Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > alt.folklore.computers > #234501 > unrolled thread

Today, 50 years ago

Back to article view | Back to alt.folklore.computers

Contents

  Today, 50 years ago ram@zedat.fu-berlin.de (Stefan Ram) - 2026-03-31 19:58 +0000
    Re: Today, 50 years ago Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-03-31 20:59 +0000
      Re: Today, 50 years ago Jason H <jason_hindle@yahoo.com> - 2026-04-14 22:55 +0000
        Re: Today, 50 years ago Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-04-15 05:47 +0000
          Re: Today, 50 years ago oldernow <oldernow@dev.null> - 2026-04-15 11:38 +0000
            Re: Today, 50 years ago cross@spitfire.i.gajendra.net (Dan Cross) - 2026-04-15 13:13 +0000
              Re: Today, 50 years ago oldernow <oldernow@dev.null> - 2026-04-15 13:57 +0000
              Re: Today, 50 years ago moi <findlaybill@blueyonder.co.uk> - 2026-04-15 19:18 +0100
                Re: Today, 50 years ago cross@spitfire.i.gajendra.net (Dan Cross) - 2026-04-15 18:28 +0000
            Re: Today, 50 years ago Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-04-15 22:08 +0000
          Re: Today, 50 years ago Peter Flass <Peter@Iron-Spring.com> - 2026-04-15 07:35 -0700
            Re: Today, 50 years ago Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-04-15 22:05 +0000
              Re: Today, 50 years ago Leonard Blaisdell <leoblaisdell@sbcglobal.net> - 2026-04-16 00:01 +0000
                Re: Today, 50 years ago Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-04-16 00:44 +0000
                  Re: Today, 50 years ago Nuno Silva <nunojsilva@invalid.invalid> - 2026-04-16 09:23 +0100
        Re: Today, 50 years ago cross@spitfire.i.gajendra.net (Dan Cross) - 2026-04-15 13:11 +0000
          Re: Today, 50 years ago Nuno Silva <nunojsilva@invalid.invalid> - 2026-04-15 14:31 +0100
            Re: Today, 50 years ago cross@spitfire.i.gajendra.net (Dan Cross) - 2026-04-15 14:04 +0000
              Re: Today, 50 years ago John Ames <commodorejohn@gmail.com> - 2026-04-15 11:21 -0700
                Re: Today, 50 years ago cross@spitfire.i.gajendra.net (Dan Cross) - 2026-04-15 18:41 +0000
                Re: Today, 50 years ago Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-04-15 22:03 +0000
            Re: Today, 50 years ago scott@slp53.sl.home (Scott Lurndal) - 2026-04-15 14:55 +0000
          Re: Today, 50 years ago Peter Flass <Peter@Iron-Spring.com> - 2026-04-15 07:44 -0700
            Re: Today, 50 years ago cross@spitfire.i.gajendra.net (Dan Cross) - 2026-04-15 16:13 +0000
          Re: Today, 50 years ago scott@slp53.sl.home (Scott Lurndal) - 2026-04-15 14:48 +0000
            Re: Today, 50 years ago cross@spitfire.i.gajendra.net (Dan Cross) - 2026-04-15 16:14 +0000
              Re: Today, 50 years ago John Levine <johnl@taugh.com> - 2026-04-15 17:19 +0000
                Re: Today, 50 years ago cross@spitfire.i.gajendra.net (Dan Cross) - 2026-04-15 18:43 +0000
                Re: Today, 50 years ago Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2026-04-15 19:55 +0000
                Re: Today, 50 years ago scott@slp53.sl.home (Scott Lurndal) - 2026-04-15 20:08 +0000
                  Re: pretty good chips, Today, 50 years ago John Levine <johnl@taugh.com> - 2026-04-15 21:26 +0000
    Re: Today, 50 years ago Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-04-05 20:37 +0000

Page 2 of 2 — ← Prev page 1 [2]

#234692

On Wed, 15 Apr 2026 11:21:01 -0700, John Ames wrote:

> Irix's "Interactive Desktop" was much nicer and more polished than
> the GUIs offered by other commercial Unices at the time. Had a
> decent GUI package manager, as well.

