Intel Restates Support for Apple During Transition to Apple Silicon

[cmaier]

What we should all consider is that your knowledge and experience in designing computer chips is the same as mine in performing nano neurosurgery.

Zero.

Here's a partial bibliography:

High Speed Microprocessor Cache Memory Hierarchies for Yield-Limited Technologies

A 533-MHz BiCMOS superscalar RISC microprocessor

This 533-MHz BiCMOS very large scale integration (VLSI) implementation of the PowerPC architecture contains three pipelines and a large on-chip secondary cache to achieve a peak performance of 1600 MIPS. The 15 mm/spl times/10 mm die contains 2.7 M transistors (2M CMOS and 0.7 M bipolar) and...

ieeexplore.ieee.org

Former AMD engineer talks about Bulldozer

X-bit Labs reports an ex-AMD engineer spoke out about some of the reasons why Bulldozer disappoints: The reason why performance of the long-awaited Bulldozer was below expectations is not only because...

www.guru3d.com

A very wide bandwidth digital VCO using quadrature frequency multiplication and division implemented in AlGaAs/GaAs HBT's

A digital voltage-controlled oscillator (VCO) is described which uses frequency multiplication and division to achieve very wide bandwidth. The VCO uses current-mode logic and does not require reactive elements such as inductors, capacitors or varactors. A novel, fully symmetric exclusive-OR...

ieeexplore.ieee.org

Accurate high-speed performance prediction for full differential current-mode logic: the effect of dielectric anisotropy

Integrated-circuit interconnect characterization is growing in importance as devices become faster and smaller. Along with this trend, interconnect geometry is becoming more complex, consisting of an increasing number of wiring levels. Accurate numerical extraction of three-dimensional (3-D)...

ieeexplore.ieee.org

Embedded at-speed testing schemes with low overhead for high speed digital circuits on multi-chip modules

The difficulty of cost-effectively identifying Known Good Die (KGD) is increased in circuits requiring multiple die packaged in Multi-Chip Modules (MCMs). Such circuits typically have high frequency I/O signals which are difficult to measure using inexpensive test equipment. The cost of full...

ieeexplore.ieee.org

A 2-GHz clocked AlGaAs/GaAs HBT byte-slice datapath chip

A byte-slice datapath for exploring multi-chip RISC processor development in AlGaAs-GaAs heterojunction bipolar transistor (HBT) technology has been designed, fabricated and tested. The circuits are implemented using differential current-mode logic (CML) and emitter-coupled logic (ECL) with...

ieeexplore.ieee.org

Care to share your credentials?

 

[GrumpyCoder]

I do not like the Apple Shriners at all. Chairman Mao weaponized that type for his Cultural Revolution.

So you have a personal problem then, independent of the technology. Good to know, because from a technological point of view, any of my computer science students (even the early semesters), can spot there's nothing there from you.

 

[Ballis]

What else are they gonna say?

Lot of furious people who just bought the new MBA, I’d imagine.

Good luck selling any Mac products right now.

Yea totally. Im in desperate need for a new mac, but there is just no way Im buying one prior to ARM now. They cant come soon enough.

 

[SystemHasFailed]

Is the MP worthwhile? Well, it's a machine for professionals making money with it. If you're making money with it, then you should have that $6k (or whatever config you get) back in no time. It's a bad idea to invest in a computer and break even in 5 to 10 years. Ask yourself if you really need the power or just want it. For an investment running it at home, for fun, it's a bad idea.

Here's an example, I recently blew over $200k on three Dell machines for research. That investment was made for a

A fair question.. for me it's a question of longevity of the computer I buy. The Mac Pro I use now I've gotten 8 years out of... If I needed to replace something, Apple shipped me the part and I shipped the old part back. Meanwhile the MBP that my work bought for me is already falling after less than ONE year.. my son's MBP that I got him for college has gone back to apple for repair 2 times so far and now has to go back for a third because keys are falling off the keyboard. The MBP I spec'ed out to buy for personal use was at 4K.. so for 2K more I can buy a arguably far more powerful and expandable machine. I cannot afford to be without a machine for multiple days if/when a problem occurs.

Since I don't use my machines "just" for home use.. I'm probably already answering my own question about it.. but I don't want to be in the position PowerPC users were in 5 years after Apple went to Intel, either.

