Vijesti i novosti o Intel procesorima i čipsetima

[Disco]

[Mladenxy]

Ha ha Stevo rasturio Intel ko beba zvečku Ovi bi fakat morali nešto napraviti sa svojim reklamnim odjelom, ovo je živa komedija.

[The Exiled]

Bokte, ovo od Intela djeluje ko neki Monty Python skeč. Stvarno je sve umjetno, a čini se da sami ne vjeruju to kaj pričaju.

[Disco]

Stevo amer je jedan od jačih komedijaša sa ozbiljnom facom (video sa pastom za proc mi je i dalje najdraži ).
Ali izgleda da ga ekipa iz Intela pokušava skinut sa trona.

[zwer54]

Salu na stranu, no evo ganjao neke benchmarke po hwbotu i ugl sve i jedan 3d mark i slicno sa limitom na 4.5 ghz intel dominira, odnosno baca daleko bolje rezultate od amda lockanog na 4.5...

[The Exiled]

Ovo su brojke koje zapravo najviše muče Intel, dok se istovremeno nalaze u poprilično nezahvalnoj situaciji, lišeni doticaja s realnošću.

[Disco]

Zvuči dobro, no vidjet ćemo.

[zwer54]

U naslovu intel & intel i onda 5 minuta slusam o amd i nekim vpn reklamama... Nisam imao volje ni slusat dalje.

Jel sam to samo ja ili ima jos ljudi kojima je drazi pisani clanak od tih youtube sranja?

[Libertus]

Autor zwer54

Jel sam to samo ja ili ima jos ljudi kojima je drazi pisani clanak od tih youtube sranja?

+1

Rijetko gledam bilo što youtube kad je u pitanju review ili neki tutorial. Pisana riječ je efikasniji prijenos informacije, izvučem što meni treba umjesto da čekam njega dok ne složi nekoliko suvislih.

[The Exiled]

OK, veće punokrvne Core jezgre u sljedeće dvije generacije ne idu dalje od 8C/16T kombinacije, dok budu manjih Atom Gracemont nagurali koliko god mogu u nadi da će to biti dosta za Zen 4 i sve kaj slijedi.

[Starsylph]

Sve vise i pisani oblik sadrzi rekla-kazala pa čak i da sadrži tocnu informaciju, možeš se izgubiti. Kad sam pocetkom godine trazio neke informacije potpuno sam bio zbunjen internetskim sadrzajem.Trebalo bi izbaciti takozvana nagađanja iz profesionalnih članaka iz tehnologije.

za AL u videu navodi i piše da će k verzije imati recimo za i7 8c+4lc(little core).
a Raptor Lake generacija 8c/16lc.
te male jezgre će biti fizičke, pa me zanima dali je već negdje navedena učinkovitost u odnosu na recimo 8c/16t.
zapravo usporedba sa threads virtualkama, little core vs threads?

[Disco]

Autor The Exiled

OK, veće punokrvne Core jezgre u sljedeće dvije generacije ne idu dalje od 8C/16T kombinacije, dok budu manjih Atom Gracemont nagurali koliko god mogu u nadi da će to biti dosta za Zen 4 i sve kaj slijedi.

Pa šta nije rekao u videu da su male jezgre Skylake? Ili je to krnji Skylaek atom? Nisam taj dio baš pokopčao.

A što se tiče slušanja vs čitanja, ja npr radim neš po kući, BT sluje u ušima i slušam....ni ne gledam video.
Ali da, za neke stvari je tekst puno bolji, a grafovi/slike još najbolji.

[The Exiled]

Autor Disco

Pa šta nije rekao u videu da su male jezgre Skylake? Ili je to krnji Skylaek atom? Nisam taj dio baš pokopčao.

Male jezgre su zapravo bazirane na postojećim Atom modelima (1 - 2 - 3 - 4), a s posljednjom Gracemont arhitekturom, te Atom jezgre bi bez HT-a prema dosdašnjim vijestima i nagađanjima trebale biti u rangu orginalne Skylake arhitekture.

Znači, Alder Lake za desktop ne ide dalje od 8C/16T za veće Core jezgre, dok mobilni Alder Lake modeli idu samo do 6C/12T kombinacije za veće Core jezgre, a sve ostalo će onda popuniti manjim Atom jezgrama.

I tu se onda vraćamo na ono kaj pričamo već godinu dana - zanimljivo bude vidjeti kak bude sve skupa na kraju funkcioniralo i na koji način će se Core + Atom jezgre nostiti s punokrvnim Zen 3 i kasnije Zen 4 modelima.

Autor igofanatic

te male jezgre će biti fizičke, pa me zanima dali je već negdje navedena učinkovitost u odnosu na recimo 8c/16t. zapravo usporedba sa threads virtualkama, little core vs threads?

Sve jezgre, male i velike će biti fizičke, samo su veće bazirane na punokrvnoj Core arhitekturi, dok za manje idu na Atom koje nemaju HT.

Bez toga nemaju (kao AMD sa Zenom) mogućnost složiti toliki broj jezgri, a da potrošnja i zagrijavanje ne odu još više, jer sad su s Alder Lakeom opet na 200W+.

Za konkretne usporedbe budemo morali pričaketi prve službene recenezije.

