Apple dumps Intel and shows Final Cut Pro on new Mac with Apple CPU !

https://www.macworld.co.uk/news/details-of-apples-silicon-graphics-plans-emerge-3794918/

Apparently the same A14X processor, with its integrated GPU (codenamed Tonga), will be used for the new 12in MacBook and iPad.

According to the report it's being produced using TSMC's 5nm process.

The iMac, when it gains a Silicon processor at some point in 2021, will feature a GPU know by the codename Lifuka. (Lifuka is an island that is part of the Kingdom of Tonga in Polynesia).


https://www.patentlyapple.com/paten...ergize-pc-market-as-possibilities-beckon.html

Integrated discrete GPU chips: Thanks to its high-performance GPU core driven by TSMC 5nm, the M1 is more powerful than the mid-range standalone GPU graphics cards. However, it is still insufficient for content creators. Therefore, Counterpoint believes Apple Silicon will integrate AP and GPU into a single chip by using advanced package"

https://appleinsider.com/articles/2...nderbolt-3-egpu-gives-hope-for-future-support

The lack of support for external GPUs in Apple Silicon M1 Macs may only be temporary, as a connected eGPU is still detected, but doesn't do anything....

However, testing by Mac4Ever corroborated by AppleInsider's own trials, offers hope to eGPU owners. While eGPUs are not officially supported by the latest M1 MacBook Air, MacBook Pro, and Mac mini, macOS 11 still sees the enclosure, and the PCI-E card inside.

This discovery leads to the possibility that Apple may reintroduce eGPU support in the future. It would have to arrive as an update to macOS Big Sur, a new M-series chip launch, or possibly both.
 
The rumors are currently a M1X chip that doubles the performance cores from 4 to 8 and keeps the same 4 efficiency cores for a total of 12 cores. This makes sense since 4 low power cores are probably all that is needed for simple tasks.

What the rumors are not saying right now is what that means for the GPU. On the CPU side the M1 is already extremely impressive and matches or beats most other mobile or desktop CPUs at a fraction of the heat and power. If the 16" really does double the performance cores we are looking at a CPU score that matches the 16 core Mac Pro in a super power efficient laptop.

Its the GPU power I'm a little concerned about. Apple spanked other integrated graphics and some lower end dedicated GPUs like the 560X and the 1050 but still falls short of even the 2060. Even if they doubled the GPU performance from 8 cores to 16 cores it may still not match the performance of the 2060 which we now find on low cost PC laptops. When it comes to FCPX it likely doesn't matter much since FCPX makes extremely good use of the M1 GPU and does better than eGPUs on Intel Macs right now which is insane. On the Resolve, Adobe, 3D rendering and gaming side however I think even doubling the GPU cores will still fall short. Hopefully Apple has another trick up its sleeve for the 16" GPU.

I also heard a rumor once that they are working on a Apple Silicon dedicated GPU for the iMac. Not sure what the heck that means yet but maybe by splitting up the CPU and GPU they can make the GPU use 32 cores which would compete very well with higher end Nvidia cards. They will likely use that in the new Mac Pro as well with the added benefit of adding multiple units of that GPU to the MP.

By the way these things are selling like hotcakes. A custom M1 with 16 GB will now not ship until the end of January. Maybe that time will get better after the holidays but every single Apple store within a 400 mile radius of where I live is sold out of every single M1 Mac. Its getting very hard to get one of these things.

Thanks for the insight. Definitely looking forward to upgrading my MBP. I am also curious about the GPU, and whether they will introduce a killer discreet GPU, rely on existing discreet mobile GPUs (such as AMD), or offer an option for both. Beyond performance in video editing, I am also curious how an Apple designed discreet GPU would fair for gaming, not just for performance but for compatibility. Both native and under Rosetta. Not because I am buying a laptop for such, but because with so many long flights having access to GOOD desktop/console class games via Steam would be great. I've killed a lot of hours on "Into The Breach" on plane rides and having something to turn off with on the road and make a 14 hour flight disappear is nice. But it's a simple game, and I'm hoping whatever they conjour up will put this potential to use and allow for flexibility.

Either way, I'm sure I'll be extremely satisfied with my upgrade from my late 2013 model!
 
I haven't even received mini yet and I'm already thinking of limitations that bother me. Mainly how to maintain a backup archive. My current Mac Pro I have 6 internal Sata ports.
sata 1: 1tb ssd system
sata 2: 10tb hard drive archive
sata 3: 10tb hard drive archive back up

For the mini I'll need either a external mirror raid or a dual drive dock. I've read drive docks have dreadful drive to drive performance like 5gbs.
The mini only has 2 thunderbolt ports. With a monitor I'm down to 1 port. If I use an external ssd enclosure now I have no more thunderbolt ports.

