* Posts by ThomH

2913 publicly visible posts • joined 18 Jun 2009

The Stonehenge of PC design, Xerox Alto, appeared 50 years ago this month

ThomH

Re: nope, Smalltalk80 was The Bomb..

> Anyone who uses MI really does not know what they are doing. Or writing toy code.

Or using pure abstract base classes as a surrogate for protocols.

Enter Tinker: Asus pulls out RISC-V board it hopes trumps Raspberry PI

ThomH

Re: Wrong

I think the Richard M Stallmans of the world are very interested. And possibly curious developers, especially those in the bare metal niche.

But no, these are not substantial groups.

Adidas grapples with $1.3B in unsold Yeezy sneakers after breaking up with Kanye West

ThomH

Re: Doing a Musk

To be clear, Adams' most direct comment was "the best advice I would give to white people is to get the hell away from Black people. Just get the f*** away. Wherever you have to go, just get away.", i.e. he advocated in favour of segregation, treating all black people and all white people as homogenous groups.

I don't need to know who agrees or disagrees with that to have a strongly negative opinion about it.

AmigaOS 3.2.2 released for those feeling nostalgic

ThomH

The ST was substantially cheaper for most of its lifespan; e.g. at launch in 1985 an 8Mhz 512kb Atari ST with a colour monitor (the more expensive option) was $1000, compared with $1,295 for a 7.14Mhz 256kb Amiga with no monitor. Leading to the ST being the world's first computer to offer 1mb for less than $1000.

Around 1990 when the difference was only about £100 it's not so important, admittedly.

Can we interest you in a $10 pocket calculator powered by Android 9?

ThomH

Re: Ok... I have to ask

With the landfill Android factories pumping out devices that can be sold for $10, what would justify the effort of somebody designing a board without Wifi?

Linux app depot Flathub may offer paid-for software

ThomH

Indeed, it does rather suggest that it would be logically impossible to plead guilty to a dishonesty offence.

Twitter algorithm to be open sourced 'next week,' says Musk

ThomH

Spoiler

Tweets recommendations() { return find(.author = "Elon Musk"); }

What Brit watchdog redacted: Google gives Apple cut of Chrome iOS search revenue

ThomH

Re: Flawed logic

It'd be about as successful as iTunes Ping, while probably consuming more than the staff levels still assigned to macOS; if anything I'm surprised that Microsoft has clung on for so long.

Generative AI is out of control: Nothing, Forever is a Seinfeld spoof about nothing... forever

ThomH

Re: BIg Bang Theory

MASH is the worst to try to watch here in my US sojourn; a huge laugh track over almost everything. I'm pretty sure it showed in the UK without a laugh track.

Bringing the first native OS for Arm back from the brink

ThomH

Re: 64-bit port

Tediously: the 1984 Macintosh did file types and application associations as resource fork key/value pairs, allowing per-file associations and not putting any trace of the type into the file’s name. I doubt it was the first, and — to be overt — RISC OS was way ahead in uncountably many other areas.

A brand new Linux DRM display driver – for a 1992 computer

ThomH

Re: Falcon, ST, AmigaOS et al

> What that processor lacked, and what you’re maybe thinking of, was an MMU.

Most precisely, the stack frames it produces upon a bus error aren’t sufficient to know how complete an instruction was when it threw, and no support is offered for restarting a half-complete instruction.

So you can’t bolt on an MMU in a general-purpose fashion because any instruction that causes a page fault will have done an unknowable amount of its operation — and side effects — before throwing but can only be restarted from the beginning.

ThomH

Re: Falcon, ST, AmigaOS et al

> There would have been no DMA to free the CPU while the floppy drive was being read though

The floppy drive and ACSI (i.e. proto-SCSI) port are the two things other than video in an ST that have a DMA interface to RAM.

The ST doesn’t even then have to do any further decoding as it receives decoded original bytes, whereas the Amiga receives an MFM stream and uses the Blitter further to decode.

ThomH

Re: Good.

640x200 at 2bpp, 320x200 at 4bpp are both a step up from the 8-bit machines; fairly predictably for machines sold on the size of their databus, they’re approximately double the bandwidth of anything on an 8-bit micro at the time.

E.g. the C64’s 320px mode is attribute based, much like a ZX Spectrum, and the CPC’s is 2bpp.

