M1 as an "Industry Game Changer"

[leman]

Did you notice that it is a mobile phone being benchmarked?

The A14 is also getting lower scores than the M1, despite having the same cores.

I think that's a good showing from Qualcom and I expect more performance to come.

Of course it’s a mobile phone. But it gives us numbers to talk about. Performance scales linearly with CPU frequency, so we can pretty much make an informed guess how an X1 will perform in a laptop.

A GB score of 1100 at 2.8 GHz translates to a score of around 1400 at 3.5 GHz (about maximal clock we can expect from these chips). We know that phone manufacturers restrict the peak TDP of their chips to around 5 watts, so we can assume that X1 at 2.8 GHz will need around 5 watts. This in turn means that at 3.5ghz it will probably need around 15 watts. This is definitely not bad, but it’s also not very good. It’s similar perf per watt as latest x86 CPUs, with lower performance. And the Qualcomm chip is already 5nm, so it is using a better process than both AMD and Intel.

ARM CPUs need to offer a big improvement over x86 in order to justify the expensive and risky platform migration. X1 doesn’t seem to offer that yet - AMD Cezanne at 7nm will have similar efficiency and performance. Where ARM machines could have an advantage is battery life (because of BIG.little) ... but I don’t know if that alone is a strong enough incentive for big players to invest big money.

 

[Pressure]

For mobile computing, you are correct.

If the question is: Which brand posts the highest single thread performance?

The answer is that Zen3 and the M1 are tied. If one were to unlock the M1, it would win. The M1 is only available in a locked configuration though.

It depends entirely on workload.

Did you notice that it is a mobile phone being benchmarked?

The A14 is also getting lower scores than the M1, despite having the same cores.

I think that's a good showing from Qualcom and I expect more performance to come.

Just a clarification but M1 and A14 share the same base IP but there are differences like increased L2 cache and double the memory bandwidth.

 

[leman]

It depends entirely on workload.

I haven't so far seen a microbenchmark where either 5950X or M1 would have a decisive lead. I think there are two interesting questions here:

- what would be the relative performance of Zen 3 and M1 if both ran at M1 power consumption levels (5 watts)?
- what would be the relative performance of Zen 3 and M1 if both ran at Zen 3 power consumption levels (20 watts)?

While I don't know the exact answer to these questions, I would bet money on the direction of the trend

 

[magbarn]

Not really. They have pretty much lost the entire pc gaming base to AMD. If I remember, many servers are also swapping over to AMD. Not sure if thats true but could have sworn I read it somewhere.

They are gradually losing more and more markets. Intel needs to change asap

If it weren't for AMD's supply constraints because of TSMC making everything for them (try finding a Zen 3 in stock, especially the 59XX variants), they would be blowing Intel completely out of the water. Even the 11th gen desktop parts next year will still be on 14nm.

 

[dingclancy23]

If it weren't for AMD's supply constraints because of TSMC making everything for them (try finding a Zen 3 in stock, especially the 59XX variants), they would be blowing Intel completely out of the water. Even the 11th gen desktop parts next year will still be on 14nm.

But that is the thing with these machines when you talk about "industry-changing". Are you able to impact things at volume?

Some people think that this change is limited to the Mac and so it is not industry changing.

For myself, while Zen 3 and Cortex X1 is the state of the art outside of the Apple world, by the time they are mass produced and be in the hands of the masses, the M1X or M2 will already have been launched at scale.

This is similar to iOS vs Android. You all get the coolest new stuff in the latest Android, but before it gets to the users, adoption takes 2 years., while Apple products, be it M1's or iPhones can affect things at scale the moment they launch.

So when people say Zen 3 will beat the next M chip, even without considering the vertical integration, the software-hardware stack, and even acceleration, the time to market for Apple devices is what will put a major mark in this next-gen chips.

