Apple Introduces M1 Pro and M1 Max: the Most Powerful Chips Apple Has Ever Built

[R0H1T]

Now this would be interesting

For those who think the M1 Pro and M1 Max in the MacBook Pro are impressive, the new Mac Pro desktop is expected to come in at least two variations: 2X and 4X the number of CPU and GPU cores as the M1 Max. That’s up to 40 CPU cores and 128 GPU cores on the high-end.

https://twitter.com/i/web/status/1450233957350576129

 

[Vya Domus]

You are the one who insisted that knowing the lower bound of the IPC was important!!

Nah, don't spin this around, you first claimed that IPC is a system specific metric, which it isn't, it's a property of the processor. Then you went on about how you're only interested in finding out what the IPC is in applications that people use, at which point I asked what do you do with that information and you never gave a proper answer.

to understand differences in generic application performance and where the differences might come from.

That makes no sense, IPC is simply used to estimate generic performance, you can't infer what the differences might be between applications from that, that's the point.

 

[Dredi]

IPC is simply used to estimate generic performance

How?? There is no ”generic software” to use in measuring IPC.

you first claimed that IPC is a system specific metric

It definitely is. Try measuring IPC on your PC without a motherboard, or main memory. I’m not holding my breath.

 

[Vya Domus]

It definitely is. Try measuring IPC on your PC without a motherboard, or main memory. I’m not holding my breath.

What a terribly unintelligent response. Was that supposed to be a gotcha moment or something ?

A processors executes instructions, not the motherboard, not the memory. IPC is a CPU metric.

 

[Turmania]

Love'em and or Hate'em, you have to respect Apple and what they are doing.

 

[Dredi]

What a terribly unintelligent response. Was that supposed to be a gotcha moment or something ?

A processors executes instructions, not the motherboard, not the memory. IPC is a CPU metric.

But the memory speed specifically has a high impact in how quickly the CPU manages to execute instructions, for some software.

if you measure the IPC of some software with ddr5 and separately with ddr4 memory, which measurement is ”the IPC”, if the processor stays the same?

And please, do explain how you measure the generic IPC you keep touting of.

 

[defaultluser]

I applaud Apple for pushing the boundaries, but their target audience for these machines is tiny - there will be no Linux or Windows support on these things!

The uptick in sales from the first generation ARM devics will die -off pretty quickly, And Apple will become yet-another-Sun.

 

[R0H1T]

And Apple will become yet-another-Sun.

With 1000-10000x in lifetime profits ~ yeah not happening

 

[Vya Domus]

But the memory speed specifically has a high impact in how quickly the CPU manages to execute instructions, for some software.

Then you're just measuring performance for that specific price of software, the CPU is still a constant that always behaves the same given the same input. If you however still decide to divide the number of instructions executed by the number of clock cycles elapsed while running that program and call that figure "the IPC", fine, but it's completely useless because it will vary for literally every single program for a million reasons.

Assume you run a program and you're trying to measure "the IPC", at some point that price of software has to communicate over the network, which obviously takes some time. At the end you draw the line and you say, well this CPU can do X instructions per clock. Do you see nothing stupid about that ? What did you actually measure ? The IPC of that specific CPU or system or whatever ? Really ?

Ultimately, if you want to measure IPC as precisely as possible you have to isolate the CPU from everything else, including memory speed. I don't know why that isn't obvious.

if you measure the IPC of some software with ddr5 and separately with ddr4 memory, which measurement is ”the IPC”, if the processor stays the same?

You tell me, according to you the IPC of a processor is an ever changing value dependent on everything and all values that you get are valid, somehow.

 

[Dredi]

Then you're just measuring performance for that specific price of software, the CPU is still a constant that always behaves the same given the same input. If you however still decide to divide the number of instructions executed by the number of clock cycles elapsed while running that program and call that figure "the IPC", fine, but it's completely useless because it will vary for literally every single program for a million reasons.

Assume you run a program and you're trying to measure "the IPC", at some point that price of software has to communicate over the network, which obviously takes some time. At the end you draw the line and you say, well this CPU can do X instructions per clock. Do you see nothing stupid about that ? What did you actually measure ? The IPC of that specific CPU or system or whatever ? Really ?