Also remember, like any *nix, the GUI was a separate, replaceable,
modular layer, not baked into the OS kernel.

Not sure this was true of NeXT. Any sign of modularity certainly
disappeared by the time Apple took over.

[toc] | [prev] | [next] | [standalone]

#234675

Nuno Silva <nunojsilva@invalid.invalid> writes:
>On 2026-04-15, Dan Cross wrote:
>
>> In article <10rmgl8$gjuj$1@dont-email.me>,
>> Jason H  <jason_hindle@yahoo.com> wrote:
>>
>>>To be fair to them, the whole move to UNIX and, much later, Apple Silicon
>>>was ultimately pretty innovative when compared to the rest of the market.
>>>Unified memory ftw. 
>>
>>
>> Eh....  I dunno.
>>
>> Apple Silicon is pretty nice, though I don't know about ARM.
>> The obvious alternative is RISC-V, but it was not mature enough
>> for use in a high-end product at the time they moved away from
>> Intel.
>>
>> Unix as the basis for an OS wasn't super innovative,
>> particularly when you consider that they were building on NeXT's
>> technology, which was already Mach+4.3BSD, and predated macOS by
>> a few decades.
>
>On that field, I guess there's also SGI Irix? (Better known by its stage
>name, "It's a UNIX system, I know this!")

Irix was a flavor of unix, yes.  When I was at SGI, I was hired to work on
a distributed version of Irix (called Teak).   I had just left Unisys
after spending almost a decade working on a distributed version of SVR4
for the Unisys OPUS MPP machines.

There were some nice things in Irix, but it was also hobbled by
some of the warts in the MIPS architecture (particularly
the software-based TLB handling).

The Teak project ended up being cancelled as we moved to linux
development and worked on an early hypervisor (called Crucible)
to run Windows and Linux simultaneously on a single dual
processor Kayak (1998/1999).

[toc] | [prev] | [next] | [standalone]

#234673

On 4/15/26 06:11, Dan Cross wrote:
> 
> Apple Silicon is pretty nice, though I don't know about ARM.
> The obvious alternative is RISC-V, but it was not mature enough
> for use in a high-end product at the time they moved away from
> Intel.
> 

 From what I read, RISC-V still isn't quite there yet. I think the 
performance is still below other chip families, but, of course, a lot of 
engineering went into squeezing more performance out of, for example. 
Intel. I expect RISC-V will soon be on par.

[toc] | [prev] | [next] | [standalone]

#234679

In article <10ro892$10a46$3@dont-email.me>,
Peter Flass  <Peter@Iron-Spring.com> wrote:
>On 4/15/26 06:11, Dan Cross wrote:
>> 
>> Apple Silicon is pretty nice, though I don't know about ARM.
>> The obvious alternative is RISC-V, but it was not mature enough
>> for use in a high-end product at the time they moved away from
>> Intel.
>
> From what I read, RISC-V still isn't quite there yet. I think the 
>performance is still below other chip families, but, of course, a lot of 
>engineering went into squeezing more performance out of, for example. 
>Intel. I expect RISC-V will soon be on par.

Yeah, this is true.  Some new chips are on the horizon that look
promising, but the ecosystem hasn't had nearly enough time to
mature as, say, ARM has had.

	- Dan C.

[toc] | [prev] | [next] | [standalone]

#234674

cross@spitfire.i.gajendra.net (Dan Cross) writes:
>In article <10rmgl8$gjuj$1@dont-email.me>,
>Jason H  <jason_hindle@yahoo.com> wrote:
>>On 31/03/2026 21:59, Lawrence DOliveiro wrote:
>>>On 31 Mar 2026 19:58:11 GMT, Stefan Ram wrote:
>>>
>>>> On April 1, 1976, Apple Computer Company was founded
>>>>   (incorporated January 3, 1977).
>>>
>>>Back in those early years, the company was more about technological
>>>innovation than trendiness.
>>
>>To be fair to them, the whole move to UNIX and, much later, Apple Silicon
>>was ultimately pretty innovative when compared to the rest of the market.
>>Unified memory ftw. 
>
>
>Eh....  I dunno.
>
>Apple Silicon is pretty nice, though I don't know about ARM.
>The obvious alternative is RISC-V, but it was not mature enough
>for use in a high-end product at the time they moved away from
>Intel.