Not to mention that the Zen 3 Ryzen chips will cost $700!, while you are now able to buy an M1 at $699 which is not just the SOC, but the whole computer in the Mac Mini. And for $200 more you get a 15-hour battery and a P3 screen, aka an actual laptop!

So no, Zen 3 is only a competitor to M1 in benchmarks and a lousy competitor at that given the power envelope.

 

[russell_314]

But that is the thing with these machines when you talk about "industry-changing". Are you able to impact things at volume?

Some people think that this change is limited to the Mac and so it is not industry changing.

For myself, while Zen 3 and Cortex X1 is the state of the art outside of the Apple world, by the time they are mass produced and be in the hands of the masses, the M1X or M2 will already have been launched at scale.

This is similar to iOS vs Android. You all get the coolest new stuff in the latest Android, but before it gets to the users, adoption takes 2 years., while Apple products, be it M1's or iPhones can affect things at scale the moment they launch.

So when people say Zen 3 will beat the next M chip, even without considering the vertical integration, the software-hardware stack, and even acceleration, the time to market for Apple devices is what will put a major mark in this next-gen chips.

Not to mention that the Zen 3 Ryzen chips will cost $700!, while you are now able to buy an M1 at $699 which is not just the SOC, but the whole computer.

So no, Zen 3 is only a competitor to M1 in benchmarks and a lousy competitor at that given the power envelope.

Everyone says well ____ CPU can beat it and sure. M1 isn't the fastest CPU. It's a base model low power CPU so people don't realize how silly comparing a high end desktop CPU to it is. Wait for M1x or whatever Apple calls it then let's compare.

 

[4sallypat]

Everyone says well ____ CPU can beat it and sure. M1 isn't the fastest CPU. It's a base model low power CPU so people don't realize how silly comparing a high end desktop CPU to it is. Wait for M1x or whatever Apple calls it then let's compare.

Agree - people saying their CPU beats the M1 is silly - I just care about my devices running with more efficiency, less delay, less heat, and more life out of a charge.

When I took apart my old trusty 2012 i7 Mini Server to put on the sales block, inside is so packed with components that it would be hard to keep cool compared to the M1 Mini interior.

I do like the fact that I can now run the M1 Mini without a cooling fan pad underneath !

For an entry level processor, it runs circles around my Intel Mac with a lot less heat - I can't wait to see what newer processors come out of Apple.....

 

[Falhófnir]

Aware of that. Again, these aren't magical apple pixie dust. They're just dedicated ASICs doing stuff that anyone else could also add if they're on TSMC 5nm manufacturing process (and thus have a state of the art manufacturing node) and have the silicon die space.

The big problem will be on the software side, but Nvidia already has a pretty good foothold there with regards to ML/AI.

What I'm saying is that Nvidia or someone else could implement on die resources to do the same sort of things. It isn't magic that only apple know how to do. It just that Apple are the only company to try it and actually put a product out to market. Yet.

What apple is doing in M1 is cutting edge for sure - but it isn't stuff that nobody else understands how to do. It simply that Apple had the financial resources to spend on it without it being a "safe bet". Now Apple HAVE done it and proven what is possible (and more importantly, financially viable, and producible in volume), others will follow.

Agreed, a lot of this ‘others can’t possibly match Apple’ talk seems to now be filling in for the ‘Apple can’t possibly beat Intel’ spiel now it’s no longer valid. Apple have advantages through expertise and good planning, as well as their business model - it doesn’t mean others can never catch up, they just have more work to do. Apple is good, Apple is big, but they aren’t bigger and better than the rest of the PC industry combined if they all decide to get behind this and start pushing.

 

[KPOM]

I am curious as to why people think Intel need to up their game. Intel only needed to up their game if they wanted to still be the CPU of choice for Apple.

The only people that are going to buy Apple macs are existing Apple mac owners, employees who use mac's at work and therefore want a mac for themselves and students who use mac's at colleges and universities and therefore want a mac themselves.

