Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Apple Silicon M1 Die-Shots

techinsights.com

Post reply on HN
Print view View original

41–50 of 114 posts

Re: Apple Silicon M1 Die-Shots

#41

Post by MichaelZuo »

wtallis wrote: post #33
wmf wrote: post #3

Earlier quoted context omitted.

For very old chips reverse engineering is relatively easy; here's one example: http://www.righto.com/2020/04/inside-am2901-amds-1970s-bit-s... For modern chips you need an electron microscope and there are a lot of metal layers to deal with.

You don't need an electron microscope to discern large-scale repeated structures like CPU and GPU cores, or to tell the difference between random logic and highly regular memory arrays (usually cache, but sometimes you can see microcode or register files if you look at a small enough area). An electron microscope is only useful if you want to fully reverse-engineer the entire circuit schematic, but that's overkill fo…

So none of the underlying layers matter much for the large and intermediate scale structures?

Re: Apple Silicon M1 Die-Shots

#42

Post by MichaelZuo »

mdonahoe wrote: post #20
systemvoltage wrote: post #16

Earlier quoted context omitted.

The inquiry I am making is more general - infantilization of consumers and the idea of dumbing things down for consumers like they're blobs of meat with cash to spare. You don't need to see the cartoonized version of it either. May be its easier as a schematic to explain? Not sure. A list of M1's features would work equally well. I think its just marketing pizzaz.

They are simplifying the presentation for educating a wider audience of non-ECEs. This is not a new phenomenon in human communication. But we’re all just blobs of meat, that’s true. Maybe one day Apple will better tailor their marketing to sentient Bitcoin algorithms.

Though there’s still the possibility that they could add it into the detailed specifications page, or perhaps even a link in the footnotes of the specifications page? Surely that would be a better match for the target audience that would care.

Re: Apple Silicon M1 Die-Shots

#43

Post by im3w1l »

culturestate wrote: post #39
userbinator wrote: post #38

Earlier quoted context omitted.

The degree to which they hype UMA (a term of art within the PC industry for decades ) on that page is amusing --- if Intel did the same sort of marketing for its integrated GPUs, they would almost certainly be laughed at.

This is consumer marketing; “people who have a solid grasp of fundamental SoC architecture” are not the target audience.

What do people with such a grasp have to say about it?

Re: Apple Silicon M1 Die-Shots

#44

Post by sydney6 »

dilippkumar wrote: post #2

How does one identify which regions of a die-shot is the CPU or GPU? Are the structures on silicon easily recognizable from such pictures? How far can someone with time and money go towards reverse engineering a chip from such pictures?

This [1] is a Talk from John McMaster about "Low-Cost IC Emission Reverse Engineering" using, iirc, commodity cameras.

[1]https://www.youtube.com/watch?v=iAuLXg2fVz0

Re: Apple Silicon M1 Die-Shots

#45

Post by wmf »

im3w1l wrote: post #43
culturestate wrote: post #39

Earlier quoted context omitted.

This is consumer marketing; “people who have a solid grasp of fundamental SoC architecture” are not the target audience.

What do people with such a grasp have to say about it?

M1's performance is great, but there isn't much evidence to attribute it to the memory subsystem. Also, people wanting to fit 32GB worth of apps into 16GB is just wishful thinking that isn't supported by evidence either.

Re: Apple Silicon M1 Die-Shots

#46

Post by mamp »

wmf wrote: post #45
im3w1l wrote: post #43

Earlier quoted context omitted.

What do people with such a grasp have to say about it?

M1's performance is great, but there isn't much evidence to attribute it to the memory subsystem. Also, people wanting to fit 32GB worth of apps into 16GB is just wishful thinking that isn't supported by evidence either.

Quite a lot of testing in Final Cut and Logic etc showing very good performance and efficient swap. So for most consumer apps it works well. No magic if you need a big block of memory for ML or analytics.

Re: Apple Silicon M1 Die-Shots

#47

Post by wtallis »

MichaelZuo wrote: post #41
wtallis wrote: post #33

Earlier quoted context omitted.

You don't need an electron microscope to discern large-scale repeated structures like CPU and GPU cores, or to tell the difference between random logic and highly regular memory arrays (usually cache, but sometimes you can see microcode or register files if you look at a small enough area). An electron microscope is only useful if you want to fully reverse-engineer the entire circuit schematic, but that's overkill fo…

So none of the underlying layers matter much for the large and intermediate scale structures?

With the right camera modifications, you don't have to remove any layers from a chip to discern the core count: https://www.flickr.com/photos/130561288@N04/48258310027/ That's an IR photograph taken with a Sony APS-C camera that the IR-blocking filter has been removed from. That's enough to make out the cores and L3 cache slices. (The rest of Fritzchens Fritz's flickr uploads are worth checking out too, especially the video showing how he preps a die for visible light photography.)

Re: Apple Silicon M1 Die-Shots

#48

Post by MichaelZuo »

wtallis wrote: post #47
MichaelZuo wrote: post #41

Earlier quoted context omitted.

So none of the underlying layers matter much for the large and intermediate scale structures?

With the right camera modifications, you don't have to remove any layers from a chip to discern the core count: https://www.flickr.com/photos/130561288@N04/48258310027/ That's an IR photograph taken with a Sony APS-C camera that the IR-blocking filter has been removed from. That's enough to make out the cores and L3 cache slices. (The rest of Fritzchens Fritz's flickr uploads are worth checking out too, especially th…

That’s cool! How would the major companies reverse engineer these chips?

Re: Apple Silicon M1 Die-Shots

#49

Post by manigandham »

amelius wrote: post #21
systemvoltage wrote: post #6

Kind of sad to think about this - Why does marketing dept not actually show these and instead put up a cartoonized version of Apple's chips [1]? That's in every keynote. This is a rhetorical question, the real mystery is why does an average consumer want to see unreal things to convince them its cooler/better? You'd think seeing actual die shots would be more convincing to average consumer, but they love marketing th…

I think if Apple wants to be a tech leader they should open up more. The secrecy is totally unnecessary.

Are you claiming that they are not a tech leader right now?

Re: Apple Silicon M1 Die-Shots

#50

Post by dannyw »

segfaultbuserr wrote: post #9

Only a die-shot, it's disappointing that there's no comment or explanation. But the website advertises: > A DFR report will be available to TechInsights subscribers on Apple's M1 disruptive chip architecture in January 2021. So I assume the actual write-up is still a work-in-progress. Perhaps we just need to be patient and come back a month later.

The actual report will be thousands of dollars because it's for industry professionals trying to compete with Apple's M1.
Post reply on HN