi gibi g\u00f6revleri \u00fcstlenen yar\u0131 iletkenler. Ayr\u0131ca n\u00f6ral i\u015flem \u00fcnitesini (NPU) destekleyen bir yard\u0131mc\u0131 yapay zeka h\u0131zland\u0131r\u0131c\u0131s\u0131\u00a0rol\u00fc de \u00fcstleniyor.<\/p>\n

Samsung\u2019un GPU\u2019yu kendi b\u00fcnyesinde geli\u015ftirme karar\u0131, genel ama\u00e7l\u0131 GPU\u2019lar\u0131n yapay zeka \u00f6zelliklerini tam anlam\u0131yla kar\u015f\u0131layamad\u0131\u011f\u0131<\/strong> d\u00fc\u015f\u00fcncesine dayan\u0131yor. Genel ama\u00e7l\u0131 GPU\u2019lar g\u00fc\u00e7l\u00fc olsa da, bir\u00e7ok marka ve cihazda \u00e7al\u0131\u015fmak zorunda olduklar\u0131 i\u00e7in Samsung yaz\u0131l\u0131m\u0131yla tam optimizasyon sa\u011flanam\u0131yor<\/strong>; bu da g\u00fc\u00e7 t\u00fcketimi ve i\u015flem kapasitesinde israfa yol a\u00e7\u0131yor. Buna kar\u015f\u0131n, \u015firket i\u00e7i GPU kullan\u0131m\u0131 Samsung\u2019un belirli i\u015f y\u00fckleri ve i\u015flevler i\u00e7in \u00f6zelle\u015ftirme yapmas\u0131na imkan tan\u0131yor. Google, Amazon ve Meta gibi devlerin AI h\u0131zland\u0131r\u0131c\u0131lar\u0131n\u0131 i\u00e7selle\u015ftirmesinin nedeni de bu.<\/p>\n

\u00c7ok say\u0131da m\u00fchendis i\u015fe al\u0131nd\u0131<\/b><\/p>\n

Rapora g\u00f6re Samsung, son \u00fc\u00e7 y\u0131l i\u00e7inde yar\u0131 iletken b\u00f6l\u00fcm\u00fcnde \u00e7ok say\u0131da GPU m\u00fchendisi i\u015fe ald\u0131. <\/strong>Bu m\u00fchendislere y\u0131ll\u0131k 300\u2013400 milyon Kore wonu (yakla\u015f\u0131k 207 bin \u2013 277 bin dolar) seviyesinde maa\u015flar \u00f6dendi\u011fi belirtiliyor. Ayr\u0131ca \u015firket k\u0131sa s\u00fcre \u00f6nce, daha \u00f6nce AMD, Broadcom ve Intel\u2019de g\u00f6rev yapm\u0131\u015f ve GPU alan\u0131nda tan\u0131nm\u0131\u015f bir isim olan John Rayfield\u2019\u0131 da kadrosuna katt\u0131.<\/p>\n

Samsung, Exynos 2800 ile kendi GPU\u2019sunu tan\u0131tt\u0131ktan sonra bu i\u015flemcileri otonom ara\u00e7 platformlar\u0131nda, insans\u0131 robotlarda ve ak\u0131ll\u0131 g\u00f6zl\u00fcklerde kullanabilir. Bunun yan\u0131nda, Broadcom ve Marvell\u2019in yapt\u0131\u011f\u0131 gibi; Amazon, Anthropic, Google, Microsoft ve OpenAI gibi \u015firketler i\u00e7in geli\u015ftirilen yapay zeka odakl\u0131 ASIC \u00e7\u00f6z\u00fcmlerinde de bu GPU mimarisinden faydalanmas\u0131 m\u00fcmk\u00fcn g\u00f6r\u00fcn\u00fcyor.<\/p>\n\n

Kaynak\u00a0 :\u00a0https:\/\/www.donanimhaber.com\/samsung-kendi-gpu-larini-gelistirmeyi-planliyor–200128<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

Samsung, 2021 y\u0131l\u0131nda Exynos i\u015flemcilerde kullan\u0131lacak GPU\u2019lar i\u00e7in AMD ile ortakl\u0131k kurdu\u011funu duyurmu\u015ftu. Bu i\u015f birli\u011fiyle geli\u015ftirilen GPU’lar Samsung’un \u00fcst seviye i\u015flemcilerinde kendine yer buldu. Ancak daha \u00f6nce de g\u00fcndeme geldi\u011fi gibi …<\/p>\n","protected":false},"author":1,"featured_media":157402,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[68,2991,7231,500,211],"class_list":["post-157401","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-teknoloji","tag-amd","tag-exynos","tag-gelistirilen","tag-gpu","tag-is"],"_links":{"self":[{"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/posts\/157401","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/comments?post=157401"}],"version-history":[{"count":1,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/posts\/157401\/revisions"}],"predecessor-version":[{"id":157404,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/posts\/157401\/revisions\/157404"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/media\/157402"}],"wp:attachment":[{"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/media?parent=157401"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/categories?post=157401"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/tags?post=157401"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}