The only 'game changer' in my opinon is if Apple are able to produce an ARM machine that has a dedicated GPU for gaming and high end video editing whilst able to maintain low power usage. Until that day arrives Intel has nothing to worry about in my opinion.

You don’t think some executive at Dell (which has spent the last 4 years upping their game with beautifully designed XPS models that attract premium prices) has an MacBook Air on his or her desk and thinks that they need something to compete with it? If the Snapdragon 888 is any indication, QualComm could at least match the performance of Tiger Lake with lower TDPs. Add Windows ARM support for x86-64 apps (coming next year), and I think we’ll see more OEMs at least experiment with Windows ARM notebooks.

 

[leman]

If the Snapdragon 888 is any indication, QualComm could at least match the performance of Tiger Lake with lower TDPs.

You think so? I'm rather skeptical. They would need to clock that chip almost 40% higher, at which point the TDP advantage is likely out of the window.

 

[cgsnipinva]

Aware of that. Again, these aren't magical apple pixie dust. They're just dedicated ASICs doing stuff that anyone else could also add if they're on TSMC 5nm manufacturing process (and thus have a state of the art manufacturing node) and have the silicon die space.

The big problem will be on the software side, but Nvidia already has a pretty good foothold there with regards to ML/AI.

What I'm saying is that Nvidia or someone else could implement on die resources to do the same sort of things. It isn't magic that only apple know how to do. It just that Apple are the only company to try it and actually put a product out to market. Yet.

What apple is doing in M1 is cutting edge for sure - but it isn't stuff that nobody else understands how to do. It simply that Apple had the financial resources to spend on it without it being a "safe bet". Now Apple HAVE done it and proven what is possible (and more importantly, financially viable, and producible in volume), others will follow.

But I think that is the point for saying this is an industry game changer. That does not mean Apple will control it all - no one company can. What I see as the industry changing aspect of this - is Apple's approach leveraging ARM instruction set standard, their own SOC design and with their very focused approach towards efficiency points the way to future performance gains that are now starting to become more difficult with the current x86/CISC industry standard.

I think Qualcomm and NVIDIA are absolutely looking at this to determine what could be gained through their SOC designs/approach.

I see the primary impact is resources being rerouted to this ARM/SOC approach as a means of leveraging market opportunities. That means new entrants. Apple is just forging the path - but they won't be the only company on this path.

As a nerd and IT consumer - it offers something new and interesting to watch and look forward to.

 

[cgsnipinva]

You don’t think some executive at Dell (which has spent the last 4 years upping their game with beautifully designed XPS models that attract premium prices) has an MacBook Air on his or her desk and thinks that they need something to compete with it? If the Snapdragon 888 is any indication, QualComm could at least match the performance of Tiger Lake with lower TDPs. Add Windows ARM support for x86-64 apps (coming next year), and I think we’ll see more OEMs at least experiment with Windows ARM notebooks.

Excellent point. OEMs and other chip manufacturers are looking at what Apple is doing and seeing an opportunity.

Intel and AMD are hitting the x86 ceiling that brute force clock speeds and power draws can no longer address. Apple's approach which delivers comparable performance at much lower power draws. Their relentless focus on efficiency in the development of their mobile SOCs has provided a solid platform to deliver a Desktop/Laptop architecture that has more performance headroom that doesn't require a brute force approach like Intel/AMD has relied upon.

I absolutely believe that Microsoft is looking at how Apple was able to port its entire software stack to the same architecture - something it has wanted to do without outstanding success. Keep in mind - Apple just didn't stick stock ARM cores and stock GPUs and AI cores onto the same die. The designed and engineered each component of their SOC to work effectively together. That is why it has worked - not magic - just a focused and methodical approach leveraging well thought out designs and solid engineering.

Microsoft has the resources to build its own team and leverage industry partners to execute a similar approach. The only stumbling block for MS is they would need to refactor their software to integrate with this approach and new SOC architecture. That is the other part of Apple's approach - integrating the software development with the hardware development to maximize efficiency.

