PC Ekspert Forum

PC Ekspert Forum (https://forum.pcekspert.com/index.php)
-   NVIDIA (https://forum.pcekspert.com/forumdisplay.php?f=7)
-   -   Vijesti, novosti i glasine o nadolazećim nVidia grafičkim karticama (https://forum.pcekspert.com/showthread.php?t=301561)

Odinarius 23.07.2021. 08:26

Sub-400$ gotovo niti ne postoji u ovoj generaciji, tako da je ovo samo potvrda postojećeg stanja.

Disco 24.07.2021. 07:41


Dr. Strange 24.07.2021. 17:12

Jednom kad se pomaknu granice neće se nikad vratiti na staru razinu, plus još što je sve skuplje i kompleksnije prelaziti na niži node i izvlačiti performanse, treba prihvatiti to i gotovo.

demetrius 24.07.2021. 22:18

https://videocardz.com/newz/nvidia-g...using-tsmc-5nm

Tajl3r 04.08.2021. 17:31

https://twitter.com/3dfxofficial/sta...14559870455809

smh :stoopid:

thunder11 18.08.2021. 15:35

evo "malo" trača i špekulacija:

https://www.gamestar.de/artikel/rtx-...h,3372183.html

Disco 18.08.2021. 15:41

Citiraj:

Autor thunder11 (Post 3539639)

Odlično, trebat će mala elektrana sa frižiderom betnonskom armiranom konstrukcijom. :zdero2:

Manuel Calavera 20.09.2021. 14:14

Citiraj:

The Hopper GPU, which is primarily aimed at the Datacenter & AI segment, is allegedly taping out soon and will feature an MCM architecture. NVIDIA won't be using an MCM design on its Ada Lovelace GPUs so they will keep the traditional monolithic design.

AMD on the other hand is going on board with offering both MCM and Monolithic chips in its RDNA 3 and CDNA 2 families. The CDNA 2 GPUs will be MCM only while RDNA 3 will feature a mix of MCM and monolithic design as detailed here. It isn't confirmed which process node each company will be relying on but based on speculation, AMD's RDNA 3 & CDNA 2 family is going to be a mix of 6nm & 5nm nodes featuring new and refreshed GPUs while NVIDIA is expected to utilize TSMC's 5nm process node for its Ada Lovelace GPUs though it is not confirmed if those will be fabricated on the N5 or N5P node. Intel, on the other hand, may also rely on TSMC's 6nm process node for its own ARC Alchemist lineup of GPUs which enter mass production later this year and launch in Q1 2022.

Dr. Strange 20.09.2021. 17:14

Dostupnost će biti ekstra koma, sve 3 kompanije na 5/6nm procesu.

Antony 20.09.2021. 18:06

Citiraj:

Autor Dr. Strange (Post 3547321)
Dostupnost će biti ekstra koma, sve 3 kompanije na 5/6nm procesu.


TSMC već odavno nije bottleneck, već supstrati. AMD/Intel/nvidia svaki imaju svoj supply chain što se toga tiče. To je jedan od razloga zašto bi npr. štancanje DG2 GPU-ova na N6 nodeu trebalo poboljšati dostupnost grafa pošto Intel ima jako dobar supply chain (MLID je to već objašnjavao u par svojih videa).


Također -> https://semiwiki.com/semiconductor-m...-versus-apple/


Citiraj:

To be clear, wafer agreements are signed 2-3 years before the chip makes it into HVM and TSMC can build fabs faster than that so there will be no N3 shortages for anyone who signed a wafer agreement (apple, AMD, NVIDIA, QCOM, etc…). If they need more chips than what they signed up for, which happens, there may be shortages. This is how TSMC and the foundry business works. It’s all about the wafer agreements.

RaneZ 21.09.2021. 14:56

Citiraj:

Autor Dr. Strange (Post 3534269)
Jednom kad se pomaknu granice neće se nikad vratiti na staru razinu, plus još što je sve skuplje i kompleksnije prelaziti na niži node i izvlačiti performanse, treba prihvatiti to i gotovo.

To govorimo nas par vec 5 godina ali uporno se pojavljuju "kad padne cijena/pričekati ću..." odgovori :lol2: :chears:

Nema više nazad. Može samo gore uz sve sto se atm dogadja u svijetu.

demetrius 22.09.2021. 18:46

https://videocardz.com/newz/nvidia-g...specifications

samo rumorsi ali najvjeroatnije će bit tako da koštat će te isto ko 3060 ti možda i više 3060 super sa više cuda coresa i 12 gb rama.