Your internal SATA drives sound like regular HDD which means they likely top out at 180 MB/S. The USB3 ports on the Mini have bandwidth for close to 600 MB/S. You can pick up a $10 USB3 hub and hook up at least 3 HDD drives for the same bandwidth you currently use them at inside your MP. Thats just using the dedicated bus speed of the USB3 controller without touching the TB controller bandwidth. While USB3 can do 600 MB/S no HDD can get anywhere near that speed except for a SSD. The HDD are the actual bottleneck for speed in your case.

If you need more USBC ports then you do have a nice new option. https://eshop.macsales.com/shop/owc-thunderbolt-dock This adds more TB ports turning one port into three ports. When it comes to TB what matters really is how much bandwidth TB3 has which is 40gbps. USB3 is 5gbps which is about the speed of a normal SATA based SSD drive or 550 MB/S. You could hook up 8 SSD drives before you eat up all the TB3 bandwidth. Thats why as long as you can daisy chain or add a hub like the one I linked to you can add plenty of drives to the Mini.

I have a 2018 i7 6 core Mac mini that does have 4 TB3 ports but it really only has two that I can use since I want to keep the full 40 gbps bandwidth for my eGPU. Every two TB3 ports uses one TB controller. So pair one has 40 gbps and pair two has another 40 gbps. Thats why I keep pair one to just drive my Vega 56 eGPU and pair two to drive numerous USB3 devices, hubs and drives. Since the M1 Mini does not support eGPU you are left with no fewer ports than I use on my Intel Mini. Keep in mind the iMac also only has 2 TB3 ports. It does have 4 USB3 ports but those 4 ports still share the same 5 gbps USB3 controller. Adding a small USB3 hub to the two USB3 ports is no different really than the iMac having four USB3 ports.

The short version of all this rambling is don't worry about it. You have a ton of bandwidth to work with. 40 gbps has enough bandwidth for 27 HDD drives which is a heck of a lot more than the SATA ports on your MP. Thats not even including the USB3 ports which gives you another 3 drives for a total of 30 drives. TB3 is a massive amount of bandwidth.
 
If true, why does this stuff always happen out of nowhere by some ridiculous amounts.

Like 50 million, 100 million times faster; OKAY.

But a cool 100 trillion times faster?
 
If true, why does this stuff always happen out of nowhere by some ridiculous amounts.

Like 50 million, 100 million times faster; OKAY.

But a cool 100 trillion times faster?

You haven't been following quantum computing? Its only useful for certain types of things (I think things with probabilistic quantum variables, because that's how it thinks). IIRC, there are 3 states that each unit can be stored in, not 2. And it doesn't have to decide which state it's in until the computation is completed.

They're exponentially faster than classical computers at relevant problems, hence the tremendous factors of magnitude advantage. Theres basically no comparison- like 10 seconds for a quantum computer but 10000 years for a supercomputer

One area they are expected to be very useful for is breaking encryption algorithms

Translation: start saying ni how to the Chinese century (millennium?)
 
What's the news here?

For quite some time, IBM has cool web environment for developers where you can test your algorithms on quantum computer.

It's true that application is rather limited, but don't forget that "modern" computers only can do addition and logic operations.

When your only tool is "hammer", you just have to be creative in solving problems.
 
What does that have anything to do with the/their breakthrough?

Everyone's racing to build a quantum computer that has enough processing capacity to actually be useful. They're functional, but only taking baby steps (dozens of processing units where they will want thousands or something)

Supposedly the chinese quantum computer is 10 billion times faster than Google's quantum computer, which suggests to me that they got it working with more processing units. That is the breakthrough being chased - scaling up the design
 
Everyone's racing to build a quantum computer that has enough processing capacity to actually be useful. They're functional, but only taking baby steps (dozens of processing units where they will want thousands or something)

Supposedly the chinese quantum computer is 10 billion times faster than Google's quantum computer, which suggests to me that they got it working with more processing units. That is the breakthrough being chased - scaling up the design

Also Quantum computers are usually designed for specific purposes. The main use is chasing Post Quantum Cryptography.
 
We might see 32-core Apple Silicon Macs and/with 16-core and 32-core GPUs late next year or early 2022:

https://www.theverge.com/2020/12/7/22158264/apple-32-core-processor-arm-mac-macbook-pro

https://www.bloomberg.com/amp/news/...ac-chips-with-aim-to-outclass-highest-end-pcs

Additionally:

"For later in 2021 or potentially 2022, Apple is working on pricier graphics upgrades with 64 and 128 dedicated cores aimed at its highest-end machines, the people said. Those graphics chips would be several times faster than the current graphics modules Apple uses from Nvidia and AMD in its Intel-powered hardware."
 
