FullyFunctional

This is fascinating, but I was curious about the last two QR codes. The left one is scannable on my iPhone (iOS 17.4.1) leading to http://example.com/AAE..._w8fL whereas the one on the right gets only http://example.com (both Safari and Firefox). Is this an iOS URL length limitation?

Also not OP, but 90% of everything I do happens in an SSH session. It doesn't really matter what I run on as long as the terminal isn't completely broken. I run `ssh TheHost exec tmux a` from all sorts of devices. A NeXTstation would do fine except I prefer my custom USB-only keyboard.

That was exactly my experience as well. I'm not really sure why that was, but I think that the breadboard scaled much farther and could be extended with more of them (mine were interlocking). I built some moderately complex circuits, but debugging them could be fairly tedious.

I don't know the current status, but iOS's design meant apps at least used to be _far_ more efficient than the equivalent Android, and thus Apple could ship lower end hardware to deliver the same or better performance to the end user. The was a measurable fact, thus objective.

The same is still somewhat true for macOS where for the average web browsing user, macOS + Safari needs less memory than you would need on Windows. This is also a fact, but it doesn't help power users.

To add more evidence to your point: SJ loved wall gardens and consistently fought against extensibility. The Apple II only got extension slots because the other Steve insisted. All of the compact Macs have very limited to no extensibility.

It's so ironic that Apple was pushing the (open) Web apps in the early days of the iPhone (out of necessity of course).

Links/pointers? I know only of Achronix' original Speedster series, which was alas abandoned for a conventional architecture ("Power" was the reason I was told).

The piece is brilliant and gives great insight.

It had previously been reported (and exposed here) that the "Osborne effect" wasn't what killed the company, but (IIRC) quality issues with parts (floppy drive?) that delayed the Executive. Regardless, essentially the fall of the Osborne Computer Corp. is the age old story of a company growing faster than its organization and eventually collapsing.

This is really old news and sadly there wasn't anything new in this article, the issue remains. IBM [used to?] use eDRAM for their last-level caches which is a classic latency/capacity trade-off that seems reasonable.

There are some amusing typos, like "single-core"/"dual-core" where they meant port rather than core.

I apparently can't edit the submission.

It's interesting to see this the world fastest counter (at 600+ GHz) is implemented in this technology.

I'm sorry I thought I did, but maybe I lost it. I'll see if I can fix it.

That's far from clear to me. However, his current passion isn't silicon; I shared it with HN: https://news.ycombinator.com/item?id=39449596

Ivan Sutherland on the next revolution (and it isn't quantum computing). Note, the Q&A session following the presentation elaborates on these ideas.

There have been other successes since, perhaps most famously Fulcrum (acquired my Intel in 2019) with their fully asynchronous networking switches.

ADD: another impressive result was Ivan Sutherland's (yes, that Ivan Sutherland) crazy fast 8x8 cross bar. He's still active in the topic: https://arc.cecs.pdx.edu/publications/

and Montalvo, Intrincity, and probably more I don't know about.

As always "the overnight success were years in the making".

That's WAY-oversimplifying it. There are so many compromises and design choices and not just "One Cool Trick". Generally speaking the M-series traded frequency for IPC by going for wider and much deeper reorder window. This is easier to do with Aarch64 than x86 and generally is more power efficient (but always the devil is in the details).

The higher memory bandwidth doesn't do anything for ST perf.

Exactly. There are so many dimensions across which to evaluate it. What I care about the most is 1) ST thread (running my personal workload which is inherently single threaded), and 2) Rust compilation (MT compile/ST link).

For 1) my fastest iron is i9-13700KS and Apple M2. They are very close. My Zen 3 is great and is notably more power efficient, but I'll evaluate 14700KS-Zen 5-M3 when possible.

ADD: because of winter I'm loving my i9-13700KS (not kidding, my office would be freezing without it), but come summer I'll care about efficiency.

I don't love Rust because of "fast code", I love it because it leaves me less stressed and less frustrated, IOW, happier.

There's so much emphasis on the memory safety of Rust that it overshadows everything else that is awesome about it. Rust (largely) gives me the power of Haskell strong types, but with fantastic tooling and a much much richer library (if you include crates.io).