Disco 03.12.2021. 08:50

Izgleda da gosn jakna ipak nije baš toliko moćan protiv mobilne ekipice (Apple, Samsung, Qualcom...):



I dobro da je deal propao, bio bi to opak monopol.

Dr. Strange 03.12.2021. 16:45

Citiraj:

Autor Disco (Post 3565892)
Izgleda da gosn jakna ipak nije baš toliko moćan protiv mobilne ekipice (Apple, Samsung, Qualcom...):



I dobro da je deal propao, bio bi to opak monopol.

Šta je on stvarno bio toliko optimističan i mislio da će deal proći? Pa svaki laik je znao da će to propasti. ARM mora ostati neutralan, a znamo koliko je Nvidia draga prema konkurenciji, i kakvim se sve metodama služe.

The Exiled 03.12.2021. 18:44

Š.T.E.T.A.:):D, jer riječ je o preduzeće koje voli da im cvate cveće i sad su ih zakinuli za inovacije kakve svijet još nije biJo vidiJo, a sad ni neće.:frend:

Kaj se Toma tiče, lagano postaje karikatura samog sebe, kad izdaje ovakve video uratke u kojima opet i po ko zna koji put mora naglasiti da je sve ovo već prije prvi saznal.

Jebeno, nema kaj, i onda od jučer kad je cijeli net pun vijesti o tome da FTC blokira nVidia + ARM selpo, dajte da vam Tom veli ekskluzivu kroz izvore iz UK vlade i vojnih satelita.:chears:

Disco 06.12.2021. 07:41

Citiraj:

Autor The Exiled (Post 3566020)
Š.T.E.T.A.:):D, jer riječ je o preduzeće koje voli da im cvate cveće i sad su ih zakinuli za inovacije kakve svijet još nije biJo vidiJo, a sad ni neće.:frend:

Kaj se Toma tiče, lagano postaje karikatura samog sebe, kad izdaje ovakve video uratke u kojima opet i po ko zna koji put mora naglasiti da je sve ovo već prije prvi saznal.

Jebeno, nema kaj, i onda od jučer kad je cijeli net pun vijesti o tome da FTC blokira nVidia + ARM selpo, dajte da vam Tom veli ekskluzivu kroz izvore iz UK vlade i vojnih satelita.:chears:

A izgleda da i Jim-a i njega hebe ego.
Tj Jim se više volio obračunavat sa trolovima.
Ali dobro, meni su uvijek fora za poslušati.

The Exiled 06.12.2021. 09:21

IMHO Toma najviše muči to da uvijek (ne) može reći da je prvi imal te informacije i to stvarno s vremenom postaje sve veći fokus, nego informacije s kojima raspolaže.

Odinarius 06.12.2021. 10:03

Meni je malo dosadan s onim "one of my best sources...", sve za različite teme i područja. Nemaju neke vrijednosti ti leakovi jer 50% budu točni, 50% ne, zabavno za poslušati makar nekakvu linije spekulacije i to je to. No tu i tamo ubode sa podcastom (Broken Silicon). I tu je fifty fifty što se tiče gostiju, no ponekad budu zbilja dobri, bude jasno u prvih 5min da li ću potrošiti 2h na to.
One koji su samo on i brat sam prestao slušati.

The Exiled 06.12.2021. 10:47

Da, brat mu je davež koji je tu samo zato, jer može, a kaj se gostiju tiče, stvarno su super u dosta slučajeva. Jedino kaj Tom neovisno o gostu uporno gura gaming tematiku.

Mladenxy 06.12.2021. 12:14

Gora ono što smatra da će osigurati klikove.

delavi 06.12.2021. 13:23

Problem je kaj ovi "ispod 35e" gotovo slijepo vjeruju svakoj budali na youtubeu. To je baš problem generacije. Mi smo rasli bez tih svih sranja pa smo se sami morali naučiti misliti haha

Disco 09.12.2021. 20:13

GeForce RTX 3090 Ti to launch on January 27th, RTX 3070 Ti 16GB January 11th, RTX 3050 January 27

Blue Spirit 04.01.2022. 07:08

Nevidija bome duboko posegnula u kasu da bi osigurala buducnost i konurentnost sa TSMC -ovim 5nm protiv nadolazecih kartica iz AMD-a koje ce također biti na TSMC -ovoj 5nm litografiji.

Citiraj:

NVIDIA is reportedly coughing up big money to TSMC (Taiwan Semiconductor Manufacturing Company) to ensure that it secures plenty of next-gen 5nm wafers for its upcoming Ada Lovelace GPUs that will power the next-gen GeForce RTX 40 series graphics cards.

