Live data from Hacker News

Intel's disruption is now complete

jamesallworth.medium.com

281–290 of 638 posts

Re: Intel's disruption is now complete

#281
post #196
post #141

Earlier quoted context omitted.

Intel's current design are 1-3 years late depending on which roadmap and sector you are counting. Caused by their 10nm delay. Intel had Tigerlake / WillowCove Design but they cant ship it. They also have Golden Cove. If you have put the design they were suppose to be using now, on a IPC basis Intel might still be in the lead.

So basically given the same process, all companies have almost a similar offering?

Disclaimer: I’m a chip design enthusiast and not a professional, but this is how I understand things right now.

If you read the anandtech article describing the A14/M1 chips, it seems that the ARM ISA lets Apple do some things that are either exceptionally hard or impossible on x86, like making individual cores much “wider” in terms of instruction decode, allowing the A14/M1 cores to do significant parallel execution. This allows their cores to be significantly more efficient in terms of instructions per clock cycle. To be clear this isn’t just the benefits of ARM - Apple is far ahead of other ARM designers as well. On the other hand, the wider the execution pipeline, generally the slower the cores have to be clocked (Im actually not sure why this is, to be honest). This is part of how Intel is able to produce chips in the 5.0+ GHz range.

So it’s not that every designer converges to the same design (internally, they’re quite different). Instead, what’s happening is that all the relevant trade offs (pipeline width and depth, ISA, fab tech, power targets, etc) are leaving competitors within an order of magnitude in performance. And the question of who’s going to take the lead is a question of who’s executing best on all these fronts, and who’s currently got a lot of talent working on areas that will see continued improvement.

But whatever happens, I wouldn’t bet against Apple. Their team has been delivering improvements like clockwork, year over year for quite a long time now.

Re: Intel's disruption is now complete

#282

The entire world has their eggs in one basket - Taiwan Semiconductor Manufacturing Company. Say whatever about Intel, I am hoping that they get back on their feet in next few years with 5nm manufacturing. Since when is competition and choice a bad thing? If you're a chip design firm and you have a 8-month backlog to get your masks made at TSMC, what do you do then? TSMC can be world's best fab but we need to address…

Samsung is not that bad. They just announced a 5nm chip; it's not as good as TSMC but it's not as far behind as Intel either.

Re: Intel's disruption is now complete

#283

My question is why Qualcomm while using the same arch, has not produced a chip as fast as the M1?

Nobody has mentioned it in the comments: because their money maker is in 5G modems (and of course LTE et al before). The ARM SoC is a bonus they throw on for their customers that need the integration (Android vendors, but not Apple).

Re: Intel's disruption is now complete

#284

As someone who worked on Intel's phone chip: we definitely didn't win it. We fucked it up twelve ways to Sunday. Why: giant egos. There were turf wars between Austin, Santa Clara and Israel over who would design it, and the team that won out had long since lost its best principle engineers and had no clue how to spin the architecture to meet the design win. Otellini's hindsight hedge is pure spin: we knew the landing…

Didn't Steve Jobs usually greenlight 2 or 3 teams to design then just pick the most promising one to go to production?

Sometimes it makes sense to have a bit of competition, but having multiple prototypes (often with slightly different design parameters - e.g. a reliable design, an advanced design, and a moonshot) is a better competition than having managers attacking each other with passive-aggressive Powerpoint shows in a conference room.

Re: Intel's disruption is now complete

#285

Earlier quoted context omitted.

> employees had to pay for everything (snacks/drinks/etc.). That’s not entirely true, at least in Hillsboro (I interned there recently). Drinks were free, and so was fruit, but you had to pay for e.g. donuts or chips.

The fruit's hilarious because if you're a contractor the fruit's not free. It's forbidden fruit.

Ah yeah, I forgot to mention that part. Contractors have to pay for a weekly subscription to the drinks and fruit.