The MSX 2 and Master System both muddy the water… but both postdate the ST.

GitLab versus The Zombie Repos: An old plot needs a new twist

ThomH

I’m on GitHub rather than Lab but to offer an example from further on the spectrum: amongst my set, I have one repository that is multiple gigabytes in size.

It’s test cases, in volume, spelt out in JSON as insurance against bit rot. That’s even with a decent portion of them being GZipped prior to addition to the repository.

It’s popular with a decent subset of people, but we’re talking dozens only. Would almost certainly fail an objective cost-benefit analysis.

Windows Subsystem for Android declared ready for prime time

ThomH

Re: WSA or ASW?

Windows Subsystem for Android = an adaptation of Android that works on Windows.

Windows 11 = an adaptation of the number 11 that works on Windows.

Microsoft Office = the version of Microsoft written by Office.

“640kb ought to be enough for anybody” = an apocryphal claim that 640kb will be happy no matter who uses it.

iPhone 14 car crash detection triggered by roller coasters

ThomH

Re: A million hours

Maybe they paired up with Tesla’s self-driving team?

ThomH

Re: GPS ?

if (in_a_theme_park) goto fail; goto fail;

Linus Torvalds's faulty memory (RAM, not wetware) slows kernel development

ThomH

Re: Macbook?

but the new Apple Silicon Macbooks are actually hostile to Linux and only support MacOS.

Meanwhile, back in reality, the Asahi Linux FAQ states:

Apple allows booting unsigned/custom kernels on Apple Silicon Macs without a jailbreak! This isn’t a hack or an omission, but an actual feature that Apple built into these devices. That means that, unlike iOS devices, Apple does not intend to lock down what OS you can use on Macs (though they probably won’t help with the development).

So I guess we're defining "actually hostile" to mean "won't help with the development", and "only support MacOS" to mean "Apple allows booting unsigned/custom kernels"?

Elsewhere from the FAQ, to get ahead of the topic:

Apple still controls the boot process and, for example, the firmware that runs on the Secure Enclave Processor. However, no modern device is “fully open” - no usable computer exists today with completely open software and hardware ...mainstream x86 platforms are arguably more intrusive because the proprietary UEFI firmware is allowed to steal the main CPU from the OS at any time via SMM interrupts, which is not the case on Apple Silicon Macs. This has real performance/stability implications; it’s not just a philosophical issue.

ThomH

Re: Take This With a Grain of Salt

> 640K -- all Intel could easily access -- was a lot in the early 1980s.

*cough* 'Intel' (i.e. 16-bit x86) can easily access 1mb of address space.

The decision to make only 640kb of that general-purpose RAM was IBM's; the rest is reserved for video card ROM and RAM, the BIOS ROM, etc.

Is it time to retire C and C++ for Rust in new programs?

ThomH

Re: C++ and memory safety

I managed one the other day by parallelising some unit tests, which are completely independent and were running without hiccup serially.

The reason, as I eventually discovered? One of the functions on the line declares a local variable that ends up being ~480kb; meanwhile the particular parallel for on my particular OS provides a much smaller stack to things dispatched in that manner.

In practice one of the threads was stomping on the memory area of the next before it had started up, so the actual crash occurred when a thread I don't own mysteriously appeared to start up with a single-item stack trace, being a crash inside an object with this = 0.

So, ummm, either I'm arguing that it is possible to hit subtle memory access bugs even with modern RAII code, or I'm revealing that I am a 'bad' programmer.

Apple co-founder Steve Jobs memorialized with online archive of emails, guff

ThomH

Re: Elon Musk told GQ, "The one time I met Steve Jobs, he was kind of a jerk."

Luckily I think it only takes six to know eight, so the chain attenuates.

Apple debuts iPhone 14, Watch 8, other sparkly things

ThomH

Re: Mini is cancelled

The iPhone 13 Mini is still on sale, now at a reduced price, and with the same processor as the iPhone 14. You're to make do with only two-thirds as much RAM and a fractionally-worse camera though.

That said, the iPhones were at 4gb RAM for three years before this year's increase to 6gb so you should probably also be wary of that being the arbitrary cut-off for some version of iOS at some point.