But in new rumors from MyDrivers, NVIDIA is paying much more than usual to get 5nm wafer from TSMC with the company reportedly paying TSMC a large $1.64 billion in Q3 2021... but that increases to $1.79 billion in Q1 2022 as they ramp up to the Ada Lovelace GPU being made and printed onto next-gen GeForce RTX 40 series GPUs.

A larger long-term deal has also reportedly been struck, which would see NVIDIA coughing up $6.9 billion to TSMC in order to secure next-gen 5nm wafers for Ada Lovelace. We've heard before that NVIDIA's next-gen GeForce RTX 40 series GPUs would be made by TSMC -- and not Samsung -- and its upcoming 5nm node.

MyDrivers reports: "According to industry sources, TSMC's requirements for Apple, MediaTek, AMD and other three major customers are relatively low. They do not need to pay too much deposit in advance to stabilize production capacity. Customers like NVIDIA need to pay huge advance payments in advance if they want to obtain 5nm production orders".

Samsung has been making Ampere GPUs for NVIDIA and its GeForce RTX 30 series, but we all know they aren't the most power-efficient GPUs when you get to the high-end. AMD has hit its stride with its RDNA 2 GPU architecture, with TSMC baking their Radeon RX 6000 series chips on the 7nm node -- the same node used to make the semi-custom chips inside of the PlayStation 5 and Xbox Series X consoles -- and they're fantastic with power.

NVIDIA is making a move that it needs to in order to... well, frankly survive. The company makes its bread and butter from GPUs and if it can't sell enough -- especially of the very best, which TSMC will be capable of providing with its new N5 node -- then well, that's not going to be good at all. NVIDIA can't have another misfire as things will be radically different for Team Green in 2022, 2023, and beyond.

AMD will have its next-gen RDNA 3 architecture here with an MCM-based design -- MCM being a multi-chip module, so multiple GPU chiplets on the same card, similar to how the Zen CPU chiplets are on the Ryzen CPUs -- and will offer some truly monster performance on the next-gen Navi 31-based Radeon RX 7900 XT.

Not only that, but the GPU market is going to be completely different this time next year with CPU giant Intel joining the GPU market with its Xe-based Arc Alchemist. Later in the year we'll see how Intel has gone, and what comes next with Intel's Arc GPUs, and then in 2023 we've been hearing that is when things will really ramp for Xe and Arc... and NVIDIA knows that.

It gets worse, as the ongoing pandemic has been wreaking havoc on every single industry -- and the tech industry hasn't been shy from being slammed into the ground. The entire supply chain is f***ed right now, and Taiwan is in the middle of a shit-flinging fight with China + Japan, with the threat of nuclear strikes from China on Taiwan if Japan intervenes in any form.

NVIDIA's next-gen Ada Lovelace GPU and the flagship GeForce RTX 4090 or GeForce RTX 4090 Ti should be capable of 4K 120FPS gaming without a problem, but I think next-gen 8K 60FPS and better yet, 8K 120FPS gaming to be in the sights of NVIDIA. Maybe not raw, but with a next-gen DLSS 3.0 up its sleeve... it's entirely possible, and that is very, very exciting.

AMD's next-gen Navi 31 GPU has recently taped out, the very first MCM-based GPU that should form into a new flagship Radeon RX 7000 series graphics card in the second half of 2022. NVIDIA will be fighting back with the GeForce RTX 40 series and new flagship GeForce RTX 4090, which should offer 2x the performance and power consumption of the GeForce RTX 3090.

Read more: https://www.tweaktown.com/news/83634...022/index.html

The Exiled 04.01.2022. 11:30

Da, to je nastavno na ovu priču, a sad kad su svi fino ovisni o TSMC-ovim kapacitetima, nVidiji i Intelu ostaje koliko su voljni platiti za usluge koje se nude.:)

demetrius 24.01.2022. 19:19

https://videocardz.com/newz/gpu-pric...ing-to-decline

https://www.guru3d.com/news-story/nv...in-europe.html

:D

jel rastu ili padaju??

mislio sam reć guru3d objavljuje da cjene rastu dok videocardz objavljuje da cjene padaju što ćemo naravno vidjet kroz 2-3 mjeseca ili/i prije.

