Sunday, 17 January 2021

MacBook is expected to be equipped with an upgraded version of M1: Apple self-developed GPU exposure


 It is reported that a new generation of MacBook is expected to be equipped with an upgraded version of M1: Apple self-developed GPU exposure


During the online communication at CES2021, CEO Si Ruibo was asked whether he would consider outsourcing the production of some chips, and he said "maybe". Afterwards, many news pointed out that Intel has selected TSMC, foundry 7nmDG2 independent display, 5nm Core i3 and even 3nm chips.


However, with Intel's lightning change of coach now, Si Ruibo's decision seems to be withdrawn.

According to media reports, Si Ruibo told employees that the company's idea of ​​outsourcing production will have to wait for Pat Gelsinger (Pat-Kissinger) to make a decision.


Once Kissinger puts aside outsourcing or does not plan to outsource at all, TSMC’s planned $28 billion record capital expenditure for 2021 can be embarrassing.


Kissinger is the first CTO in Intel's history. Starting with the 80486 processor, he has participated in the development of too many generations of Xeon and Core processors, and he also defined Intel's Tick-Tock strategy. Before returning to Intel, Pat Kissinger was VMware CEO.


In addition, in 2013 and 2018, Kissinger was considered as a CEO candidate by the Intel board of directors, but he publicly stated that he did not want this position.

Friday, 15 January 2021

AMD Does Not Believe Its Mobility Processors Require PCIe 4.0 Right Now

AMD has been the darling of the technology sector for quite a while now, pushing the envelope for technological advancement ever forward, but it looks like the company feels that its next-generation Ryzen 4000 mobility processors do not require PCIe 4.0 right now and will keep that particular platform at PCIe 3.0. This is interesting because Ryzen 4000 series processors, while being on the 7nm process, also do not contain Navi graphics as far as the GPU IP e5 2667v4 goes. AMD EPYC 7702

AMD: Ryzen 4000 series mobility processors will not support PCIe 4.0

The interview was conducted by Japenese publication 4gamer.net

and was attended by AMD's Scott Stankard (Senior Product Manager, Client Business Unit). AMD has been aggressively pushing PCIe 4.0 on its desktop side of things and this is one of the biggest selling points of its Zen 2 based processors right now. It was, therefore, eyebrow-raising, to see AMD not use PCIe 4.0 on its mobility side of things as well and give a response that seems very synonymous to the ones Intel gave for not pushing technology envelopes.

Here is a translation of the pertinent part of the interview that I was able to get from a friend:

Interviewer: The Zen 2 generation desktop platform adopted PCI Express (PCIe) 4.0 as one of its features. What about Ryzen 4000 (series procesosrs) for notebook PCs?

Scott Stankard: [Ryzen 4000 series] is compatible with PCIe 3.0. This is because in our judgement there aren't many scenarios [on the notebook side of things] that are able to adequately utilize PCI 4.0.

AMD is, of course, correct in their statement that there aren't many situations in a mobility environment that would necessitate the use of PCIe 4.0, but this has never stopped the company before. In fact, the fact that Ryzen 4000 mobility processors feature PCIe 3.0 along with the older Vega IP in their GPU (instead of the newer Navi IP) looks more and more like the company has somewhat compromised as far as mobility goes. This might not be a bad thing as far as sales and market shares go (because it would allow a lower ASP and broader adoption) but is certainly enough to raise eyebrows in enthusiasts.



Online specifications of the Ryzen 7 4800H "Renoir" APU list PCIe 3.0 as the adopted standard.

The company also confirmed in the same interview that the APU's SmartShift technology does not support synchronous capability and that users will not be able to run the iGPU and a dedicated Radeon GPU simultaneously. This is something that could have raised the value proposition of all-AMD builds exponentially (a laptop that featured an AMD CPU, AMD iGPU, and AMD dGPU, with the latter two running in sync would be a much powerful build than one with any other combination in the same budget).

Interviewer: Can "AMD SmartShift" support simultaneous operation of both iGPU and dGPU, dynamically distributing power between the iGPU and dGPU dpending on graphics processing load?

Scott Stankard: No, iGPU can dGPU cannot operate simultaneously. When one is working, the other is turned off completely.

That said, AMD's Ryzen 4800H processors are going to be extremely disruptive to the mobility market as these are one of the first processors shipping on a sub 14/10nm process. Since AMD will be able to tap into the cost and power savings of the 7nm process while having access to a much larger supply than Intel's 10th generation mobility processors, it should be able to achieve much higher economies of scale. If they stick to their usual pricing policy we should also see AMD laptops ship for about a $100 cheaper than other configurations of a similar power envelope. The cost of gaming, it would appear, is on its way down with every passing day.