 

[PCMA]

You're the target audience. Apple is again marketing to the "Shriners" those weirdos who own and collect ancient Apple products and have a little temple to Steve Jobs in their residence.

The Shriners will eat up the bogus tricked out presentations Apple releases on ARM performance.

I am a Shriner — loud and proud... and I am completely pissed off with Tim Cook and Apple.

Apple has stopped being "for the rest of us," and become an elitist luxury brand.

More relevantly to this discussion... they take years to roll out new products. e.g. how long did it take them to bring back scissor switches? Where is FaceID on my Macbook Pro?... and I could go on... and on... and on.

P.S. 73% of all Pro XDR Displays ever made/sold were used as props behind Craig Federighi and his colleagues.

 

[cmaier]

I am a Shriner — loud and proud... and I am completely pissed off with Tim Cook and Apple.

Apple has stopped being "for the rest of us," and become an elitist luxury brand.

P.S. 73% of all Pro XDR Displays ever made/sold were used as props behind Craig Federighi and his colleagues.

The original mac sold for $2500. The original IBM PC sold for $1500.

When was Apple anything other than a luxury brand?

 

[Flight Plan]

I'm looking forward to advancement in technology, for sure.

But having said that, I'm not fond of subscriptions. So Creative Cloud and MS Office aren't selling points for me. Non-subscription versions of creativity and office products WOULD be selling points for me.

 

[PCMA]

The original mac sold for $2500. The original IBM PC sold for $1500.

When was Apple anything other than a luxury brand?

 

[incoherent_1]

Can continue? They haven’t started yet. The main reason no benchmarks are allowed is to avoid showing how unscalable the A12z embedded SoC is in the full OS X and to halt the ensuing paranoia Apple knows will happen when disclosed.

The hardware teams have years to go before this vision is made reality, and neither AMD nor even Intel designs of today will be indicative of performance in designs two years from now.

Hell in 3 months, enthusiasts will be salivating at Zen 3, Big Navi and Nvidia’s latest GPUs and how far all three have upped the game in computing.

Apple has been designing SoCs for years for its iOS line.....

What incentive would they have to convert to ARM if they can’t outperform x86? Makes no sense.

 

[Jon Rowlison]

"We support Apple's right to run laptops at a MUCH slower compute rate, using custom chips for which they can control what features they do not want to include... instead of getting all of the features with only 2/3 the battery life of the A14." -Intel

 

[cmaier]

View attachment 926656

Okay. $1299. I don’t remember what PCs cost then, but probably these weren’t “elitist.”

So, for a few minutes, Apple was not elitist.

 

[Jon Rowlison]

Apple has been designing SoCs for years for its iOS line.....

What incentive would they have to convert to ARM if they can’t outperform x86? Makes no sense.

It's just a question of usage. No doubt the ARM chip will be MUCH slower than a power-hungry x86/x64. The advantage is that Apple can control how they optimize their chip to overcome its limitations and minimize how much slowdown (or speed-up if they optimize correctly) they see for how their OS works behind-the-scenes.

 

[teknikal90]

Intel in mac will be supported atleast for another 5 years! They have 60k$ Mac Pro with intel Xeon out and still selling for high powered video editing and all that everyone relax! We will ok for a long time!

I remember (yes I'm that old) during the powerpc/intel transition...the powermac crowd said the same thing...then the macbook pro got announced 6 months later.
Steve jobs also quoted 2 years transition back then. they did it in less than a year if I recall correctly

 

[Macbookprodude]

Don't be surprised if this relationship deteriorated rapidly, especially when silicon units start shipping from Apple.

Its just great that Apple will be going back to PowerPC through ARM V2. Silicon is a lame name, just call it - PPC V2 ARM on steriods.

 

[v3rlon]

No they don't. ARM CPUs will never be as powerful as x86. That's just a plain fact and always will be. Apple is literally dropping support for all the people who used their hardware for development. Let's see that Ax CPU compete with a high end 16 core/32 thread x86 CPU when it comes to 3D modeling or 8k video editing. These new Macs will be good for browsing and light duty tasks.

Electric lights will never be as bright as kerosene lamps. Who wants to be attached by a wire to a generator anyway?

Your cute little horseless carriage will never be as fast as a real horse. And where are you going to get the gasoline? Horses are self reproducing. You going to buy a new 'automobile' every 10 years? That's just crazy.