This is why a chip company acting on their own won't yield the same performance results Apple has experienced.

 

[KPOM]

But I think that is the point for saying this is an industry game changer. That does not mean Apple will control it all - no one company can. What I see as the industry changing aspect of this - is Apple's approach leveraging ARM instruction set standard, their own SOC design and with their very focused approach towards efficiency points the way to future performance gains that are now starting to become more difficult with the current x86/CISC industry standard.

I think Qualcomm and NVIDIA are absolutely looking at this to determine what could be gained through their SOC designs/approach.

Exactly. What Apple has done is prove the following:

• ARM is a viable alternative to x86 for desktop-class systems.
• It's possible to build a translation layer for x86-64 to ARM that performs well.
• ARM brings clear advantages on its own, but it is better to customize chips for the OS.

 

[KPOM]

I absolutely believe that Microsoft is looking at how Apple was able to port its entire software stack to the same architecture - something it has wanted to do without outstanding success. Keep in mind - Apple just didn't stick stock ARM cores and stock GPUs and AI cores onto the same die. The designed and engineered each component of their SOC to work effectively together. That is why it has worked - not magic - just a focused and methodical approach leveraging well thought out designs and solid engineering.

Remember Apple had the advantage of having to translate only 64-bit instructions, giving developers a 2 year advanced notice (with Mojave's warning messages and Catalina's lack of 32-bit support entirely). I wouldn't be surprised that they saw Microsoft's struggles with Windows on ARM translating 32-bit instructions and decided early on not to support 32-bit apps (though they likely would have dropped 32-bit support even had they stayed on Intel).

Perhaps Microsoft will use this to double down on 64-bit, telling developers they are all in on ARM but that Win32 compatibility will be adequate at best, and that Win64/UWP is the future.

 

[cgsnipinva]

Remember Apple had the advantage of having to translate only 64-bit instructions, giving developers a 2 year advanced notice (with Mojave's warning messages and Catalina's lack of 32-bit support entirely). I wouldn't be surprised that they saw Microsoft's struggles with Windows on ARM translating 32-bit instructions and decided early on not to support 32-bit apps (though they likely would have dropped 32-bit support even had they stayed on Intel).

Perhaps Microsoft will use this to double down on 64-bit, telling developers they are all in on ARM but that Win32 compatibility will be adequate at best, and that Win64/UWP is the future.

Agreed. That would simplify Windows migration and better enable them to optimize their software stack for a new architecture. If I were them, I would seriously start conversations with Nvidia. They own ARM and they could also offer chip design and engineering services to support them.

Of course this would raise alarm bells in AMD and Intel. They would respond as well. How is a bit more unclear.

 

[KPOM]

Agreed. That would simplify Windows migration and better enable them to optimize their software stack for a new architecture. If I were them, I would seriously start conversations with Nvidia. They own ARM and they could also offer chip design and engineering services to support them.

Of course this would raise alarm bells in AMD and Intel. They would respond as well. How is a bit more unclear.

The NVIDIA transaction won't close until late next year or early 2022 because of all the regulatory considerations.

 

[cgsnipinva]

The NVIDIA transaction won't close until late next year or early 2022 because of all the regulatory considerations.

You are correct - but all signs point to it being approved and executed. Which would still support the business case for NVIDIA to investigate a custom ARM/SOC approach.

 

[Pressure]

You are correct - but all signs point to it being approved and executed. Which would still support the business case for NVIDIA to investigate a custom ARM/SOC approach.

They already have that. For example Tegra X1 in Nintendo Switch.

They have an Arm license as well.

 

[cgsnipinva]

They already have that. For example Tegra X1 in Nintendo Switch.

They have an Arm license as well.

Yeah - I mentioned that NVIDIA actually own ARM now. I knew about the Tegra X1 - but what we are talking about is something that would have to be customized to a higher degree to support desktop/laptop computing.