Last edited:
Since Apple is alleged to be working on a 128 core chip - ARM had those for about a year and a half now - it is exploring the server/cloud market. Marvell and Ampere have released their 96-cores and will/should soon release their 128-core versions too. Apple, however, will be the only one covering the entire CPU market, from the mobile phones to the multi-million dollar servers.
 
Since Apple is alleged to be working on a 128 core chip - ARM had those for about a year and a half now - it is exploring the server/cloud market. Marvell and Ampere have released their 96-cores and will/should soon release their 128-core versions too. Apple, however, will be the only one covering the entire CPU market, from the mobile phones to the multi-million dollar servers.

Apple? Servers? Probably only its own. It would be a long uphill battle to get into a small market with very different dynamics than iPads and iPhones, with opponents that have been serving that market for decades, and it's a market that is notoriously finicky about reliability and support. There's no enterprise that would give up having access to 24-hour on-site service in case of problems of any kind for any reason no matter how it's marketed. Apple wouldn't be able to compete with anyone on price, because reliability and support are far more important, and enterprise support has never been one of Apple's strengths. A dufous bar just doesn't cut it.

My guess is that on the off chance that Apple is working on 128 core datacenter oriented processors, it's to build an IoT infrastructure for its iDevices and replace its workstation market with online services using its svelte devices as thin clients. Otherwise, it would be a huge investment in a trivial sized market that would take years to break into and where none of its strengths help.

But Apple doesn't need to convince anyone of anything in order to get Apple processors into Apple's datacenters, and it doesn't need to build a large scale global network of people who can show up at your doorstep within four hours of a call at any time of day any day of the week including holidays to repair or replace a server because a mouse chewed through a cable or a transformer blew out a server or a spinning disk failed or whatever else went wrong.
 
https://eshop.macsales.com/blog/68484-thunderbolt-on-the-m1-mac-mini/

Well apparently Apple slipped in another surprise for us with M1. Unlike Intel machines with TB3 where every two ports use one TB controller the M1 has one controller per port. So the M1 Mini with two TB ports has as much external bandwidth as the Intel Mac Mini has with four TB ports. You do get 40gbps per port on the M1. I believe the same is true for the 13” MBP as well. Have not heard if it is true for the MBA.

While they do have less physical ports with TB4 we finally have the ability to add hubs that add extra TB ports. Only having two ports suddenly doesn’t seem so bad anymore.
 
Apple? Servers? Probably only its own....

On November 30, Amazon began to offer Mac Minis as part of its cloud service (announcement). And I can see them offering servers based on the M1, if technically possible, especially since they already offer ARM servers (Graviton).

You're right that a company with its own data center would unlikely restock it with Macs. But most of the new tech companies don't build their own data centers. They rent them from one of the cloud companies: Amazon, Google, or Microsoft. And even the old companies are starting to move their stuff to those cloud providers.
 
Amazon is buying M1 minis to run OSX specific software. Very few datacenters fit that bill. There's also a very small chance that Apple would be willing to sell its processors to anyone to build their own servers around, both because Apple's SOC is optimized for its own use, and because Apple probably isn't interested in customizing processors to make anyone else happy, because the margins for its consumer electronics are higher.
 
https://eshop.macsales.com/blog/68484-thunderbolt-on-the-m1-mac-mini/

Well apparently Apple slipped in another surprise for us with M1. Unlike Intel machines with TB3 where every two ports use one TB controller the M1 has one controller per port. So the M1 Mini with two TB ports has as much external bandwidth as the Intel Mac Mini has with four TB ports. You do get 40gbps per port on the M1. I believe the same is true for the 13” MBP as well. Have not heard if it is true for the MBA.

While they do have less physical ports with TB4 we finally have the ability to add hubs that add extra TB ports. Only having two ports suddenly doesn’t seem so bad anymore.

This is excellent
 
Apple? Servers? Probably only its own...
iPhone + iCloud sounds like a winning combination. Apple moves quarter billion phones a year and iCloud already has close to a billion members.

Google does the same thing and they also include a pro market. Apple isn't going to let it sit idly.
 
The M1 is an ideal chip for use in servers, many posts ago I guessed the market would move in this direction. Its a server the code can be written to serve (see what I did there ).
Many moons ago at an SAP research lab PR event, Intel turned up giving there sales pitch on how they could coax their hardware to run a few degrees hotter, the promise was that this small increment could reduce power and cooling costs for large server farms.
Now along comes the M1, very powerful and frugal on power with minimal heat output.
If Apple doesn't reenter the server market perhaps Dell etc and hammering on their door to buy.
 
Back
Top