ARM is already faster than some x86, and that is running in a passively cooled, thermally limited environment. If you can throw $$$ and man hours at the problem, you can make ARM as big and bad as you like. More cores working together, and work on the speed as you throw an actual heat sink on the processor. If only Apple had spare cash lying around to do this. Oh. They have so much spare cash they are looking for things that need billions of dollars to spend on. Well, maybe if people actually wanted to work for them. Oh. One of the top companies people apply to. You know, this might actually work.

 

[Brien]

Lets be honest, the A12x is not what is going into their desktops; they have something else in the pipeline that will blow the pants off even that for laptops and desktops. I agree that active cooling will help greatly. We know what the A12x can do without cooling; but they have something monumental when it comes to your next MacBook Pro. Something so good you will want to upgrade; it won't be marginal. This is a big step to move code away for Intel native and the ecosystem that already exists; it needs to be worth it besides saving $$ on Intel silicon.

Or they'll do a stupid Apple thing, stay with passive cooling on the notebook lines and make them THINNER!!!!11!1

 

[v3rlon]

No they don't. ARM CPUs will never be as powerful as x86. That's just a plain fact and always will be. Apple is literally dropping support for all the people who used their hardware for development. Let's see that Ax CPU compete with a high end 16 core/32 thread x86 CPU when it comes to 3D modeling or 8k video editing. These new Macs will be good for browsing and light duty tasks.

Did you miss the part where they ran Maya during the Keynote? Or the one with 3 simultaneous 4k videos adding filters in realtime? And that is still the beta.

 

[SocialKonstruct]

Sorry Intel but your desktop CPU’s aren’t that great anymore. I switched to AMD to build every single desktop for the past three rounds.

 

[bergert]

I was a beta-tester for SoftPC, running on a 1993 Centris 650 (Motorola 68k CPU). There were many people which said it would not work, as PC's used 486 and were more powerful than my 68K. It took quite a bit to start up, but once started, the DOS applications were useable- not sluggish. They PC-emulation at BIOS/hardware level was pure genius.

When moving to PPC, we had the Emulator - and Mac applications ran emulated faster than on the Centris. All apps, except drivers, and system extensions just worked. Very smooth transition. Had a G4 (1999) desktop - great computer. After years, still no faster CPU from IBM; MDD (2003) solved it by using two overclocked G4's (dual single core).

Then came INTEL with the Rosetta (1) emulation of PPC. 2008 my MacPro (dual Xenon quad core) would run apps faster than my MDD.

Long story short: Apple did an excellent job back then, and will do with Rosetta (2). By 2022 we'll wonder how we ever kept waiting-and waiting-and waiting: the i9 low-power consumption comes 10 years too late. The i5 to drive 4x16 lanes (USB-C) came 5 years too late. We pay to INTEL (not Apple!) $7500 for the 28-core MacPRO Xenon privilege - for about $1000 we can get 10 x A13, each outperforming an INTEL i9 (single-core), and all 10 besting the Xenon. INTEL is so dead.

Who cares - only Windows users (and the trolls in this forum): the mess called Win/RT is Microsoft's own.

 

[bergert]

Since I don't use my machines "just" for home use.. I'm probably already answering my own question about it.. but I don't want to be in the position PowerPC users were in 5 years after Apple went to Intel, either.

Your argument does not make sense:

I am a home/hobby user, I use my gear around 6 years typically; starting 1989 with the MacIIx. 2008: MacPro dual-quad, 2014: MacPro 6-core (trashcan) is my current. I was planning to get the 12-core MacPro - but now I'll wait for the new Axx-MacPro (2022). My computer is now 6.5 years old. No problem, still great with FCPx, the week of the lockdown I got a TB3-RAID with SAS-SSD. Best gadgets to create video.

And if the Axx does not get as hot, I would love to get the iMacPRO .... I do render jobs for many hours every week. The iMac has the CPU muscle, but it cannot dissipate the heat the INTEL-Xeon creates for multi-hour render jobs.

 

[seek3r]

No they don't. ARM CPUs will never be as powerful as x86. That's just a plain fact and always will be. Apple is literally dropping support for all the people who used their hardware for development. Let's see that Ax CPU compete with a high end 16 core/32 thread x86 CPU when it comes to 3D modeling or 8k video editing. These new Macs will be good for browsing and light duty tasks.