 

[jdb8167]

The M1 has been lauded, by tech journalists and users alike, as an industry game changer for personal computing. I'm curious: Do you agree with that statement? If so, how do you think the personal computing industry has been forever changed by the advent of the M1 and Apple Silicon Macs at large?

Personally, I do think it was a fantastic move for Apple and I would argue that it's a game-changer for the Mac. But I don't see the personal computer industry changing or adopting similar strategies to this as a result of Apple doing it. Do I think we'll see more SoCs in non-Mac personal computers? Abso-friggin-lutely. But we're never going to see a computer maker own the entire hardware and software stack the way Apple now does with Apple Silicon Macs like the ones we now have with M1. Microsoft may have an SQ1 or SQ2 for the Surface Pro X, but that thing is a Qualcomm SoC. Samsung makes SoCs for its phones and tablets, but Samsung isn't Samsung's only customer for those SoCs. And while they do have their own version of Android (albeit one of the worst ones out there), it's not their OS underneath it all! I think Microsoft and Samsung have the best chance of trying to follow Apple on something like this. Maybe NVIDIA, now that they own ARM Holdings. But I think any one of the three of them doing it would take so much time to catch up to Apple. So, no, I don't think it's an "Industry Game Changer"; though I do think it's a massive game changer for the Mac itself. What say you all on this? Do you think the personal computing industry will forever be changed by this? And if so, how and when?

I have too many thoughts on this to put in a forum reply. I wrote a blog post instead. It's long.

Is the M1 an Industry Game Changer

 

[Pressure]

Yeah - I mentioned that NVIDIA actually own ARM now. I knew about the Tegra X1 - but what we are talking about is something that would have to be customized to a higher degree to support desktop/laptop computing.

They don’t, yet. The acquisition is pending regulatory approval.

 

[cgsnipinva]

They don’t, yet. The acquisition is pending regulatory approval.

True - but odds are its going through

 

[Marshall73]

It's a game changer to Apple but with regards to the 'industry', nope not at all.

Thinking you are wrong there. Pressure will come from the likes of HP and Dell on intel and AMD to provide similar power per watt so they can sell slim laptops boasting 18+ hr battery. AMD are well placed to make the jump to unified memory in a laptop (using their APU but packing ram on chip) intel? Not so much.

 

[JohnnyGo]

Everyone says well ____ CPU can beat it and sure. M1 isn't the fastest CPU. It's a base model low power CPU so people don't realize how silly comparing a high end desktop CPU to it is. Wait for M1x or whatever Apple calls it then let's compare.

Patience is something lacking for most people these days...

In 12 months Apple will have an M1X or M2 powered iMac to challenge desktop CPUs and mid-tier GPUs

In 24 months, who knows ? If Apple can pull off an Apple Silicon based Mac Pro (with tons of RAM and dedicated GPUs) no one will be able to compare that performance to anything in Intel/AMD arsenal

 

[throAU]

I don't see MS holding back the WoA licensing for much longer , otherwise it will be a dead platform

We can only hope. I say that as someone who doesn't particularly like windows, but DOES want to see the death of x86 as a platform - its such a massive bodge with 40 years worth of band-aids and sticky tape and at least 15-20 years of big security problems to unravel.

But.... this is the company that heard the negative feedback from Windows 8 re the start menu and "rectified it" with a crappy start button in Windows 8.1. It took them until windows 10 to sort of get the point, and we still don't have something that was as fast, useful and easy to modify as the Windows 7/vista start menu.

It's also a company that has spent the past 10 years almost seemingly deliberately crippling their ARM devices with less memory, crap screens, slow storage, etc.

HOPEFULLY this is also a wake up call to Microsoft as well - that ARM is a valid high end platform and should be treated as such. And HOPEFULLY the past decade has resulted in a shift to a more easily steered company. Microsoft has never been very fast to change course, hopefully this is changing.