Re: Intel's disruption is now complete

#286

The entire world has their eggs in one basket - Taiwan Semiconductor Manufacturing Company. Say whatever about Intel, I am hoping that they get back on their feet in next few years with 5nm manufacturing. Since when is competition and choice a bad thing? If you're a chip design firm and you have a 8-month backlog to get your masks made at TSMC, what do you do then? TSMC can be world's best fab but we need to address…

Completely agree. The world's supply of semiconductors is dependent, in some ways, on the benevolence of the CCP. TSMC is like Arakkis. We need to diversify the supply chain.

Isn't TSMC building a fab in AZ? Anyway I'd say the "spice" in semiconductor production is rare earth elements, since the CCP controls most of their extraction (and has artificially lowered exports in the past)

Re: Intel's disruption is now complete

#287

As someone who worked on Intel's phone chip: we definitely didn't win it. We fucked it up twelve ways to Sunday. Why: giant egos. There were turf wars between Austin, Santa Clara and Israel over who would design it, and the team that won out had long since lost its best principle engineers and had no clue how to spin the architecture to meet the design win. Otellini's hindsight hedge is pure spin: we knew the landing…

I have a theory about why turf wars like this happens: the more successful and wealthy an entity, the greater opportunity there is for a manager to take wealth instead of make it. When you’re scrappy and broke, the only path to success is to make the company successful. When the company is rich and multitudinous, an individual can gain more from politics and turf wars rather than actually trying to push an already hi…

Reminds me of this article I just read. In a totally different domain, but if there is so much wealth, it is easier to just a part of it, instead of creating new wealth.

https://pedestrianobservations.com/2020/11/13/surplus-extrac...

Re: Intel's disruption is now complete

#288

Maybe this is a bit of a side topic, and I'm a layman, but I'm interested to know why the M1 chip is able to perform so surprisingly better? And why Intel or others can't just come up with a similar design, if it's clear from the specs/die images what's being done? Is it years of work to figure out the placement of CPU/GPU/memory units on the die, routings, and come up with new instruction sets? I imagine this wasn't…

Apart from cache, anandtech notes https://www.anandtech.com/show/16226/apple-silicon-m1-a14-de... What really defines Apple’s Firestorm CPU core from other designs in the industry is just the sheer width of the microarchitecture. "Width" enables greater parallel execution of instructions.

One of the interesting comments in that article was how they pinned limited width of x86 decode implementations on variable instruction length. There are obvious code density benefits to the x86 variable length approach (especially in immediate encoding), but I guess the need for realignment creates long critical paths on the frontend. I wonder if the more constrained variable instruction length of RV64GC (32- and 16-bit instructions only) will be able to similarly scale up to 8-wide instruction decode like Apple has been able to do with AArch64.

Re: Intel's disruption is now complete

#289

As someone who worked on Intel's phone chip: we definitely didn't win it. We fucked it up twelve ways to Sunday. Why: giant egos. There were turf wars between Austin, Santa Clara and Israel over who would design it, and the team that won out had long since lost its best principle engineers and had no clue how to spin the architecture to meet the design win. Otellini's hindsight hedge is pure spin: we knew the landing…

I used to work there too but from software perspective, they simply did not understand how software world worked. S/W was also relegated as bunch of coding guys. It was not uncommon to see some random benchmark touted to assert superiority, while customers were using it differently. It'll end up becoming a manufacturing company that builds based on spec from Amazon/Google/Apple etc.

Unlike NVIDIA, which is a “Software Company That Produces Chips”. And unlike Apple, which is a software company that sells hardware to run that software.

Re: Intel's disruption is now complete

#290

Earlier quoted context omitted.

I'm typing this on a Tiger Lake, too, and it is very fast and draws modest power. But, are they making any money on it? If they lose Apple as a laptop customer how much does that hurt their business.

There was a lot of discussion on a related, recent post about Apple buying many of Intel’s most desirable chips. Will be interesting to see whether the loss of a high-end customer translates into more woes for Intel.

Maybe that will just make PCs better now that Apple isn't buying out the best chips.
Post reply on HN