Gail Wellington, former Commodore executive, has died | Hacker News Reader

Gail Wellington, former Commodore executive, has died

legacy.com

106 pointserickhill50 comments

Open articleView on HN

Further info:

https://cdtvland.com/2025/05/16/in-memory-of-gail-wellington...

https://commodore.international/2021/11/21/gail-wellington-f...

https://www.youtube.com/watch?v=QBP257fGu8Q

I read the linked page and googled here name. I found a very detailed and intesting page ( to me ) at

Commodore International Historical Society Link below )

Gail Wellington: far more than just a herder of CATS and ... https://commodore.international/2021/11/21/gail-wellington-f...

Looks to be an excellent page, excellent information about Commodore computer history too.

>The CDTV was a brilliant product for its time.

It was a 1985 computer selling $300 retail in 1991 packaged with $300 retail CD-ROM. Commodore got the crazy idea to try and sell this "the whole is less than the sum of its parts" at $1000 because black case and remote control! Nobody got fooled. Zero effort went into trying to cost optimize it, or even make it a desirable product. It was as ridiculous as Philips mega flop CD-I shipping similarly bad internals at same price point.

I think there was at least one iteration of the CDTV with somewhat lower-cost internals though.

Never seen a CDTV-CR in real. I have an original with keyboard, mouse, diskdrive and remote. I did like the device. Somehow the CD playback sounded better than my Sony CD player, but very clunky with the caddy and somewhat unintuitive interface.

I think that was after Gail Wellington was fired, and didnt exit prototype stage.

Unfortunately I concur with this assessment. Commodore was too busy phoning it in towards the end and effectively wasting the talents of Gail, as well as others in engineering, like Dave Haynie.

Obviously they lost focus if they laid Gail off.

I feel like they just wanted to coast on the previous success, and not having to put more capital in to bring the platform back up to the lead.

Sad. I was an Amiga user from roughly 1989 through 1994. Commodore barely updated the Amiga platform for most of its life. The major updates, like AGA, were too little, too late.

One of my biggest regrets in my "journey" with computers is walking away from Commodore when Amiga was released. I felt it was superior to anything else I had seen, and I knew the Commodore 64 inside-out, but I just felt like PCs were for grown ups, and I needed to grow up. I needed to skate to where the puck was going, and that wasn't Commodore. I guess I was right, but I still regret it because all the dudes I knew that stuck with Commodore, with the new Amiga platform - well, they all seemed to be having more fun. I learned macros for Lotus 1-2-3, which was more practical (I made $$$ as a teenager teaching stiffs about macros), but my Amiga friends were making cool drum beats and sample-based music and remixes on their Amigas - totally impractical, but also fun as shit back then. So yeah, they were all having fun with their Amigas, while I became Alex P. Keaton.

I stayed with Commodore, but with my 128, since I couldn't afford an Amiga. I envied my PC friends their VGA graphics, but yeah, the Commodores were a lot of fun. And I learned to program on it, which none of my friends were doing with their PCs. Once they went to PC, it seemed like they just used them to run software, and weren't interested in tinkering.

For me anyways: Amiga was learn what a computer can do, PC was learn how a computer works. PC was also my path to linux and linux changed my circumstances in life drastically. I get what the comment above you is saying, but I don't think the Amiga was more fun at the end of the day, PC ended up being much more challenging as a power user (therefore fun).

I still miss features from AmigaOS, to the point I've started adding some to my software. E.g. I've added "assigns" to my shell (very superficial integration), and it's so nice to be able to just create an alias for any path.

It's hard to explain why it feels better, because you could "just" create a bunch of symlinks in one location, but being able to "cd projects:" instead of ~/Desktop/Projects just feels nicer. I've also added "implicit cd" when entering paths, so I can also do just "projects:" like on the Amiga.

What made the Amiga special wasn't just the hardware, but a whole host of small extra niceties like that.

Another favorite is datatypes: A uniform plugin-based API to open files of any type, ensuring Amiga software written decades ago can open modern image formats.

And another one I love was the ability to open a new console window as a file path, so you could redirect shell output to a new window just as if it was a file.

(it also took way too long before I remembered who Alex P. Keaton was...)

I also enjoyed these features! I used to have "downloads:" for all my BBS downloads, "docs:" for my papers, etc. The Amiga OS had some other interesting features, like the ability to attach comments to files as metadata ("filenote" command.)