Why not? x86 doesn't have exclusivity on being a fast architecture....

 

[Fabio_gsilva]

Why Apple is not redesigning the iPhone and the iPad to make the battery swap possible at user level ???

2020, dude. Get over it.

 

[seek3r]

Unless buying that $6000 Mac Pro is mission critical, I’d really avoid that unless it’s going to be a tax write off.

If it's a business machine it's going to be a tax writeoff, and be depreciated in 3-4 years anyway - and most Mac Pro customers aren't buying them for light home use.

 

[SystemHasFailed]

If it's a business machine it's going to be a tax writeoff, and be depreciated in 3-4 years anyway - and most Mac Pro customers aren't buying them for light home use.

Exactly. If I'm just browsing the web and doing email I'll use an iPad..

 

[GrumpyCoder]

Since I don't use my machines "just" for home use.. I'm probably already answering my own question about it.. but I don't want to be in the position PowerPC users were in 5 years after Apple went to Intel, either.

You'll not get the long time support you're looking for with the Intel MP. Unless they have two different chip-classes in the pipeline, a ARM version to replace the I3/5/7/9 class of processors and one to replace the Xeon class CPUs. That's in the case they need longer to release a ARM CPU which competes with the latter. Only Apple can tell you.

Here's my take on it. I like others have been saying for ages that the switch is coming, while the nay-sayers claimed it's never going to happen. It was clear to me it had to be this year or next year latest for it to happen. I've bought the MBP 16" fully spec'ed (except for SSD which I picked 2TB) when it came out, knowing it would most likely be my last Intel-based Mac. I've used iMacs, Mac Pros in addition at home. In the past, also at the University, but since I need RTX 8000 and V100 GPUs for my research and teaching, I switched to Dell (Desktops/Servers) there (still using my MBP there too).

I bought the MBP for the next 3/4 years, for my mobile computing needs, expecting we'd have ARM based MBPs after that time.
I've also used some Hackintoshes at home in addition to my "real" Macs. For me, they've been rock-stable. Might be easy to say as someone who's in computer science, but it worked. I wouldn't miss a iMac or MP.

Knowing this, I didn't buy the latest MP when it came out as I always saw it as bad investment. I actually expected the announcement to be ARM in the first place. I also don't need the horsepower at home anymore, the really heavy lifting is done in the cluster. So unless you can make the $6k back in a year, stay away. If you have something that works and does the job, stay away. Unless a MacMini or iMac doesn't work and you need the horsepower to make money and stay in business, stay away. If there's a chance I could get away with something less power, say a I7/9 Extreme / whatever, I'd put my money in a proper Hackintosh for now. They can be rock stable. If I couldn't afford a single day downtime, I'd run macOS via Proxmox. Create a snapshot before anything I install and roll back if it doesn't work. That would get me over the 2 or 2 1/2 years until a MP worthy ARM replacement comes along.

In fact, I'll soon place an order for a i9-10900K to be put in a Hackintosh build, which is powerful enough for my needs at home and fire up the one or two games I play every year.

Sorry Intel but your desktop CPU’s aren’t that great anymore. I switched to AMD to build every single desktop for the past three rounds.

This has nothing to do with Intel and AMD. AMD isn't better in every aspect, Intel still has some advantages depending on your usecase. If all you do is play some games and check your mails, then AMD is better value. On the other hand, how's that 1TB of RAM working out on your Ryzen? Oops, doesn't work at all.

Apple could have easily gone AMD, but for what? 20 Years ago Athlon was the holy grail and Intel was doomed, soon to vanish. How did that turn out? What we see with AMD right now, is exactly what we saw 20 years ago. What's happening now is a switch to a completely new architecture, not dragging around decades of old stuff for backward compatibility. x86 is old, it's flawed, SMT never worked out the way expected, new issues / security threats showed up. It's time to finally leave the old stuff behind. Goodbye x86.

This won't happen over night, it won't happen next year, but ultimately there's no way around something new. People will complain, they also did when CD drives replaced floppy drives claiming floppys would be around forever. They complained when USB mice/keyboards replaced PS/2, 68k to IBM and IBM to Intel and so on. Just live with it.