Ultimately, if you want to measure IPC as precisely as possible you have to isolate the CPU from everything else, including memory speed. I don't know why that isn't obvious.



You tell me, according to you the IPC of a processor is an ever changing value dependent on everything and all values that you get are valid, somehow.

Just describe to me how IPC should be measured, as you would do it.

My method is both correct, and produces results that can be used for something.

IPC that isn’t affected by memory speed requires it to be measured running software that does not utilize the memory controller. How relevant are such performance metrics?


And btw, I have for the entirety of this discussion said that IPC is application specific. Your post makes no sense.

Next you’ll probably state that IPC is only for single threaded software.

 

[defaultluser]

With 1000-10000x in lifetime profits ~ yeah not happening

Profits didn't matter for IBM, either - they still sailed their ship straight onto insignificance

I'm just saying that Sun is the closest thing to Apple's corporate direction (a lot more profitable, but every corp find's it's own unique ways to become irrelevant)

 

[TheoneandonlyMrK]

But the memory speed specifically has a high impact in how quickly the CPU manages to execute instructions, for some software.

if you measure the IPC of some software with ddr5 and separately with ddr4 memory, which measurement is ”the IPC”, if the processor stays the same?

And please, do explain how you measure the generic IPC you keep touting of.

Your on about application specific performance of a system, that is not IPC.
How a system leverages it's intrinsic ability (IPC)
Is system level performance.

IPC instructions per clock is a hypothetical unbending max of the hardware. . .

It's not a constantly varying parameter depending on the surrounding system or the software running.

Sisoft Sandra.
Cpuz
Etc.

How relevant is your version of IPC when everyone buys different ram SSD motherboard?! Then runs different games apps.

 

[Dredi]

IPC instructions per clock is a hypothetical unbending max of the hardware. . .

No it is not. You are mixing IPC with theoretical throughput (usually given as FLOPS/IOPS). What point is in your idea of IPC? What use is it to anyone? Can you link some text where it is used in that manner?

How relevant is your version of IPC when everyone buys different ram SSD motherboard?! Then runs different games apps.

It is not. IPC is a metric one can use in the quest of understanding arcitectural differences, nothing more. It has little to do with actual real life performance.

 

[Vya Domus]

Just describe to me how IPC should be measured, as you would do it.

I already said how, you try and remove as many variables that don't have to do with the CPU. Otherwise you are not measuring IPC, you are measuring something else, memory performance or I/O or whatever else.

Say you had two CPUs, one a lot faster than the other one but paired with memory that's a lot slower and you'd try and calculate IPC your way by running a memory intensive program and say that for arguments sake you find out both give you extremely similar figures. Then, according to you, it would be totally valid to claim that both CPUs have the same IPC, which would be obviously wrong and extremely idiotic.

 

[Darmok N Jalad]

I was responding to you stating that you don't think anyone buys $3000+ laptops for "office work", and your arguments against Apple knowing their audience. It's pretty clear that they do (in part because they've been pissing off their core creative audience for years, and are now finally delivering something they would want).

What often gets lost when looking at pricing of Apple products is when just doing a strict comparison of performance vs PC internals while not considering the rest of the machine. In the case of the MBP, they also include very very good displays (high-refresh, miniLED, ultra wide gamut, and very color accurate), and really good input components. Once you start shopping PC equivalents with these factors (display especially), not only does the cost difference start to dwindle, but so does the list of options. For example, I tried an XPS17 9700, and the display and keyboard were really good, but the trackpad was total crap. I mean really bad, mushy, ruined the entire experience. To your point specifically, it’s a major expectation of Apple customers, and was why the butterfly keyboard models were widely hated by Apple customers. Add the lack of ports, the dreaded touchbar, and the trash can Mac Pro to the list of big misses. What Apple announced yesterday has got to be making their customers very happy. While they really don’t just come out and admit that they were missing the mark for years, they are at least back to delivering on customer expectations again. That era of portless, overly thin, “this is what you should want” product launches just might be over.

In the end, Apple doesn’t care if everyone buys into their benchmarks, as long as people buy their products. When I look at the MBPs as the sum of their parts, I see them being incredibly successful and way better than what they replaced. Apple has never wanted to own all the marketshare in the desktop space, but rather continue to sell in their existing high-end market.