The Exiled 24.01.2022. 19:29

Za koji dan stiže RTX 3050 za kakvih 5.000 kn, dok istovremeno nudi GTX 1660 rang performanse i to će samo dodatno potvrditi da budžet segment stvarno nudi najbolji omjer uloženog i dobivenog.:):D

Kaj se samih cijena tiče, tj. pitanja da li cijene grafičkih kartica rastu ili padaju, odgovor je tak jasno - da.:frend::goood:
EDIT:
Ma zajebancija s moje strane kaj se cijena tiče:), ali nažalost izgleda da nema kraja ovom ludilu u neko skorije vrijeme.

demetrius 24.01.2022. 19:47

stara cjena za rtx 3060 ti 419 eura nova cjena 443 eura di to ima rtx 3060 ti za 443 eura??

The Exiled 24.01.2022. 20:13

Nema, to je već dulje vrijeme kartica od 1.000€, a MSRP je tu reda-radi.

RimtuTiTuki 25.01.2022. 12:04

Citiraj:

Autor demetrius (Post 3579109)
jel rastu ili padaju??

Oboje :D

Nvidija diže MSRP, a cijene padaju jer pada i potražnja za grafičkim.

Ali s obzirom da su cijene jako prenapuhane, u dućanima cijene padaju bez obzira štoi nvidija pokušava naplatiti koliko može

sakomako 25.01.2022. 18:12

Ove visoke cijene će samo većinu korisnika natjerati da kupuju CPU koji u sebi ima uključenu i grafiku. Koliko smisla imati za jednu grafulju dati od 7K kuna naviše da bi se na njoj igralo?


Druga stvar je kada se kupuje nešto za posao i kada ti je to sredstvo za rad.


I nije samo u pitanju visoka cijena već i period kroz koji će to što si kupio zastarjeti. A mnoge od najavljenih mogućnosti i potencijala same grafičke kartice možda se nikada neće niti iskoristiti u budućim igrama jer će u međuvremenu doći nešto bolje i ove mogućnosti će postati zastarjele.

The Exiled 30.01.2022. 12:42

nVidia Hopper might have huge 1000 mm² die monolithic design:kafa:
Citiraj:

Hopper is still sporting a classic monolithic die design despite previous rumors, and it appears that nVidia's performance targets have led to the creation of a monstrous, ~1000 mm² die package for the GH100 chip, which usually maxes out the complexity and performance that can be achieved on a particular manufacturing process. This is despite the fact that Hopper is also rumored to be manufactured under TSMC's 5nm technology, thus achieving higher transistor density and power efficiency compared to the 8nm Samsung process that nVidia is currently contracting. At the very least, it means that the final die will be bigger than the already enormous 826mm² of nVidia's GA100.

If this is indeed the case and nVidia isn't deploying a MCM (Multi-Chip Module) design on Hopper, which is designed for a market with increased profit margins, it likely means that less profitable consumer-oriented products from nVidia won't be featuring the technology either. MCM designs also make more sense in nVidia's HPC products, as they would enable higher theoretical performance when scaling - exactly what that market demands. Of course, nVidia could be looking to develop an MCM version of the GH100 still; but if that were to happen, the company could be looking to pair two of these chips together as another HPC product (rumored GH-102). ~2,000 mm² in a single GPU package, paired with increased density and architectural improvements might actually be what nVidia requires to achieve the 3x performance jump from the Ampere-based A100 the company is reportedly targeting.
Izvor: TechPowerUp

The Exiled 24.02.2022. 18:55

RTX 4000 rumors: September release, over 800W TDP:kafa:
Citiraj:

nVidia’s upcoming RTX 4000 series of graphics cards, codenamed Lovelace, are rumored to arrive this September. But in addition to availability and price concerns, you’ll probably need a new PSU for the high-end models, which are said to consume over 800W of power. We heard reports last year claiming the RTX 4000 cards would use some of the most power-hungry GPUs we’ve ever seen. That’s especially true of the AD102 that’s rumored to have a TGP of 800W or more. The latest rumors come from prolific leakers Kopite7kimi and Greymon55, both of whom say they’ve heard claims that the AD102 GPUs will have scarily high TGPs. The former thinks the RTX 4080 will carry a 450W TGP and the RTX 4080 Ti will be 600W. The RTX 4090, meanwhile, will come with a monstrous 800W rating. According to a report from Hardware Times, nVidia is prepared to pay dearly for the ability to meet growing demand for increasingly powerful graphics cards. Specifically, it may be offering up to $10 billion to TSMC for a significant chunk of its 5nm manufacturing capacity. Other companies tapping TSMC’s N5 process node are Bitmain, AMD, and Apple.
Izvor: TechSpot

Lewis 24.02.2022. 18:57

ono dvije grafe u boostu + CPU = 2kW, strasno :(

The Exiled 24.02.2022. 19:02

Iskreno, koliko god pretjerano zvučalo, ne bi me čudilo da budu tu negdje, jer već sad imaju 450W i 500W BIOS-e za RTX 3090 modele.

