Hacker News

Top stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Laya the open source version of Jev(convaiinnovations.com ↗)
    34comments
  2. Human brain is two separate organs, Stanford Medicine-led research finds(stanford.edu ↗)
    146comments
  3. AI-generated posters don’t have to be horrible(john.hartnup.uk ↗)
    297comments
  4. “The Secret Life of Circuits” is here(coredump.cx ↗)
    31comments
  5. Android 17 is the first since 3.x to add new APIs without releasing to the AOSP(grapheneos.social ↗)
    500comments
  6. GPT-6 Astra Solves a WWI German Radio Cipher(prinzai.com ↗)
    103comments
  7. San Francisco Onion Futures Company(onionfutures.com ↗)
    88comments
  8. If math is more than proof, we need to better celebrate the rest of it(terrytao.wordpress.com ↗)
    139comments
  9. Communication by means of modulated Johnson noise(pnas.org ↗)
    12comments
  10. Cloudflare Quick Tunnels(cloudflare.com ↗)
    294comments
  11. From Stonemasons to Carpenters(thelastsoftwareengineer.substack.com ↗)
    3comments
  12. How to Write with an LLM(sockpuppet.org ↗)
    352comments
  13. You can run Git on object storage if you re-make packfiles(tigrisdata.com ↗)
    21comments
  14. SDCC – Small Device C Compiler(sourceforge.net ↗)
    20comments
  15. Ray Ozzie and the Optimism of Being Early(reproof.app ↗)
    2comments
  16. Saving another 100TB of RAM(cloudflare.com ↗)
    86comments
  17. Science Is Open Software(jepedersen.dk ↗)
    45comments
  18. Why building a Rust LSP is hard(rust-glancer.github.io ↗)
    42comments
  19. How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip(ieee.org ↗)
    105comments
  20. Ctenophores: Wonders of Biology(quantamagazine.org ↗)
    6comments
  21. NASA-IBM Lunar Foundation open-Source Geospatial AI Model(usra.edu ↗)
    5comments
  22. The first new cat species discovered in 100 years(nationalgeographic.com ↗)
    113comments
  23. OpenJev(openjev.com ↗)
    273comments
  24. Goroutine Leak Profiles(go.dev ↗)
    6comments
  25. Show HN: I wrote a custom assembler for CHIP-8 in C++(github.com/tackx ↗)
    3comments
  26. Show HN: Cactus Needle 3: 8-29MB automation models can match DeepSeek V4 Flash(cactuscompute.com ↗)
    89comments
  27. Photon-Emission-Guided Laser Fault Injection Enables RP2350 Secure Debug(ledger.com ↗)
    76comments
  28. Veronese's Dogs(publicdomainreview.org ↗)
    2comments
  29. Warez: The Infrastructure and Aesthetics of Piracy (2021)(archive.org ↗)
    94comments
  30. Cache-to-Cache: Direct Semantic Communication Between LLMs (2025)(arxiv.org ↗)
    16comments

Apple Computer: Scorpius Architectural Specification (1988)

63 pointsby 6y agoarchive.org
20 comments
6y agoHN ↗

The archive (seemingly erroneously) says this was created in 1989. The text itself, however, says it was from 10/88.

6y agoHN ↗

You should probably link to this instead: https://archive.org/details/scorpius_architecture

(This way has more compatibility with various computers and software that you try to view it. Also, for some reason the "FULL TEXT" link does not link directly to the file, nor does there seem to be a link from there as far as I can find, but if you change "stream" in the URL to "download" then it will access the file.)

6y agoHN ↗

I disagree with that. You can hit "← Back to item details" to trivially get to the more information panel, while this presents content in a better way than linking to that link.

6y agoHN ↗

Thanks for this. I couldn’t really read it on my iPhone otherwise

6y agoHN ↗

Coincidentally (or not?), Power Mac G5's later used the PowerPC 970MP (a dual-core part) which also carried the code name 'Antares'.

6y agoHN ↗

So... how soon can we expect a HDL implementation of this architecture? :)

6y agoHN ↗

I can't believe the paucity of comments on this article. Surely, this is of interest to HNers? Am I just the only one that's never heard of this? It sounds awesome, and confirms my personal bias that an Apple+Sun merger (prior to the Oracle acquisition) could have been terrific. We all complain about Apple's slipping quality these days, but they used to (?) have some amazing engineers.

Edited to add: Does anybody know why this got nixed? Would love to know the backstory to this.

6y agoHN ↗

1988 would put this squarely in the 'Apple trying to convince Motorola to develop the 88k' era, most likely this was a backup plan in case they failed (they did, but by the time the 88k was killed off, they'd managed to get AIM founded and the 601 was 'in the pipeline')

6y agoHN ↗

I didn’t know that apple considered the M88K. I used DG Aviion servers with DG/UX, which were all multi cpu 88ks, and they were awesome, a staple of the industry I was in at the time (medical laboratories).

6y agoHN ↗

Wow, now there's a name I haven't seen in a long long time... =)

6y agoHN ↗

There was an AMD 29000 Macintosh. But Apple execs were afraid that would piss off Motorola too much, and they still needed their parts. So then there was an 88k Macintosh. Execs decided that Moro didn’t have the volume to compete with Intel, so they put them together with IBM to form AIM and develop PPC (they considered HPs Precision RISC as well) and went with that until it was clear even those two couldn’t keep up with Intel- or with Apples direction.

6y agoHN ↗

Was there any serious consideration for an ARM Macintosh derived from Mobius, or was that strictly meant as an Apple II replacement and then canned in favor of Newton?

6y agoHN ↗

I’m fascinated by this.

Hey, at least we know they had proper Plan B’s!

I see nothing else about this online, so this must be pretty recent public information

6y agoHN ↗

This was canceled at the point where Apple needed to commit to spending a lot of money to implement the design in silicon. I heard it was canceled because upcoming COTS processors looked like they are going to be faster and cheaper. (Source: I worked in a different part of Apple ATG and was talking with some of the people on the project.)

6y agoHN ↗

I’m actually one of the architects of this processor. The story that I tell wasn’t so much that they were at the point of needing to spend real money, but it was pointed out that to compete with Intel/Motorola they would have to pay for development over and over- and the couldn’t afford (then) to spend the money to do that.

Ironically, that decision led (pretty much directly) to the formation of ARM (and PowerPC, of course)

6y agoHN ↗

Really interesting architecture. The way they tried to make the processor units flexibly become SIMD or MIMD is thought provoking. Thanks for sharing this.

6y agoHN ↗

Late-1980s Apple’s plan for architecting their own RISC CPU. Never made it to production Macs.