I applaud Apple for pushing the boundaries, but their target audience for these machines is tiny - there will be no Linux or Windows support on these things!

The uptick in sales from the first generation ARM devics will die -off pretty quickly, And Apple will become yet-another-Sun.

When M1 came out, porting for Linux began. They have already got it to booting to the GUI, though GPU acceleration is not there yet and is a bigger hurdle. Apple said running Windows on M1 is in MS’s court, but MS has shown no interest in licensing their ARM version of Windows, especially now that you can just license a hosted version of Windows. Parallels works on M1 Macs.

 

[Dredi]

Say you had two CPUs, one a lot faster than the other one but paired with memory that's a lot slower and you'd try and calculate IPC your way by running a memory intensive program and say that for arguments sake you find out both give you extremely similar figures. Then, according to you, it would be totally valid thing to claim that both CPUs have the same IPC, which would be obviously wrong and extremely idiotic.

But it would be the same! Specifically for that application!

I already said how, you try and remove as many variables that don't have to do with the CPU.

Can you link to some examples where your ideas are applied, preferrably for PC/mac processors?

 

[dragontamer5788]

IPC includes memory performance.

Heck, you can almost always tell when memory-performance is the limiter. Your IPC figures are 0.1 or so (yes, one instruction every 10 clock ticks). Actually, that's something like L3 cache, main-memory is roughly one-instruction every 200 clock ticks or something, because CPU is way way faster than RAM these days.

In a minority of cases, people can write highly optimized code + have very good data organization (so it takes a bit of luck: not all problems can be optimized this way) to be CPU-limited instead. In these cases, you see IPC reach into 2 IPC, 3 IPC, or 4IPC.

 

[Richards]

It's absolutely important to know the lower bounds because that tells you what the worst case scenarios is.



Huh ? You always aim to use the most out of processor given whatever the time constraints are, I don't know what you are talking about.


That's such a bizarre thing to say. OK, you find out that a CPU can achieve X IPC in a certain application. What can you do with that information ? Absolutely nothing, people measure IPC to generalize what the performance characteristics are. If you are only interested in an application, as you say, then IPC measurements are pointless, you're actually just interested in the performance of that application.



I guess ? It's still stupid and that doesn't tell you anything about how relevant the hardware is if you argue that people who didn't actually needed one would buy it anyway.



I have no idea but I fail to see why it contradicts anything that I said. I just said that more L3 cache doesn't always translate to much improved performance. If your code and data mostly resides in L1 cache then messing around with the L3 cache wont do anything. Obviously real world workloads are a mixture of stuff that benefits more or less from different levels of caches or from none of them at all.


I read many of his articles and while they're very good I can't help but notice he has a particular affinity for everything Apple does.


A wide core with huge caches and most importantly a very conservative clock speed. That's why I am not impressed, trust me that if their chips ran at much higher clocks comparable to Intel's an AMD's chips while retaining the same characteristics then I'd be really impressed. But I know that's not possible, ultimately all they did is a simple clock for area/transistor budget tradeoff because that way efficiency increases more than linearly. I just cannot give them much credit when they outperform Intel and AMD in some metrics while using who knows, maybe several times more transistors per core ?

Glad you noticed that andrei from anantech is a big apple fan... plus anantech founder works for apple

 

[defaultluser]

When M1 came out, porting for Linux began. They have already got it to booting to the GUI, though GPU acceleration is not there yet and is a bigger hurdle. Apple said running Windows on M1 is in MS’s court, but MS has shown no interest in licensing their ARM version of Windows, especially now that you can just license a hosted version of Windows. Parallels works on M1 Macs.

I'm sorry man - forgot that several million people ran Linux GPU-less on their PS3.

Oh wait, no they didn't

If you have no GPU-support, then you have no public interest in the project (why would I spend gargantuan amounts of money for one of these things and shove Linux on it when you can't use the proprietary GPU or tensor system for anything?)

The PS3 worked better than Mac ARM Linux (because they had an official port,) but in the end-of-the-day nobody actually used it for home use! Mac ARM Linux, by-comparison going to remain a hack-job for years (therefoore, DOA!)

