Live data from Hacker News

Apple Silicon M1 Disruption

mondaynote.com

61–70 of 201 posts

Re: Apple Silicon M1 Disruption

#61
An easy way to understand the M1 is that Apple simply "iPhone-ized" the PC (or the Mac). Instead of having the modular architecture that PCs are known for (motherboard with separate CPU, RAM, hard drive, GPU, and chips for ports etc), everything is mashed together. This gives the best performance but very little modularity and zero upgradability. It's great for Apple but kind of shuts out all other ecosystems. It's unlikely anything but macOS will ever run on M1 since Apple won't open up the specs, and why should they? It is also another step into the direction of total throw-away culture.

Re: Apple Silicon M1 Disruption

#62

> Now you got a big problem, because neither Intel, AMD, or Nvidia are going to license their intellectual property I keep seeing this line and wonder - "did people miss the news about Nvidia buying Arm?". Sure, there are still plenty of hurdles to pass until it's official, but take a look at the initial piece written by Nvidia about the acquisition: https://nvidianews.nvidia.com/news/nvidia-to-acquire-arm-for... Spe…

It's also outright false - AMD licensed GPU tech to Samsung for the next generation Exynos, built custom SoC for clients in console.

I don't know if AMD could license x86 cores since I don't know what their deal is on x86 IP with Intel.

Re: Apple Silicon M1 Disruption

#63

Where are some benchmarks comparing M1 to the best x86 processors from AMD? Since M1 is only 4 big cores maybe it would be fair to compare it to the 4 core 8 thread APUs and not crush the poor thing with an 8 core version. A1 is cool. Good for Apple, but it's not the computing miracle it's made out to be.

Here they are[0]. The 4 performance cores + 4 efficiency cores of the M1 hold up pretty well against the 8 core AMD chips.

https://www.anandtech.com/show/16252/mac-mini-apple-m1-teste...

Re: Apple Silicon M1 Disruption

#64
post #46

Earlier quoted context omitted.

Loads of people didn’t think much of the iPhone when it came out either. No physical keyboard, other devices have large touch screens, no stylus, can’t copy and paste. Appreciating the iPhone is largely done in hindsight. Best not to make predictions too far into the future.

> Appreciating the iPhone is largely done in hindsight. I’m sorry but that’s definitely not true. Yes, there were a vocal minority of folks who said the iPhone was nothing special but the mass market response was huge and it was heralded as a huge deal by the vast majority. And to go back to my original point: there’s a difference between “this new invention allows you to do new things, but those things aren’t import…

Did you forget when the CEO of Microsoft laughed it off and said it was crap? [1]

[1] https://www.youtube.com/watch?v=eywi0h_Y5_U

Re: Apple Silicon M1 Disruption

#65
post #30

Earlier quoted context omitted.

I think it will benefit Linux but not that way. It will force the x86 vendors to adopt the microarchitectural aspects of the M1 that make it fast in practice, instead of spending so much area/power budget chasing SPEC benchmarks. We will get better chips through competition.

I don’t think these goals are exclusive. I am not sure about the M1, but the A13 and A14 are doing quite well in SPEC benchmarks. If they have compromised these scores, it’s not obvious to me.

I don't think they've compromised on SPEC performance at Apple, but I think they've chased it too hard at Intel. I'm sure I'd rather have more iTLB entries and more L1i cache than that third 512x512 FMA unit, if that's a trade that can be made.

Re: Apple Silicon M1 Disruption

#66
post #31
post #17

Earlier quoted context omitted.

> the true cost of x86 is finally exposed They've reinvented x86 at least twice: microcode and then x86_64. I wouldn't bury x86 yet. Plus a good chunk of the M1 computing prowess is not even the CPU, it's the additional hardware such as the neural engine cores and such. This is something you could put next to a Sparc core, it's not ARM specific.

x86 used microcode from the beginning: https://www.reenigne.org/blog/8086-microcode-disassembled/

Maybe he meant to say micro-operations.

