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

Intel Problems

stratechery.com

Post reply on HN
Print view View original

51–60 of 431 posts

Re: Intel Problems

#51

Post by mhh__ »

raverbashing wrote: post #30
mhh__ wrote: post #5

Earlier quoted context omitted.

Speaking of Itanium, if the x86 dam has truly burst, I'd much rather see something more like the Itanium than RISC-V. Something new. It's a shame the Mill is so secretive, actually, they're design is rather nice.

Nah I think the Itanic concept is dead in the water VLIW works (especially in the way it was done in Itanium - IIRC) when either your workload is too predictable or maybe if your compiler manages to be one order of magnitude smarter than it is today (even with llvm, etc) It seems even M1 prefers to reorder scalar operations than work with SIMD ops in some cases (this is one of its processors)

Itanium is dead but VLIW as a concept is still interesting to me.

If you look at uops executed per port benchmarks you can see that CPUs are far from all seeing eyes.

Re: Intel Problems

#52

Post by varispeed »

klelatti wrote: post #12

Feel this piece ducks one of the most important questions - what is the future and value of x86 to Intel? For a long time x86 was one half of the moat but it feels like that moat is close to crumbling. Once that happens the value of the design part of the business will be much, much lower - especially if they have to compete with an on form AMD. Can they innovate their way out of this? Doesn't look entirely promising…

My view is that currently the only way for Intel to salvage themselves it to go ARM route and start licensing x86 IP and perhaps even open source some bits of tech. They are unable to sustain this tech by themselves nor with AMD anymore. It seems to me when Apple releases their new CPUs I am going to have to move to that platform in order to keep up with the competition (quicker the core, I can do calculations quicker and quicker deliver the product). Currently I am on AMD, but it is only marginally faster than M1 it seems.

Re: Intel Problems

#53

Post by kolbe »

If TSMC is going to be a monopolist fab for x86, then they will ultimately suck all the profits out of the server/desktop markets. This isn't just kinda bad news for Intel/AMD, it's really bad news.

Re: Intel Problems

#54

Post by mhh__ »

sitkack wrote: post #32
mhh__ wrote: post #21

Earlier quoted context omitted.

It's not the memory model I'm thinking of but the cache design, ROB size etc. Obviously this is fairly niche but the friction to making something fast is hugely easier locally.

The vast majority of developers never profile their code. I think this is much less of an issue than anyone on HN would rank it. Only when the platform itself provides traces do they take it into consideration. And even then, I think most perf optimization is in a category of don't do the obviously slow thing, or the accidentally n^2 thing. I partially agree with you though, as the penetration of Arm goes deeper into…

This isn't really about you or me but the libraries that work behind the spaghetti people fling into the cloud.

Re: Intel Problems

#55

Post by thoughtsimple »

How does Moore's law figure into this? I suspect that TSMC runs into the wall that is quantum physics at around 1-2nm. Considering that TSMC has said that they will be in full production of 3nm in 2022, I can't see 1nm being much beyond 2026-2028. What happens then? Does a stall in die shrinks allow other fabs to catch up?

It appears to me that Intel stalling at 14nm is what opened the door for TSMC and Samsung to catch up. Does the same thing happen in 2028 and allow China to finally catch up?

Re: Intel Problems

#56

Post by oldgradstudent »

dogma1138 wrote: post #36
ashtonkem wrote: post #10

I think one day we’re going to wake up and discover that AWS mostly runs on Graviton (ARM) and not x86. And on that day intel’s troubles will go from future to present. My standing theory is that the m1 will accelerate it. Obviously all the wholly managed AWS services (Dynamo, Kinesis, S3, etc.) can change over silently, but the issue is EC2. I have a MBP, as do all of my engineers. Within a few years all of these ma…

At that point if it will be trouble for Intel it would be a death sentence for AMD... Intel has fabs, yes it’s what maybe holding them back atm but it also a big factor in what maintains their value. If x86 dies and neither Intel nor AMD pivot in time Intel can become a fab company they already offer these services, yes no where near the scale of say TSMC but they have a massive portfolio of fabs and their fabs are l…

> Intel can become a fab company

Not unless they catch up with TSMC in process technology.

Otherwise, they become an uncompetitive foundry.

Re: Intel Problems

#57

Post by klelatti »

mhh__ wrote: post #24
klelatti wrote: post #22

Earlier quoted context omitted.

Fair points, but if you're saying that if we moved to a non-x86 (and presumably Arm based) world then its business as usual for Intel and AMD then I'd strongly disagree - it's a very different (and much less profitable) commercial environment with lots more competition.

The likelihood of Intel moving to ARM is probably nil. They have enough software to drag whatever ISA they choose with them, whereas AMD bringing up an ARM core could be fairly herculean as they have to convince their customers to not only buy their new chip but also trust AMD with a bunch of either missing or brand new software.

The days when Intel could single handedly successfully introduce a new (incompatible) ISA are long gone (if it ever could). I expect they will stick with x86 for as long as possible.

Re: Intel Problems

#58

Post by lostapathy »

mhh__ wrote: post #20
ashtonkem wrote: post #16

Earlier quoted context omitted.

Professional laptops don’t last that long, and a lot of developers are given MBPs for their work. I personally expect that I’ll get a M1 laptop from my employer within the next 2 years. At that point the pressure to migrate from x86 to ARM will start to increase.

You miss my point - if I am seriously optimizing something I need to be on the same chip not the same ISA. Graviton2 is a Neoverse core from Arm and it's totally separate from M1. Besides, Apple don't let you play with PMCs easily and I'm assuming they won't be publishing any event tables any time soon so unless they get reverse engineered you'll have to do it through xcode.

This is typical Hacker News. Yes, some people "seriously optimize" but the vast majority of software written is not heavily optimized nor is it written at companies with good engineering culture.

Most code is worked on until it'll pass QA then thrown over the wall. For that majority of people, an M1 is definitely close enough to a graviton.

Re: Intel Problems

#59

Post by mhh__ »

lostapathy wrote: post #58
mhh__ wrote: post #20

Earlier quoted context omitted.

You miss my point - if I am seriously optimizing something I need to be on the same chip not the same ISA. Graviton2 is a Neoverse core from Arm and it's totally separate from M1. Besides, Apple don't let you play with PMCs easily and I'm assuming they won't be publishing any event tables any time soon so unless they get reverse engineered you'll have to do it through xcode.

This is typical Hacker News. Yes, some people "seriously optimize" but the vast majority of software written is not heavily optimized nor is it written at companies with good engineering culture. Most code is worked on until it'll pass QA then thrown over the wall. For that majority of people, an M1 is definitely close enough to a graviton.

> typical hacker news

Let me have my fun!

Re: Intel Problems

#60

Post by Someone »

mhh__ wrote: post #13
ashtonkem wrote: post #10

I think one day we’re going to wake up and discover that AWS mostly runs on Graviton (ARM) and not x86. And on that day intel’s troubles will go from future to present. My standing theory is that the m1 will accelerate it. Obviously all the wholly managed AWS services (Dynamo, Kinesis, S3, etc.) can change over silently, but the issue is EC2. I have a MBP, as do all of my engineers. Within a few years all of these ma…

I can see this happening for things that run in entirely managed environments but I don't think AWS can make the switch fully until that exact hardware is on people's benches. Doing microbenchmarking is quite awkward on the cloud, whereas anyone with a Linux laptop from the last 20 years can access PMCs for their hardware

I would think the number of developers that have “that exact hardware” on their bench is extremely small (does AWS even tell you what cpu you get?)

What fraction of products deployed to the cloud even has its developers seen doing _any_ microbenchmarking?

Post reply on HN