[The Exiled]

Intel Xeon Sapphire Rapids supports HBM and Ponte Vecchio OAM

Intel’s new Sapphire Rapids Xeon Scalable with High-Bandwidth Memory (SPR-HBM) will be coming to market. Rather than hide it away for use with one particular hyperscaler, Intel has stated to AnandTech that they are committed to making HBM-enabled Sapphire Rapids available to all enterprise customers and server vendors as well. These versions will come out after the main Sapphire Rapids launch, and entertain some interesting configurations. We understand that this means SPR-HBM will be available in a socketed configuration. Intel states that SPR-HBM can be used with standard DDR5, offering an additional tier in memory caching. The HBM can be addressed directly or left as an automatic cache we understand, which would be very similar to how Intel's Xeon Phi processors could access their high bandwidth memory. Alternatively, SPR-HBM can work without any DDR5 at all. This reduces the physical footprint of the processor, allowing for a denser design in compute-dense servers that do not rely much on memory capacity (these customers were already asking for quad-channel design optimizations anyway).

In terms of CPUs, Intel’s dominance is weakening and AMD is on the rise, as we expected. And once the exascale machines hit, AMD will have a large percentage of the GPU compute and even the CPU compute because of the hybrid architectures in use. In June 2018, Intel had 95 percent share of the CPUs used in the Top500 rankings, and in June 2019 it rose to 95.6 percent and then fell to 94.2 percent in the June 2020 list. In the November 2020 list, Intel’s CPU share of the supercomputing rankings dropped to 91.8 percent, with AMD rising to 4.2 percent, and in the June 2021 rankings Intel is down to 86.2 percent of the systems and AMD is up to 9.8 percent of systems. With all of the big pre-exascale systems and exascale systems coming down the pike based on AMD Epyc processors in their hosts, AMD’s share of machines is going to rise, but its share of CPU cores is going to explode. And the same thing is going happen with the system count and GPU count for AMD relative to Nvidia starting with the November 2021 list, when the Frontier system at Oak Ridge is expected to leap ahead of Fugaku and take the top spot.

Intel Socket LGA-1700 and LGA-1800 in detail – Exclusive data and drawings for the new CPUs starting with Alder Lake

First, let’s take a look at the new socket compared to its predecessors and consider the obvious differences. I don’t even have to explain much, because you can see it at first glance by the dimensions and the shape. In this context, the backplate is also interesting, so I picked out the relevant construction drawing of the last three LGA sockets. The first thing you might notice is the new heatspreader (IHS), which is now rectangular instead of square. This has a big influence on the optimal positioning of the heatpipes in air coolers and is especially in coolers with DHT (Direct Heat Touch), so without a real heatsink, quite an important factor for the possible performance. Because then only the ground heatpipes cool. If these are unfavorable, that’s it with the performance, because we know exactly this problem from AMD’s Ryzen CPUs and the asymmetrically placed dies.

Izvor: ServeTheHome, AnandTech, HPCwire, The Next Platform i Igor's LAB

[Starsylph]

LGA 1800, to je poslije Alder lake ?

[The Exiled]

LGA-1700 za 10nm Alder i Raptor Lake, a onda potencijalno LGA-1800 za 7nm Meteor Lake i nasljednike. Sve u svemu ista priča kao i do sad.

Eventualno dvije generacije na istom socketu uz obaveznu promjenu čipseta, pa onda opet sve kompletno novo.

[Disco]

Autor The Exiled

LGA-1700 za 10nm Alder i Raptor Lake, a onda potencijalno LGA-1800 za 7nm Meteor Lake i nasljednike. Sve u svemu ista priča kao i do sad.

Eventualno dvije generacije na istom socketu uz obaveznu promjenu čipseta, pa onda opet sve kompletno novo.

Nekako sam uvjeren da će i AMD napravit isto sa AM5 nadalje...dok se Intel i dalje čini malo pogubljen.

[The Exiled]

IMHO AMD bude za AM5 napravil isto kao i za sadašnji AM4 socket, tj. sljedećih nekoliko godina bude AM5 LGA-1718 itekak aktualan.

Samo kaj budu po prohtjevima OEM ekipe, opet složili par generacija čipseta, mada tehnički sve paše, ali - umjetna segmentacija + pare.

Apple and Intel become first to adopt TSMC's latest chip tech

Apple and Intel have emerged as the first adopters of Taiwan Semiconductor Manufacturing Co.'s next-generation chip production technology ahead of its deployment as early as next year. According to TSMC, 3-nm technology can increase computing performance by 10% to 15% compared with 5-nm, while reducing power consumption by 25% to 30%. Apple's iPad will likely be the first devices powered by processors made using 3-nm technology, sources said. The next generation of iPhones, which are to roll out next year, are expected to make use of the intermediate 4-nm tech for scheduling reasons. Intel, America's biggest chipmaker, is working with TSMC on at least two 3-nm projects to design central processing units for notebooks and data center servers in an attempt to regain market share it has lost to Advanced Micro Devices and Nvidia over the past few years. Mass production of these chips is expected to begin by the end of 2022 at the earliest. For Intel, which both designs and manufactures chips, the collaboration with TSMC is aimed at tiding the company over until it can get its own in-house production technology on track. The company has delayed the introduction of its own 7-nm production technology to around 2023, well behind Asian rivals TSMC and Samsung Electronics. The release of Intel's latest Xeon processors powered by the company's 10-nm technology has also been delayed from the end of this year to the second quarter of next year, the company said this week.

Izvor: Nikkei Asia