 

[windwhirl]

I applaud Apple for pushing the boundaries, but their target audience for these machines is tiny - there will be no Linux or Windows support on these things!

The uptick in sales from the first generation ARM devics will die -off pretty quickly, And Apple will become yet-another-Sun.

Apple's market is for the most part separated from the rest of brands. It's not just the device's brand itself, but also the ecosystem around it.

Apple will likely remain relevant for years if not decades to come.

 

[Vya Domus]

But it would be the same!

You know what, you're totally right.

 

[Darmok N Jalad]

I'm sorry man - forgot that several million people ran Linux GPU-less on their PS3.

Oh wait, no they didn't

If you have no GPU-support, then you have no public interest in the project (why would I spend gargantuan amounts of money for one of these things and shove Linux on it when you can't use the proprietary GPU or tensor system for anything?)

The PS3 worked better than Mac ARM Linux (because they had an official port,) but in the end-of-the-day nobody actually used it for home use! Mac ARM Linux, by-comparison going to remain a hack-job for years (therefoore, DOA!)

You said there was no support. I merely replied that it is being worked on. I wouldn’t expect it to “just work” even after a year. It’s a work in progress. No one is asking you to spend your money on it. I wouldn’t buy a Mac with the intent to run another OS on it as the primary OS.

The parallel with Playstation is a red herring. Very few would elect to use a game console as a desktop machine beyond the novelty of it, as it’s not exactly great hardware for the job in the first place. However, people just might pick a Mac to be a desktop machine to run an alternate OS on, as has been the case for a decade or so.

 

[TheoneandonlyMrK]

No it is not. You are mixing IPC with theoretical throughput (usually given as FLOPS/IOPS). What point is in your idea of IPC? What use is it to anyone? Can you link some text where it is used in that manner?



It is not. IPC is a metric one can use in the quest of understanding arcitectural differences, nothing more. It has little to do with actual real life performance.

Who said it was useful to anyone.

It's in the abbreviation

Instructions per clock.

it's not my fault if you and others are using it wrong ,out of context etc.

Your on about application performance in an application.

I'm on about the actual term IPC.

It's been misused for years.

First note on wiki
"In computer architecture, instructions per cycle (IPC), commonly called instructions per clock is one aspect of a processor's performance: the average number of instructions executed for each clock cycle. It is the multiplicative inverse of cycles per instruction."

Instructions per cycle - Wikipedia

en.m.wikipedia.org

I'm not at work now do go on.

And as a SYSTEM test engineer your reasoning is unsound you use known values and standards to ascertain or Benchmark the performance of a System not any or every application one might use.

You use design/mathematically provable and defined criteria to classify the spec of a Part.

 

[Dredi]

Who said it was useful to anyone.

It's in the abbreviation

Instructions per clock.

it's not my fault if you and others are using it wrong ,out of context etc.

Your on about application performance in an application.

I'm on about the actual term IPC.

It's been misused for years.

First note on wiki
"In computer architecture, instructions per cycle (IPC), commonly called instructions per clock is one aspect of a processor's performance: the average number of instructions executed for each clock cycle. It is the multiplicative inverse of cycles per instruction."

Instructions per cycle - Wikipedia

en.m.wikipedia.org

I'm not at work now do go on.

And as a SYSTEM test engineer your reasoning is unsound you use known values and standards to ascertain or Benchmark the performance of a System not any or every application one might use.

You use design/mathematically provable and defined criteria to classify the spec of a Part.

Oh, you managed to read the wiki page finally. Too bad you didn’t understand the contents.

As you quoted, IPC is the average instructions per cycle of a given processor, (for a given piece of software). It has nothing to do with the maximum instructions per cycle of a given processor, what you previously wrote.

IPC instructions per clock is a hypothetical unbending max of the hardware. . .

 

[TheoneandonlyMrK]

Oh, you managed to read the wiki page finally. Too bad you didn’t understand the contents.

As you quoted, IPC is the average instructions per cycle of a given processor, (for a given piece of software). It has nothing to do with the maximum instructions per cycle of a given processor, what you previously wrote.

I'd re read that? That's average IPC.
The technical use of the term has changed over the years but it is what it is.

I disagree with your and everyone else like you, use of it clearly.

In the time before FP units, it was easier to define IPC, it's a shit show now.