NVIDIA Officially Drops The GeForce RTX 2060 Price To $299 US, Now A Better Purchase Than The AMD RX 5600 XT

NVIDIA has officially

dropped a price bomb on the GeForce RTX 2060, bringing its price down to $299.99 US. The move was done just a few days after EVGAintroduced their GeForce RTX 2060 KO series of graphics cards which start at an impressive price of $279.99 USAMD EPYC 7742

NVIDIA GeForce RTX 2060 Now Available For $299.99 US - Get RTX, DLSS, and More In This Feature-Packed Mainstream Powerhouse

The NVIDIA GeForce RTX 2060 launched last year at CES 2019 at a starting price of $349.99 US. The graphics card has since received the SUPER treatment but despite the other RTX variants which got replaced by their SUPER brethren, the RTX 2060 was still kept in NVIDIA's inventory, making up the sub $350 US market. NVIDIA GeForce RTX 30 Ampere Gaming Graphics Cards, RTX 3080 Ti & RTX 3080, To Be Introduced on 9th September

Now, NVIDIA has officially brought the price of their mainstream offering to $299.99 US, making it the only RTX option to purchase under $300 US. The NVIDIA GeForce RTX 2060 packs the NVIDIA Turing TU106 GPU with 1920 CUDA cores, 120 TMUs, and 48 ROPs. The chip is coupled with 6 GB of GDDR6 memory running at 14.00 Gbps to deliver 336 GB/s of bandwidth along its 192-bit interface. The card has a TDP of 160W and should go along nicely with mainstream gaming PCs under $1000 US.



NVIDIA GeForce RTX/GTX "Turing" Family:
Graphics Card NameNVIDIA GeForce GTX 1650NVIDIA GeForce GTX 1650 D6NVIDIA GeForce GTX 1650NVIDIA GeForce GTX 1660NVIDIA GeForce GTX 1660 SUPERNVIDIA GeForce GTX 1660 TiNVIDIA GeForce RTX 2060NVIDIA GeForce RTX 2070NVIDIA GeForce RTX 2080NVIDIA GeForce RTX 2080 Ti
GPU ArchitectureTuring GPU (TU117)Turing GPU (TU117)Turing GPU (TU116)Turing GPU (TU116)Turing GPU (TU116)Turing GPU (TU116)Turing GPU (TU106)Turing GPU (TU106)Turing GPU (TU104)Turing GPU (TU102)
Process12nm FNN12nm FNN12nm FNN12nm FNN12nm FNN12nm FNN12nm FNN12nm FNN12nm FNN12nm FNN
Die Size200mm2200mm2284mm2284mm2284mm2284mm2445mm2445mm2545mm2754mm2
Transistors4.7 Billion4.7 Billion6.6 Billion6.6 Billion6.6 Billion6.6 Billion10.6 Billion10.6 Billion13.6 Billion18.6 Billion
CUDA Cores896 Cores896 Cores1280 Cores1408 Cores1408 Cores1536 Cores1920 Cores2304 Cores2944 Cores4352 Cores
TMUs/ROPs56/3256/3280/3288/4888/4896/48120/48144/64192/64288/96
GigaRaysN/AN/AN/AN/AN/AN/A5 Giga Rays/s6 Giga Rays/s8 Giga Rays/s10 Giga Rays/s
Cache1.5 MB L2 Cache1.5 MB L2 Cache1.5 MB L2 Cache1.5 MB L2 Cache1.5 MB L2 Cache1.5 MB L2 Cache4 MB L2 Cache4 MB L2 Cache4 MB L2 Cache6 MB L2 Cache
Base Clock1485 MHz1410 MHz1530 MHz1530 MHz1530 MHz1500 MHz1365 MHz1410 MHz1515 MHz1350 MHz
Boost Clock1665 MHz1590 MHz1725 MHz1785 MHz1785 MHz1770 MHz1680 MHz1620 MHz
1710 MHz OC
1710 MHz
1800 MHz OC
1545 MHz
1635 MHz OC
Compute3.0 TFLOPs3.0 TFLOPs4.4 TFLOPs5.0 TFLOPs5.0 TFLOPs5.5 TFLOPs6.5 TFLOPs7.5 TFLOPs10.1 TFLOPs13.4 TFLOPs
MemoryUp To 4 GB GDDR5Up To 4 GB GDDR6Up To 4 GB GDDR6Up To 6 GB GDDR5Up To 6 GB GDDR6Up To 6 GB GDDR6Up To 6 GB GDDR6Up To 8 GB GDDR6Up To 8 GB GDDR6Up To 11 GB GDDR6
Memory Speed8.00 Gbps12.00 Gbps12.00 Gbps8.00 Gbps14.00 Gbps12.00 Gbps14.00 Gbps14.00 Gbps14.00 Gbps14.00 Gbps
Memory Interface128-bit128-bit128-bit192-bit192-bit192-bit192-bit256-bit256-bit352-bit
Memory Bandwidth128 GB/s192 GB/s192 GB/s192 GB/s336 GB/s288 GB/s336 GB/s448 GB/s448 GB/s616 GB/s
Power ConnectorsN/AN/A6 Pin8 Pin8 Pin8 Pin8 Pin8 Pin8+8 Pin8+8 Pin
TDP75W75W100W120W125W120W160W185W (Founders)
175W (Reference)
225W (Founders)
215W (Reference)
260W (Founders)
250W (Reference)
Starting Price$149 US$149 US$159 US$219 US$229 US$279 US$349 US$499 US$699 US$999 US
Price (Founders Edition)$149 US$149 US$159 US$219 US$229 US$279 US$349 US$599 US$799 US$1,199 US
LaunchApril 2019April 2020November 2019March 2019October 2019February 2019January 2019October 2018September 2018September 2018

