throw34

Do you have a reference for that? 5 years is a pretty long time but I suppose it depends a lot on how one defines the start of a design.

Usually it’s what’s being run inside the terminal that matters. ssh? no big deal. ffmpeg? can slow a machine down and can have a significant memory footprint.

YMMV

Probably somewhere around two years before the M1 announcement is when development started.

It’s a lot of work to design one of these parts.

https://anysilicon.com/asic-design-flow-ultimate-guide/

I’m not sure what point you’re arguing anymore.

I’m arguing against this statement. I don’t think it’s true.

“ You're looking in the wrong place. The magic comes from TSMC, not Apple.”

My point is a good process is necessary but not sufficient to make a part competitive with Apple on performance per watt. There’s a lot of “magic” to go around.

That’s an interesting take given this thread because Apple has explicitly stated they don’t prioritize performance, they prioritize performance per watt, which is not the same thing. And that also shows there’s a whole design space here and just focusing on one dimension, process node, is overly simplistic.

But doesn’t your statement prove that there are decisions and things the user of a node can make which effect the end result? Therefore, is not just process node that matters.

I suppose traditionally this is what one learns during the fiancée stage of the relationship and given your accelerated timeline that’s definitely going to be a challenge. Unfortunately the only way I know to tell is to actually go through a couple of tough spots and see what happens and forcing it might not work. On the plus side, if you do find a test that is predictive and can also be used early in a relationship, you’ll be a best selling author of relationship books for sure.

> A lot of dating books recommend getting really specific about what you want

Also, from my personal experience, what you want is going to change over time. And even if someone checks your boxes today, they’re unlikely to be that same person in 10 years.

Being a parent changed me a lot (hopefully for the better) and 20s me, 30s me, 40s me are all pretty different guys in terms of priorities, willingness to listen to others, etc.

I echo other comments and say the priority is having a communication channel and the willingness to adapt.

I get your point but was Voyager a bank?

https://www.investvoyager.com/investorrelations/overview

“ Voyager Digital Ltd. is a fast-growing, publicly traded cryptocurrency platform in the United States founded in 2018 to bring choice, transparency, and cost efficiency to the marketplace. ”

Do you know the limit for the part? If it’s 120C then the M1 is leaving 26C on the table and the M2 left only 12C.

Without the specs these measurement don’t mean much.

"The R1000 addresses 64 bits of address space instantly in every single memory access. And before you tell me this is impossible: The computer is in the next room, built with 74xx-TTL (transistor-transistor logic) chips in the late 1980s. It worked back then, and it still works today."

That statement has to be coming with some hidden caveats. 64 bits of address space is crazy huge so it's unlikely the entire range was even present. If only a subset of the range was "instantly" available, we have that now. Turn off main memory and run right out of the L1 cache. Done.

We need to keep in mind, the DRAM ICs themselves have a hierarchy with latency trade-offs. https://www.cse.iitk.ac.in/users/biswap/CS698Y/lectures/L15....

This does seem pretty neat though. "CHERI makes pointers a different data type than integers in hardware and prevents conversion between the two types."

I'm definitely curious how the runtime loader works.

This has been a popular topic for some time in the computer architecture space.

“ The End of the Road for General Purpose Processors & the Future of Computing - Prof. John L. Hennessy”

https://www.csail.mit.edu/news/end-road-general-purpose-proc...

Or maybe have the object code for the executable as one static constant and jump to it on the one and only line of the program: speed, size and one line of code (maybe two depending on how the counting is done)

This MIT OCW takes a student from basically nothing to an almost compliant RV32i, complete with videos, worked examples and lecture notes.

https://ocw.mit.edu/courses/6-004-computation-structures-spr...

Me personally, all the time.

In fact, I am taking a short break from a problem that reduces to graph coloring with 2 colors. [1] If you interview with our team we will absolutely ask LC style algorithmic problems because one will definitely use that knowledge with us, a lot.

The problem is not LC style interviews per se. The problem is not enough companies tailoring their interview to the role they're trying to fill.

[1] https://en.wikipedia.org/wiki/Graph_coloring

And is it correct?

http://www.mathematicsmagazine.com/Articles/NicolaTesla.php#...

“Tesla’s physics requires a quite different understanding of mathematics, in some extent it is sacral in the spirit of Pythagoras. Pythagoras considered that numbers and subjects are interrelated. They correspond to each other in property due to informational and mathematical aspects of matter existing as one of the manifestations of the Divine Logos.”

Not totally sure how to interpret that…

On the other hand, it's not like coffee houses are a recent invention.

https://livesandlegaciesblog.org/2017/05/24/coffee-a-revolut... https://blog.publicgoods.com/a-brief-history-of-coffee-in-am...

Check on the section on WW2, "An average of 20 pounds of coffee were consumed a year per adult. ... This amount was cut in half to 10 pounds a year, as one pound of coffee was allotted to each person over 15 years old every five weeks."