ThomH

Re: Dynamic Island*

It's close enough to that, but the area of non-screen also seems to have touch sensors underneath it, so it can now act as a real widget to some limited extent. Or, at least, the fact that it's hidden won't come back to bite you when you accidentally tap on the wrong spot.

Zuckerberg: Yes, Facebook kept Hunter Biden's laptop under wraps

ThomH

Re: 'Take that, you evil Trump fan'

I was responding to the conversation as already digressed.

That said, if there’s one thing I thought the left was against then it’s undue power accumulating in private hands so for me it is a concern that Twitter, Facebook et al can go so far in deciding what propagates and what doesn’t, without any sort of oversight.

I don’t agree with you about the net bias, but that’s orthogonal.

ThomH

Re: 'Take that, you evil Trump fan'

> > Sigh. Sure, anyone could have made those deals. It has of course nothing to do with gaining access to power - also called soft corruption.

You seem to be accusing Hunter Biden of actions that, although deeply unethical, weren't illegal — then they'd be corruption, not soft corruption? So you're arguing:

1. Hunter Biden used his surname to obtain access to deals that he would not otherwise have been able to achieve; and

2. because the FBI (or the CIA, or whomever) didn't investigate something that isn't illegal, they must be part of a shadowy unelected cabal that undercuts whichever person the US people actually elect?

Treating your allegations as if they were true:

I don't see the logic, and I also don't see why that would be a partisan issue. I'd also have expected fans of small government to be happy if agencies don't spend time and money on things that aren't crimes.

Alternatively, if it were a partisan issue then I'd be curious as to why the FBI reopened its investigation into Clinton's illegal mishandling of official communications immediately before the 2016 election, but is failing to act now. Did four years of Trump turn the deep state from red to blue?

So, yeah: nonsensical conspiracy theory.

ThomH

Re: Deep state

Does anyone really believe that the US three-letter-agencies were *not* in the know regarding Trump's predilection for wearing a scarf as socks and running around the White House shouting "Here comes Professor Lancaster!!!"?

We're just stating random non-illegal things as facts, regardless of evidence, right?

ThomH

Re: Why do people ignore facts?

> There's far more evidence of Joe Biden running an expensive pay-for-play scheme than what was found on Hunter's laptop. Hunter's laptop has never been refuted

From the school of "if Fox, OAN, and my crazy uncle on Facebook repeat it enough to one another, it becomes plenty of evidence".

ThomH

Re: Narrative checker

I think this is why Google has sought official permission for GMail not to filter any sort of political soliciting, regardless of content — the only way to be perceived as unbiased is to keep entirely out of the filtering game. At least around election time.

ThomH

Re: "Hunter Bidens Laptop"

It was a MacBook, which I think for many round here would be the main offence.

ThomH

Why does adobob ignore facts?

> Hunter Biden's laptop shows that his father was setting up his son jobs in China, Ukraine, US, etc, and taking massive kickbacks as a result of that.

No, it doesn't; it shows that, at some point before October 2020, files alleging that were stored on a laptop that Hunter Biden had until April 2019. The agent of delivery was Rudy Giuliani, whom the New York State Bar Association is currently evaluating for disbarment based on his willingness to lie and mislead.

> I don't care if you don't like Trump

It's pretty clear that you do.

ThomH

Re: "Hunter Bidens Laptop"

Yeah, it was an attempt to recapture the magic of 'But Her Emails!' for 2020. Mishandling of official communication matters a lot, unless it's a Republican doing it.

In all probability they'll be back in charge of the House as of next January so expect endless investigations and committees on Hunter Biden's laptop, along with plenty of hearings in which the committee members take turns to give twenty-minute non-fact-checked discourses with a nominal question mark at the end.

Big Tech is building the metaverse of its own dreams. You don't want to go there

ThomH

Re: E=World

The Atari ST was 1985. The Apple II, the first home computer with colour graphics and sound, was 1977.

The Atari ST offered the first colour desktop GUI, the first < $1000 1mb machine and a bunch of other breakthroughs. Apple didn't offer a colour desktop until the IIgs in 1986, and didn't offer a decent colour desktop until the Mac II in 1987. It continued selling purely-monochrome machines at the entry level until the end of 1990.

Fun bonus observation: the Atari ST was designed by a team led by Shiraz Shivji, who had previously worked on the Commodore 64, another famous colour-and-audio machine.