The move without a doubt has been done to tackle the AMD Radeon RX 5600 XT which got its intro at CES 2020 and will be hitting retail shelves in the coming week for $279.99 US plus premium for the custom models. The EVGA GeForce RTX 2060 KO is already going to hit the sales of the card at $279.99 US, but other AIBs are also well underway to prepare their own cost-effective RTX 2060's while dropping the prices of their existing RTX 2060 graphics cards to match NVIDIA's new MSRP.

At $299 US, the RTX 2060 will offer much better value than the Radeon RX 5600 XT, even while being $20 US more expensive than AMD's Navi based offering. AMD has shown the RX 5600 XT to be up to 15% faster than a GTX 1660 SUPER and 20% faster than a GTX 1660 Ti.

The NVIDIA GeForce RTX 2060, on the other hand, is up to 20% faster than the GeForce GTX 1660 SUPER while offering added features such as RTX, DLSS, etc. It will be interesting to see how the Radeon RX 5600 XT and the price-cut GeForce RTX 2060 perform against each other in the coming weeks.

Which mainstream gaming graphics card is now the better purchase?

AMD Zen 3 ‘4000 Series’ Processors And 600 Series Chipset Landing In Late 2020, Getting USB 4.0

It is a brand new year and its time for some brand new leaks. This one comes from Electronic Times and they claim that AMD's 600-series chipsets will be landing near the end of 2020. The 600-series chipsets will not only support USB 3.2 but 4.0 as well. The new chipset is designed for Zen 3 based processors and should usher in a new era for AMD processors bringing IPC gains that will put even Zen 2 to shame. We have already heard about this particular timeline and considering this is the second leak reiterating that it is starting to look like huh721212ale600 reality. AMD EPYC 7502

AMD 600-series chipsets will be ready by end of 2020, Zen 3 launching at CES 2021?

Of course, if this report is to be believed then we are looking at the Zen 3-based Ryzen 4000 series processors launching at CES 2021 (considering it doesn't make much sense to launch the processors without the underlying 600-series chipset becoming available). Backward compatibility and support is something on which we have very little information at this point so its best to keep an open m


ind.


A520 to start production in the first quarter of 2020. The 600-series counterparts, on the other hand, will be getting some serious upgrades including a USB 4.0 controller. AMD's Zen 3 based processors are expected to be the pinnacle of Zen-based CPU architecture and will be the mature process on the 7nm process from TSMC. Speaking of which, TSMC is expecting AMD to become the largest customer on its 7nm process node this year (it is currently no. 4 on the list).

AMD is currently supply-locked since TSMC's 7nm capacity is fully maxed out with Apple taking on a huge chunk. Sometime around 2H 2020, Apple will be moving on to the 5nm process leaving AMD completely free to take ownership of the 7nm process. Word on the grapevine is that the company has already put in a request for an additional 30,000 wafers (which is a huge amount by the way). Meanwhile, Intel remains supply locked as its foundries work at peak capacity.

AMD's 600-series chipsets will be led by the X670 chipset which will replace the older X570 PCH. Here is where things get murkier, however. According to rumors, the AMD X670 PCH will retain the AM4 socket and feature PCI 4.0 support as well as increased IO support. There are also chances that Thunderbolt 3 will be featured natively on the chipset. Now, here is where things get interesting: there is a good chance AMD chooses to retire the AM4 socket with the 600-series chipsets regardless of what rumors are pointing to right now.

Why? The company is already doing so on the HEDT side of things and it makes sense to streamline the mainstream side as well. Secondly, the introduction of new tech like USB 4.0 might actually make this a business decision worth taking. Thirdly, the AM4 platform would actually have been with us for more than 4 years by the end of 2020 and it would be about time for it to go. If AMD's new platform also supports DDR5 and PCIe 5.0, then it would be a no-brainer to change the socket as well for long-term compatibility.

AMD's SVP, Forrest Norrod, has previously revealed how Zen 3 is going to bring an entirely new chip architecture along with other key enhancements. Based on the 7nm+ process node, AMD aims to deliver some major IPC improvements and key architectural changes with Zen 3 cores. In an ironic turn of events, AMD will also follow the Intel Tick-Tock cadence with Zen 2 being the Tick since it offers a new process node and an evolution to the original Zen design rather than a complete architectural change while the Zen 3 core will be AMD's Tock, offering a similar but enhanced process node (7nm+) alongside a brand new architecture.