S ovom novom opet idu na ogroman monolitan čip, dok AMD već ima spreman i funkcionalan MCM, tj. chiplet dizajn za RDNA 3.

Disco 24.02.2022. 20:33

Ima video od od Toma iz MLID koji misli da to nije istina, i kao jedan od glavnih razloga navodi da bi to bilo nemoguće hladit ako srče 800+ wata. Pa to je imho apsolutno suluda potrošnja za igranje…uz skupi HW, trebat ce imat i za struju. Nadam se da nije istina, i nadam se da ce onda bar AMD imati nesto normalno.
E da Tom misli da su to mozda i serverske grafe ili ona cudesa za AI, ili je to isto, ne sjecam se sto je točno rekao

The Exiled 24.02.2022. 20:42

Citiraj:

Autor Disco (Post 3585990)
Ima video od od Toma iz MLID koji misli da to nije istina, i kao jedan od glavnih razloga navodi da bi to bilo nemoguće hladit ako srče 800+ wata. Pa to je imho apsolutno suluda potrošnja za igranje…uz skupi HW, trebat ce imat i za struju. Nadam se da nije istina, i nadam se da ce onda bar AMD imati nesto normalno.
E da Tom misli da su to mozda i serverske grafe ili ona cudesa za AI, ili je to isto, ne sjecam se sto je točno rekao

Da, pogledal sam i sve to stoji, ali velim ništa me ne bi začudilo da slože takve potrošače, ne mora čak ni biti tih nagađanih 800W+, dosta je da bude iznad ovih 450W ili 500W koliko gutaju sadašnji high-end RTX modeli. HPC/AI/ML modeli nadolazećih kartica bi stvarno mogli i trošiti čuda tehnike uz HEP na brzom biranju, jer njima nije problem po defaultu služiti vodeno hlađenje. Kaj se AMD RDNA 3 tiče, sve ovo kaj nVidia priprema opet s monolitnim čipom, je baš zato, jer isto ko i Intel najlakše budu najboljiJi kad se potrošnja nafrlji do kraja, pa kud puklo da puklo. Tom se isto na to osvrnul da nVidia već sad zna da RDNA 3 MCM dizajn uz bolje performanse nudi i bolju učinkovitost, pa da onda odmah u startu našpanaju svoje čisto za svaki slučaj. Sve u svemu, budemo pričekali da vidimo kaj imaju spremno, ali kak god se okrene, nove Radeon i GeForce kartice budu trošile više, jedino kaj bi moglo ispasti da nVidia stvarno pretjera.
EDIT:
Citiraj:

GPU giant nVidia is investigating a potential cyberattack:kafa:
Citiraj:

US chipmaker giant nVidia confirmed today it's currently investigating an "incident" that reportedly took down some of its systems for two days. Systems impacted in what looks like a cyberattack include the company's developer tools and email systems, as first reported by The Telegraph. The reported outage is the result of a network intrusion, and it is still not known if any business or customer data was stolen during the incident. nVidia told BleepingComputer that the nature of the incident is still being evaluated and that the company's commercial activities were not affected. An insider has described this incident as having "completely compromised" nVidia's internal systems.
Izvor: BleepingComputer

Eladio 26.02.2022. 20:23

Nadam se da se jako varam, hm, moj savjet onima koji cekaju bi bio kupite grafu sad dok ne padne i Taiwan..
Brijem da nize cijene i bolju dobavljivost nego sto sad imamo, nećemo imati tako skoro :hitthewal:

Lewis 26.02.2022. 20:52

Ja sam kupio jos dvije 3090 dok nije nestasica :D

Eladio 27.02.2022. 13:58

Nisi me bas shvatio.. geopoliticki nemir uz rat u Ukrajini i mozebitni novi ratovi (Kina/Taiwan) te sankcije koje idu u paketu uz rat ne ide na ruku IT trzistu... to je jedan novi faktor... naravno nisam baba vanga, ali samo nagadjam da tko očekuje još pad cijena, mogao bi se neugodno iznenaditi ako krenu nove nestašice i poskupljenja

MadBoY 27.02.2022. 14:12

Ukrajina proizvodi 70% neona, koji se koristi u proizvodnji chipova. Bez obzira kolike zalihe proizvođači imaju, ovo će im biti izgovor za dizanje cijena.


Sva vremena su GMT +2. Sada je 01:40.

Powered by vBulletin®
Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
© 1999-2024 PC Ekspert - Sva prava pridržana ISSN 1334-2940
Ad Management by RedTyger