But objectively, here in reality, whether you like the company or not: Apple was first with colour graphics and with audio.

[if you can forgive that both were painful, being artefact colour and software toggling of a speaker position]

ThomH

Re: E=World

> Apple have never done anything first.

Do you mean other than colour and sound on a home computer?

Or, more likely, do you mean: since 1977?

Warning: Apple 'could very easily' cripple Jamf

ThomH

Re: An ostensibly nonpartisan left-wing advocacy organization

In ye standard political aside: do you coincidentally mean more-or-less since the Reagan-era abolition of the FCC fairness doctrine in 1987?

Apple to compel workers to spend '3 days a week' in the office

ThomH

Luckily for Apple the other broadly-similar employers such as Google, Meta, etc, are either reducing or completely freezing hiring in the short term.

I'm full remote these days, and I ideally wouldn't ever go back. It's doing wonders for my child time — both being with one and acting like one.

Google tells Apple to 'fix text messaging' in bid to promote RCS protocol

ThomH

Re: "Email only recently (2008-ish?) came to mobile telephony"

Even before that there were WAP front-ends to email which somebody, somewhere, probably used once, at a rate of 38p/megabyte.

ThomH

Re: they do not support modern texting features ...

You mean for all those videos that are less than about 10mb in size?

ThomH

Re: they do not support modern texting features ...

I sometimes send video of my little chap to other family members, an ocean away; not having that reduced to MMS quality is a win, without having to find a file or video sharing intermediary.

Parallels increases prices with Desktop version 18

ThomH

VMWare Fusion Player is also free for personal use, though you have to register first. It's my preferred choice.

Google's ChromeOS Flex turned my old MacBook into new frustrations

ThomH

I had a first-generation MacBook Pro with its Core Duo, and used it up until probably about 2010; it is unfortunate that Apple needed to jump ship during Intel's temporary regression to a 32-bit ISA but at least no Apple laptop ever saw a single-core x86. The original low-end Mac Mini was a Core Solo.

The PowerPC was so far behind at that point that I doubt Apple could have afforded to wait much longer, but at least SSE3 was present right out of the gate, providing a smooth transition for things like the Accelerate framework.

ThomH

Then you clearly didn't buy the original iPad, which is the semi-recent Apple device with the shortest useful lifetime.

Released in March 2010, support was dropped for it in iOS 6 of September 2012, and running 2011's iOS 5 on it leads to an incredibly slow device.

The iPad 2 of 2011 improves on the original with twice the RAM and twice the bandwidth, two processor cores instead of one and a greater-than-33% improvement per core, and what Apple claimed was eight times the GPU power though I can't still find a benchmark on that.

So I guess Apple just figured they'd brush the original under the carpet?

(Though I'm only now about to retire my 2015 iPhone 6s and Retina MacBook; the first-generation iPad is the aberration)

Apple-1 prototype hand-soldered by Woz up for auction, bids expected to reach $500k

ThomH

Re: Open source?

> In one final gasp, the Apple II supporters at Apple designed the Apple IIGS Plus, code named "Mark Twain". It had an 8Mhz 65C816, a built in SuperDrive, 2MB on the motherboard, and a hard drive. Prototypes leaked out and a user group that has one and wrote a series of articles about it. Apple management vetoed this unit.

The Mark Twain was designed in 1991, a full five years after the release and therefore likely six or more after the design of the original IIgs.

Nevertheless, per the only owner of one:

> Despite rumors that the Mark Twain is a speed demon, a standard 65C816-4 CPU running at 2.8 Mhz is found in the same physical location as it is on the other IIGS models.

ThomH

Re: Open source?

> Selective for good reason - those things you mention weren't common AT ALL back in those days on competitor systems, regardless of how good or bad the other capabilities of said systems were. So criticising the IIgs for daring to suggest that it had graphics capabilities, just because it didn't include a bunch of barely seen elsewhere hardware to help mitigate its relatively slow memory bandwidth/CPU, felt like a bit of an unfair dig at the system.

On the contrary: the issue is that the extreme and unusually-severe bottleneck to write to video memory isn't otherwise alleviated. Trying to talk about half of that equation while ignoring the other is absurd.

ThomH

Re: Open source?

Yeah, you've misunderstood the source.

The mechanism described uses banks $00 and $01 as a write-through cache for banks $e0 and $e1. Writes still occur at 1Mhz.