RISC-V didn't exist when Apple started working on the ARMv8-based
silicon (circa 2012-2013).   Apple was part of the TAB[*] (as was my
employer) and helped guide the development of 64-bit ARMv8.

[*] ARM's Technical Advisory Board

[toc] | [prev] | [next] | [standalone]

#234680

In article <tzNDR.276692$4wI6.186533@fx24.iad>,
Scott Lurndal <slp53@pacbell.net> wrote:
>cross@spitfire.i.gajendra.net (Dan Cross) writes:
>>In article <10rmgl8$gjuj$1@dont-email.me>,
>>Jason H  <jason_hindle@yahoo.com> wrote:
>>>On 31/03/2026 21:59, Lawrence DOliveiro wrote:
>>>>On 31 Mar 2026 19:58:11 GMT, Stefan Ram wrote:
>>>>
>>>>> On April 1, 1976, Apple Computer Company was founded
>>>>>   (incorporated January 3, 1977).
>>>>
>>>>Back in those early years, the company was more about technological
>>>>innovation than trendiness.
>>>
>>>To be fair to them, the whole move to UNIX and, much later, Apple Silicon
>>>was ultimately pretty innovative when compared to the rest of the market.
>>>Unified memory ftw. 
>>
>>
>>Eh....  I dunno.
>>
>>Apple Silicon is pretty nice, though I don't know about ARM.
>>The obvious alternative is RISC-V, but it was not mature enough
>>for use in a high-end product at the time they moved away from
>>Intel.
>
>RISC-V didn't exist when Apple started working on the ARMv8-based
>silicon (circa 2012-2013).   Apple was part of the TAB[*] (as was my
>employer) and helped guide the development of 64-bit ARMv8.
>
>[*] ARM's Technical Advisory Board

Adopting ARM on the desktop does make some sense from Apple's
perspective, I imagine.  They already have a robust ecosystem
for using it on handheld-devices; standardizing on one ISA
across their product lines is a no-brainer.

	- Dan C.

[toc] | [prev] | [next] | [standalone]

#234681

According to Dan Cross <cross@spitfire.i.gajendra.net>:
>Adopting ARM on the desktop does make some sense from Apple's
>perspective, I imagine.  They already have a robust ecosystem
>for using it on handheld-devices; standardizing on one ISA
>across their product lines is a no-brainer.

More than that, they already had extensive experience designing chips using ARM
components.  I gather the M series of chips in the macbooks are quite similar to
the A series in the phones and tablets.

The new Macbook Neo has an A18 Pro chip, with five GPU cores rather
than the 6 in the version they put in the phone. It was apparently a
way to use slightly defective chips with one broken GPU core they
could disable, thereby letting them sell chips they would otherwise
have thrown away.

The Neo has been so popular that they've run out of 5 core chips.  Oops.
They could ask TSMC to do another run of A16 but those would be way
more expensive than the rejects they've been using.

-- 
Regards,
John Levine, johnl@taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly

[toc] | [prev] | [next] | [standalone]

#234686

In article <10roha6$uuh$3@gal.iecc.com>, John Levine  <johnl@taugh.com> wrote:
>According to Dan Cross <cross@spitfire.i.gajendra.net>:
>>Adopting ARM on the desktop does make some sense from Apple's
>>perspective, I imagine.  They already have a robust ecosystem
>>for using it on handheld-devices; standardizing on one ISA
>>across their product lines is a no-brainer.
>
>More than that, they already had extensive experience designing chips using ARM
>components.  I gather the M series of chips in the macbooks are quite similar to
>the A series in the phones and tablets.

Exactly.  Given that the operating system distributions they use
for their various hardware products share significant amounts of
code, they had probably done most of the hard parts of porting
it to aarch64 already, anyway.