I find the difficulty of Rust overstated; much of the borrowing challenges can be avoided by cloning/copying wherever required and limiting borrowing to where that approach becomes an actual performance problem. As Python's success proves, developer productivity trumps runtime efficiency.

I find my (NorCal) PG&E bill extremely (and deliberately?) inscrutable, but for the most recent statement, the Off-peak net price was $0.427/kWh and $0.466/kWh on-peak.

WASM uses a block structured IR (that is, no arbitrary gotos) and I would characterize it as in between of the [serialized] Juice AST and the flat Java bytecodes.

From the compiler's point of view the Juice AST is the same as source code so I'd expect the Juice JIT to have slight more code generation flexibility than the type-erased WASM.

Maybe I wasn’t clear; I meant after Itanium failed, Microsoft refused to support yet another 64-bit extention of x86 as they already had AMD64/x64 (and IA-64 obviously)

Fair point, Itanium's impact on SPARC might be less than I stated, but Alpha is very clearly documented.

Mostly true, but there are exceptions:

Apple does really well on its rare acquisitions, but they aren't very public as they get successfully absorbed. PA Semi, Intrinsity, more I can't remember.

ATi and Xilinx have by all accounts worked out really well for AMD.

That was not the point of the Altera acquisition. The point was the fill Intel's fabs, but the fab fiasco left Altera/Intel-FPGA without a product to sell (Stratix 10 -- 10nm -- got years of delay because of that). Meanwhile Xilinx was racing ahead on TSMC's ever shrinking process.

This. If you follow GamersNexus, there are stories every month about just how much the Arc drivers have improved. If this rate continues and the next-gen hardware (Battlemage) actually ships, then Intel might be a serious contender for the midrange. I really hope Intel sticks with it this time as we all know it takes monumental effort to enter the discrete GPU market.

That's not at all what happened. Intel's 64-bit story was EPIC/IA-64/Itanium and it was an attempt to gain monopoly and keep x86 for the low-end. AMD64 and Itanic derailed that idea so completely that Intel was forced by Microsoft to adopt the AMD64 ISA. Microsoft refused to port to yet-another incompatible ISA.

Had Itanium been a success then Intel would have crushed the competition (however it did succeed in killing Alpha, SPARC, and workstation MIPS).

Intel has an amazing track record with acquisitions -- almost none of them work out. Even the tiny fraction of actually good companies they acquired, the Intel culture is one of really toxic politics and it's very hard for acquired people to succeed.

I wish Pat well and I think he might be the only who could save the company if it's not already too late.

Sourced: worked with many ex-Intel people.

POSTSCRIPT: I have seen from the inside (not at Intel) how a politically motivated acquisition failed so utter spectacularly due to that same internal power struggle. I think there are some deeply flawed incentives in corporate America.

OMG, please please do not turn it into an aquarium. That thing is so rare that it would be a crime! If you aren't up for restoring it, maybe consider selling it (for much $$$) to somebody who would.

The front looks great IMO. Kudos. I have a few compacts myself and one project I need to try is tossing a Pi 5 in there and having it display on the original CRT. Would make for a cute little SSH terminal.

The most impractical Mac that I'd love to see reviewed is the iLamp, aka iMac G4.

All of this is accurate and just want to add how hard it is to appreciate how revolutionary BASIC was at the time of creation (at Dartmouth, when everything was batch oriented FORTRAN). It was probably the first _widely_ available example of interactive computing, and as we have to come to understand, instant feedback is critical for learning and a boon for productivity.

However, using line numbers for editors wasn't AFAIK unusual at the time and I think making the semantics depend on them was unfortunate. However 20/20 hindsight etc.

I have working hardware from every decade, oldest is from 1982, but my daily driver and what I'm typing on now is a 2019 iMac 5K and Apple still doesn't have anything all-in-one to replace it. My keyboard is from 2002 (the USB version of the Kinesis I bought in ~ 1993).

My oldest computer still in 24/7 use is my 2014 FreeNAS, but it's using a monitor that's so old I don't even know.

Old hardware is great, but I do appreciate just how much better LCD panels have gotten over time.