The statement "there is a solution which allows the screen to be written to at near full speed." is false.

Not a surprise you've taken the positions you have given your lack of familiarity with the machine.

ThomH

Re: Open source?

That's a selective quote; the point is that the CPU has to do the work, but can't do that efficiently because of the 1MHz bus.

If the question were which other machines don't offer hardware and also offer such speed-limited memory access then the answer would be 'none'.

ThomH

Re: Open source?

> Writing odd values to odd addresses to page memory is par-for-the-course on 8-bit machines, the Commodore 64's memory map was probably just as crazy.

As per above, this isn't about writing odd values to odd addresses. It's about triggering a bunch of soft switches — which are flipped one way or the other by access to a single address — with each individual chunk of memory having two or three different reasons it might be pointing at one thing rather than another.

If you had programmed any other 8-bit machine you'd know this is not the norm. Every other machine uses: (i) regularly-sized memory areas; (ii) with a single point of control.

If you're a fan of everything being possible in two or three ways, each method being a particular combination of two or three other discrete inputs then, yeah, there's a lot to like in the world of Woz.

> The main thing that killed the IIgs was Apple deciding to artificially limit the speed of it to 2.8Mhz so it wouldn't compete with the Mac or any other 16-bit computer of the time.

The speed-limiting factor in an Apple IIgs is the available speed of RAM.

The 65816, like the 6502 before it, requires RAM accesses be completed in a single cycle.

The 68000 provides at least four cycles for any RAM access to be complete.

There is a conspiracy theory, which often nonsensically fingers Steve Jobs himself from then beyond the Apple grave, that the clock was limited, but basic facts seem to disagree.

See also the Apple IIc+, which Apple launched two years after the IIgs and which ran at a nominal 8Mhz. It achieved that by using a cache between processor and bus. And Apple achieved that by licensing the cache from a third party.

Even if they had magical superfast consumer RAM, to me it seems improbable that Apple could produce a faster Apple II machine in 1986, but had forgotten by 1988 and had to pay somebody else to help out.

> Woz I doubt was responsible for making the IIgs artificially slow

Right, because nobody was.

ThomH

Re: Open source?

Woz is massively overrated; have you ever tried programming for a system he designed?

The rule for whether base or auxiliary RAM is visible in the region $2000–$3FFF on an Apple IIe is that it'll be the latter if: (i) the 80STORE, PAGE2 and HIRES soft switches are all set; or (ii) the 80STORE switch is reset but the RAMRD switch is set. Otherwise it'll contain base RAM.

There are similar rules pertaining to the other nine irregularly-sized memory regions that fill its 64kb of address space.

Think that's just a corner the team was backed into by a desire for compatibility with the existing ROMs? Then you're wrong as per the IIgs, which amongst other things introduces shadowing (using a fast part of RAM as a write-through cache for a slow part, essentially) and invents another set of arcane rules about what situations lead to shadowing of what.

Probably its best feature is having 'graphics' in the name while offering nothing beyond a plain single-buffered frame buffer (no hardware scrolling, no blitting, nothing), and putting that behind a 1Mhz 8-bit bus.

With the IIgs Woz achieved what Apple hadn't otherwise managed: to kill the Apple II.

Apple to pay $50m settlement for rotten butterfly keyboards

ThomH

Re: Staingate et al

Most Apple fans will receive completely reliable MacBooks, as the statistical failure rate is a tiny slither of the whole. But that doesn't mean that Apple isn't culpable for some of the flaws, as here, where the issue could clearly have been avoided.

Me personally? In twenty-ish years I've had a MacBook Pro with a GPU fault that happened late enough as not to really matter and a MacBook with a logic board that has failed twice, the first time after only about three years so during what should be a machine's normal lifetime.

That's a pretty high subjective failure rate, but I'm damned if I can find any objective statistics so there's no reason to believe me any more than any of those other commenters you refer to.

CP/M's open-source status clarified after 21 years

ThomH

Re: The title is no longer required.

Probably skip the +3; the CPC can display real 8px/character 80-column text whereas for CP/M the Spectrum uses some weird 5- or 6-pixel-wide font, displaying only a portion of the width of the virtual 80-column display and jumping between the left and right portions of the screen ‘intelligently’.