>The new Macbook Neo has an A18 Pro chip, with five GPU cores rather
>than the 6 in the version they put in the phone. It was apparently a
>way to use slightly defective chips with one broken GPU core they
>could disable, thereby letting them sell chips they would otherwise
>have thrown away.
>
>The Neo has been so popular that they've run out of 5 core chips.  Oops.
>They could ask TSMC to do another run of A16 but those would be way
>more expensive than the rejects they've been using.

Hah, that's awesome.

	- Dan C.

[toc] | [prev] | [next] | [standalone]

#234688

On 2026-04-15, John Levine <johnl@taugh.com> wrote:

> The new Macbook Neo has an A18 Pro chip, with five GPU cores rather
> than the 6 in the version they put in the phone. It was apparently a
> way to use slightly defective chips with one broken GPU core they
> could disable, thereby letting them sell chips they would otherwise
> have thrown away.
>
> The Neo has been so popular that they've run out of 5 core chips.  Oops.
> They could ask TSMC to do another run of A16 but those would be way
> more expensive than the rejects they've been using.

That reminds me of the one about the company that placed an order
for parts with a Japanese supplier, specifying a 1% defect rate.
When the shipment arrived, they found several parts in a plastic
bag lying on top of the remainder of the parts.  They queried
the supplier, who replied, "You specified a 1% defect rate.
For your convenience we packaged them separately."

-- 
/~\  Charlie Gibbs                  |  Growth for the sake of
\ /  <cgibbs@kltpzyxm.invalid>      |  growth is the ideology
 X   I'm really at ac.dekanfrus     |  of the cancer cell.
/ \  if you read it the right way.  |    -- Edward Abbey

[toc] | [prev] | [next] | [standalone]

#234689

John Levine <johnl@taugh.com> writes:
>According to Dan Cross <cross@spitfire.i.gajendra.net>:
>>Adopting ARM on the desktop does make some sense from Apple's
>>perspective, I imagine.  They already have a robust ecosystem
>>for using it on handheld-devices; standardizing on one ISA
>>across their product lines is a no-brainer.
>
>More than that, they already had extensive experience designing chips using ARM
>components.  I gather the M series of chips in the macbooks are quite similar to
>the A series in the phones and tablets.
>
>The new Macbook Neo has an A18 Pro chip, with five GPU cores rather
>than the 6 in the version they put in the phone. It was apparently a
>way to use slightly defective chips with one broken GPU core they
>could disable, thereby letting them sell chips they would otherwise
>have thrown away.
>
>The Neo has been so popular that they've run out of 5 core chips.  Oops.
>They could ask TSMC to do another run of A16 but those would be way
>more expensive than the rejects they've been using.

They can also use the 6-core version and fuse out one of the cores
with a small revenue hit.

[toc] | [prev] | [next] | [standalone]

#234691 — Re: pretty good chips, Today, 50 years ago

According to Scott Lurndal <slp53@pacbell.net>:
>>The Neo has been so popular that they've run out of 5 core chips.  Oops.
>>They could ask TSMC to do another run of A16 but those would be way
>>more expensive than the rejects they've been using.
>
>They can also use the 6-core version and fuse out one of the cores
>with a small revenue hit.

They're up to the A19 now.  The A16 was their 2022 chip that went
into the iPhone 14 and 15 series, which were discontinued in 2024.

I would be surprised if they had any of the 6 core chips left.



-- 
Regards,
John Levine, johnl@taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly

[toc] | [prev] | [next] | [standalone]

#234622

Someone has put together a montage
<https://www.instagram.com/reels/DWm2oTZCOcZ/> of various startup
sounds of Apple gear (desktop/laptop/server).

What’s missing, though? I don’t see any representatives of the
first-generation Power Macs (with the PowerPC 601 processor). These
had a “guitar strum” sound, which was generally considered to be
underwhelming. Which is why later models replaced it.

Also, there were the AV Macs (with the on-board DSP chip) from the
latter part of 1993. I think they had a similar sound to other
68K-based models from around the same time.

[toc] | [prev] | [standalone]

Page 2 of 2 — ← Prev page 1 [2]

Back to top | Article view | alt.folklore.computers

csiph-web