Re: Apple Silicon M1 Disruption

#67

Since these keep floating up, I'm already exhausted with this specific rebuttal to Morehead's review: > As pointed out in John Gruber’s hard takedown of Moorhead’s piece, most of the problems he encountered were software problems, not M1 hardware issues. Morehead's allegiances and work are relevant biases, as are Gruber's. But Morehead's review wasn't of the M1 processor or architecture as a technology. It was a revi…

> Morehead's allegiances and work are relevant biases, as are Gruber's. But Morehead's review wasn't of the M1 processor or architecture as a technology. It was a review of the 13" M1 MacBook Pro as an end-user device.

Moorhead's review literally contains this sentence:

> The new M1 processor is impressive, but far from perfect- it has many warts, that nearly nobody is discussing.

He then proceeds to describe not warts of the M1 as a processor, but rather warts of the 3rd party software ecosystem. I think overall the review is fine and not in bad faith, but it's easy to see how people are latching on to that sentence. If he were honest about making it a review about the overall product experience, rather than the silicon, he wouldn't need to write that sentence.

Re: Apple Silicon M1 Disruption

#68

> Now you got a big problem, because neither Intel, AMD, or Nvidia are going to license their intellectual property I keep seeing this line and wonder - "did people miss the news about Nvidia buying Arm?". Sure, there are still plenty of hurdles to pass until it's official, but take a look at the initial piece written by Nvidia about the acquisition: https://nvidianews.nvidia.com/news/nvidia-to-acquire-arm-for... Spe…

Plus, in some cases it is perfectly fine to integrate technology by soldering a discrete package on a PCB. It is not superior to a SoC in all cases, but neither is the inverse true.

In that sense Intel, AMD, or Nvidia are absolutely conveying their intellectual property to end user products through OEMs.

Re: Apple Silicon M1 Disruption

#69
post #57

Where are some benchmarks comparing M1 to the best x86 processors from AMD? Since M1 is only 4 big cores maybe it would be fair to compare it to the 4 core 8 thread APUs and not crush the poor thing with an 8 core version. A1 is cool. Good for Apple, but it's not the computing miracle it's made out to be.

You’re right, it doesn’t stand up to the multi core performance of the absolute top of the line Xeon chip. But that chip probably cost more than the entire laptop with an M1 in it. Apple started at the bottom of their lineup. Given the performance of the processor compared not only to the processors that it replaces but to others higher in their lineups, it’s extremely impressive. Is it the absolute number one best i…

I keep seeing this divide where people that are used to Apple products using Intel chips are only capable of comparing the Apple Silicon M1 to Intel products.

Here, in response to someone discussing AMD processors, the Intel Xeon is raised as an example.

Sorry, no - the Intel Xeon is not an AMD processor. To be sure, the comparison between the APUs found in $999 laptops should be compared to the M1. The AMD Ryzen 9 4900HS, for example, is a 35W part though that's more of a $1200 laptop part. The Ryzen 7 4800U and Ryzen 7 4800H are both found in $999 laptops though. There are definitely cases where the M1 wins (many single core benchmarks, and efficiency) and other cases where the 4900HS, 4800H and sometimes even 4800U wins (several benchmarks that take advantage of the 8 cores/16 threads, and the higher TDP.)

You don't need to go to the high-end server chips, like the 128 core EPYC to find competition for the M1, but you do have to take a step outside of the Apple ecosystem that only has the old Intel and the new Apple Silicon.

Re: Apple Silicon M1 Disruption

#70
post #56

Earlier quoted context omitted.

It won't reinvent UI, but it signaled the change to ARM and by extension, RISC. To a normal consumer, that's not a big deal (although 17 hour battery life is already pretty huge). To future PC hardware and server/backend infrastructure, that's a huge deal.

Oh, for sure. It’s going to be a big deal to use techies and represents a huge shift in terms of who holds the power in processing. (that said, the vast majority of PCs won’t be able to use it!) But entire industries (e.g. ride hailing) were created in the aftermath of the iPhone’s launch. I just don’t think we’ll see that kind of change reflected in most people’s lives.

The big boosts here are commodity ARM instances on the cloud. These instances already are more power efficient than their intel counterpart and cheaper - but developer machine used to be all intel. Now that has changed.
Post reply on HN