Hacker News new | past | comments | ask | show | jobs | submit | mben's comments login

Jay Miner was responsible for Atari 8 bit home computers and later for the Amiga.


Yup, the Amiga is more Atari than the later Ataris are.


I honestly like the STs better as computer designs. They are clean and general-purpose while the Amigas ended up being hurt by an overly complicated hardware that made it expensive for Commodore to keep up with PCs.


Absolutely. That was Tramiel and Shivji's intent, I believe. Rock bottom price general purpose computer. People comparing to the Amiga are missing the point. Tramiel didn't care much about games, and was interested in "computers for the masses, not the classes" -- bringing cheap 16/32 bit computer power to the market (and as fast as possible before the competition could).

The comparison point for the ST should be the IBM PC or the Apple Mac, not the Amiga. Productivity computers that were many multiples the cost. In 1985, 1986 there was nowhere else where you were going to get that much memory and clock rate for the price Atari was charging.

That plus the crisp paper white monochrome monitor (no dubious interlaced flicker), the Mac-like UI, PC compatible floppies. They had potential for a real winner, but he problem is that Atari failed to effectively iterate. At least not until it was too late. The Falcon was a great machine.


Don't forget the built-in MIDI.

The inverse color scheme? Not so great... which people are finally admitting 40 years later.


'twas easy enough to flip the palette


Not sure it was so, but it'd be interesting if the monochrome mode used two entries of the palette. Possibly also easier to implement as well.


That's how it was. Flipping to inverse as really easy and there were lots of auto programs or desk accessories that included an option for it, including I believe Atari's stock control panel.


More like basic and barren, missing the simplest/cheapest to implement quality of life features (scrolling = video start address offset register, pcm output). Add awful Engineering, signal integrity is marginal with many hacks to keep the house of cards from falling over. "Bad DMA" chip recall (would erase your hard drive on first access) was actually caused by using too fast DMA chip picking interference on the overloaded bus.

What hurt Commodore was technologically clueless management. Just one example - they didnt update Audio/Floppy controller chip for 9 years because original designer Glenn Keller left (maybe even fired, you never know with Commodore) in 1988, nobody else knew how to and Commodore didnt want to hire/pay competent people. As a result Commodore was shipping computers with 720KB drives all the way to 1994 while on PC 1.44MB was standard since 1988.


Yes. The ST was very basic and didn't include too much in the box. A lot of it needed a couple revisions to work well. Hardware scrolling is not a huge thing on a GUI machine - the Mac didn't have it either. The ST is not a computer built out of a gaming console (much like, BTW, the C64 - which just happened to be quite good for that). Later versions improved on the on-board audio and video and at least two were built around a VME bus (probably the only home computers to ever do that). Much as Commodore, Atari was also slow to iterate (and it had a much better starting point for that, but not as much money).

> What hurt Commodore was technologically clueless management.

That, too, but all the bus timings on the Amigas tied to NTSC (it sprung out of a gaming console, after all), the complexity of the custom chips (they'd all need to be iterated to keep up with new Motorola processors AND what was being offered in PC clones), resulting in multiple buses, fast/chip RAM, and so on. The Amigas were incredibly capable for their time, but were not elegant computers and did not evolve, in part, because it'd be too costly for Commodore to do so. A lot of the features of the Amiga made sense in games (planar graphics), but made none for a general purpose computer. What's obvious now (hindsight is always 20-20) is that they should have segmented the line, have a gaming platform at the low end, and a general purpose computer, running the same OS but with things like VGA ports and preemptive multitasking with memory protection and virtual memory. And they SHOULD have taken the Sun proposal of selling 3000's as low-end UNIX machines seriously.

As for the floppy, I have no idea why they didn't just adopt a market standard FDC and let the original's floppy side rot. The audio part was quite impressive but I doubt anyone was using the FDC side without the OS in the middle.

Another thing that always surprises me is that both made PCs with little or no commonality with their "proprietary" lines. Using the same chassis for Amigas, STs and their PCs would save money, as well as putting Amiga and ST keyboards on PCs (this was all before the PC-101 layout).


All good points. But I think Commodore was essentially technically rudderless with the Amiga after Miner and related folks were no longer involved, so I don't think they'd really have had anything to offer on either end (game/consumer or general purpose workstation class machine.) Or even on the software front, who would have even developed the software on the latter? Like Atari, I get the sense that they just didn't have the resources to make that kind of thing happen.

Atari made moves in 90, 91 to get more serious about their OS development, hiring Eric Smith fulltime to work on MultiTOS, etc. Smart move, but too late.

The low-end Unix workstation thing was a meme that just didn't work out in practice. Both Atari (with the TT running SysV) and Commodore tried at this market and failed. There was no buyer for it. I still remember the snarky little headline on a snippet in UnixWorld about Atari's TT & Unix: "Up from toyland."


Sure, I'm not evaluating their quality. Just pointing out design lineage.



1090+ clicks from that one comment alone ... the internet is a pretty wild place.


Check pf-diverters [1]. We are using them in our openbsd firewalls in order to block unwanted connections.

[1] https://github.com/echothrust/pf-diverters


Yes, iRedMail (http://www.iredmail.org/) on Digital Ocean. The $10/mo VM is enough for a low traffic mail server.


Theo's opinion about BoringSSL and LibreSSL portability: http://article.gmane.org/gmane.os.openbsd.tech/37174


NetBSD has hired a developer to port Linux DRM to NetBSD with support for KMS and GEM: http://mail-index.netbsd.org/tech-x11/2013/02/06/msg001229.h...


PFSense is using an older and patched version of OpenBSD PF and as you said all the configuration is done using the gui. PF configuration in OpenBSD is very simple using the cli and following the official FAQ: http://www.openbsd.org/faq/pf/index.html


Check also the Unix Toolbox: http://cb.vu/unixtoolbox.xhtml I've got a copy of the pdf in my dropbox just in case.


Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: