The Rise and Peak of Japanese Semiconductors | Hacker News Reader

The Rise and Peak of Japanese Semiconductors

asianometry.substack.com

99 pointspicture34 comments

About yields in 1980s/1990s.

Friends in semiconductor equipment industry visited clean rooms in USA and Japan, and said the Japanese companies were much much more concerned with keeping them very clean.

Nothing to do with singing the company song in the morning, or liking bonsai https://en.wikipedia.org/wiki/Bonsai, just attention to detail.

Thank you. I can’t count how many people in those days attributed (for example) Japanese manufacturing success to visibly foreign things like karaoke and radio exercises when the real story was something vastly more interesting about factory management and how you coordinate the efforts of a lot of people to make a good car.

Then there's the pull cord on the Toyota assembly lines.

Someone sees something wrong, they reach up, pull the cord, the whole line stops until the issue is resolved. Unlike American lines, where they defective cars would just roll on by.

When Toyota and other Japanese makers started building in the USA, surprise, surprise, their cars made here were just as good as the ones from Japan.

This seems similar to Taiwan's ITRI (Institute for Technology Research Institute) that arranged a technology transfer from RCA, to what would become UMC and Mediatek. RCA even sold all their patents to UMC.

https://meet-global.bnext.com.tw/articles/view/47727

TSMC did not emerge directly from the RCA exchange, but got even better technology from Philips (and this is after the CEO's career at Texas Instruments).

Well researched, but saying "dram" over and over instead of "dee ram" makes my eye twitch.

Oh, I didn't even watch the video. I guess there does seem to be a new breed of blogs that have videos mirroring their articles. I am just glad they also have written articles.

I'm with you on that, written articles > videos any day of the week.

how many drams of whiskey to make an ounce?

He keeps doing it despite frequent complaints. Troll-marketing?

At university, many years ago, one lecturer referred to the BUSY pin signal on a chip as the "bus ypsilon" signal. English was his third language, as an East German German first, then Russian second, so we did not think it was a joke, and that was in East Germany only a few years after reunification so his English practice probably was lacking at the time. I don't think anybody ever corrected him the entire semester.

I suspect it's just a cultural thing. He lives in Taipei and clearly isn't a native speaker. I don't get why people get so bent out of shape over something so insignificant.

I'm a British english speaker, I pronounce it dram. I didn't know there are people who pronounce it D-RAM.

It's typically pronounced D-RAM I assume because RAM is already it's own common acronym people were/are familiar with. DRAM is just a specific type of RAM, Dynamic RAM, hence D-RAM.

Out of curiosity, would you pronounce S-RAM as sram?

And how about SDRAM?

I'm also a British English speaker, and I almost always hear "D-RAM". It's been a while since I'm encountered someone using "Dram".

On the other hand, I've always said "Prom" instead of "P-ROM" (for programmable ROM). I've never considered why there is a difference.

I have the same credentials but I've never encountered anyone pronouncing it as dram. Only D-RAM and S-RAM etc. Although the bike brand SRAM is pronounced as sram, which is fun.

Fourth Brit chiming in to say I’ve only heard ‘dram’ in the context of booze.

Dee-ram, essdee-ram etc for random access memory.

Dramcorp requires you to stay on brand.

There is a four-part documentary I saw on YouTube that went into much more (fascinating) detail.

Part 1: https://youtu.be/ihkRwArnc1k

Part 2: https://youtu.be/uGRNXmWng3M

Part 3: https://youtu.be/ansXGewduN4

Part 4: https://youtu.be/G40YwOg0_B8

Damn, Asianometry seems to be really informative content

I really like the channel's content, but like "The Caspian Report" channel, I kinda wonder if everything we're told is correct and objective. It's great information if it's all verifiable, but who checks the checkers?

Pick a topic you know something about that he talks about. Check him against that.

This is how I got into Caspian.

I think stringent objectivity is overrated anyhow. Id much rather read something biased and well researched than shallow and objective.

> Id much rather read something biased and well researched than shallow and objective.

Seems like a poor way to be informed.

Dont feel like you have to rationalize your reaction.

Which is why I doubt Caspian Report. Anytime he ventures into Western Europe it's either so high level it's basically meaningless (Brexit will change relationships with EU members? You don't say!) or just way off base. Eg. For UK-Newzealand to sign a trade deal, the US decides what happens because... it's geographically between them? I guess assuming the world is flat?

I find the same with Peter Zeihan. In depth researched accurate data.. used to construct always wrong predictions. Eg. Alberta to secede from Canada and become the 51st state because oil pipeline infrastructure.

>For UK-Newzealand to sign a trade deal, the US decides what happens because... it's geographically between them?

No, because it's the world's dominant naval power and NZ is completely dependent upon naval Pax Americana because container shipping.

It's the same reason Australia sent troops to Vietnam despite not having any real desire to do so.

I finally watched one yesterday after saving several to my Watch Later list due to the interesting titles. Unfortunately, I came away feeling that it wasn’t fully accurate, which has put me off watching others.

I've watched a few of his videos and I'm sure a lot of people get a lot of value from them but I've stopped watching them after noticing they are Wikipedia articles converted into a PowerPoint presentation. There is no original research, and I can read faster then I can watch a video.

Sure, but you can listen to him on a commute, some of us have those things again, and the push-"discovery" aspect (versus pull-rabbithole) of YouTube is useful "here is a topic that you seem interested in, why not fill in gaps of your knowledge during your idle time?

I've watched his YT channel for a year or so and really came to like it. It's good enough to scratch the hitch about the semi-con industry without getting into industry-specific publications levels of details.