I use a script to create a dotfile with comments about a file, and an ls wrapper that outputs the comments if present for that reason as well.

My setup is very slowly converging on getting back some of the things I enjoyed about AmigaOS... The challenge of course is that simulating it without rewriting every piece of software and/or replacing the kernel will always be a leaky abstraction, but I can at least do it for the basic desktop and terminal experience...

> The major updates, like AGA, were too little, too late.

And AGA was a mixed bag. The extra bitplanes were really welcome, but not having chunky (1 byte per pixel) mode when all the 3d coming out really required it, and having to do an expensive operation to go from chunky to planar, did really hurt efficiency.

It was a great addition that extended the existing idea of bitplanes, which was a really good one in lots of ways though.

The problem with AGA is it arrived when 386 clones with SVGA were becoming incredibly affordable. If it had arrived in early 1990 instead of late 1992, the Amiga might've had more of a chance.

Basically, Commodore should've skipped ECS entirely. ECS was essentially useless to most consumers.

Actually, Commodore should have kept their chips up-to-date with process improvements. That means releasing a new chipset with four times the performance every three or four years. So the AGA should have arrived in 1988, and the blitter should have been four times faster as original, not just twice as fast.

The Amiga 2000 should have been delayed until 1998 to include the new AGA. Keeping the cheaper Amiga 500 on the OCS would have been fine.

I do like your alternate reality better than mine!

I disagree that it needed a chunky pixel mode most of all. What you're asking for is a machine that can draw a scene byte by byte with the CPU, then just display that. But if your hypothetical Amiga is doing most of its graphical manipulation with the CPU, then it has failed as a platform. The main idea of the Amiga is to handle media data using specialized chips that are much faster than the CPU for certain tasks.

What an upgraded Amiga really needed was two things. The first is a fast blitter that could also horizontally stretch or shrink a bitmap by some fractional amount. The second was some sort of “flipper” device (or new blitter feature) that could reflect a bitmap across a diagonal line (or rotate by 90 degrees).

Here's how you'd use these for a third-person shooter. Store the wall bitmaps flipped along the diagonal; each line of those bitmaps correspond to a vertical slice of the wall. For each vertical line in the scene, find the correct wall tile and row, and blit that line of pixels into a scratch space, squishing it and shifting it by the correct amount. Then use the flipper to copy that to the screen.

Yeah, as you mentioned before, the chipset just didn't keep up. I had an A3000 for a while (68030 25mhz) You could do faster blits with the CPU. I remember running a utility called "cpublit" that accelerated window scrolling. It was noticeably faster. This was 1991 or so. ECS was long in the tooth.

But still, 3 years later, the Playstation did the same thing and it was a fantastic success.

CDTV was simply a bit early (hence the price) and a bit confused about what it wanted to be. It cost like a development machine but it was a fundamentally end-user one; it provided continuity for Amiga developers but only a hard reset for Amiga users. It also debuted in harsh economic times.

Commodore wanted $1000 for a 1985 computer combined with single speed CDROM barely able to play 160×100@12Hz CDXL video and called it multimedia! It cost less to just buy external SCSI CDROM and SCSI controller + A500.

Playstation was a $350/$299 toy with a computing power delivering 3D experience only available in top end Arcades a year prior (Namco System 22). 60fps 3D gaming plus 320x240@30Hz ~MJPEG video playback. You couldnt buy anything with similar 3d power for a couple of years, closest would be $3000 top of the line 1997 PC (P200, dx5/glide).

Can't help being a little nitpicky, but the CDTV was an A500, not A1000. So it was 1987 tech inside a stereo box + CD player, not 1985.

But at the end of the day it really didn't matter. I remember seeing a CDTV at a store and it wasn't even in the computer section. It was in the stereo department sitting in the middle of the sales floor in just a stack of boxes. You had to know what it was all about when you saw it. The sales guys didn't know what to do with it, sadly.

>A500, not A1000

Not a single technical improvement, hardly any non cost reduction difference between them. Would be like arguing breadbin vs C64c are different computers so cant call C64c 1982 tech.

The A500 came with a Kickstart ROM. The A1K required a floppy disk. The 500 could also use a 1MB Agnus chip and one could easily update trapdoor RAM. It wasn't light-years advanced over the A1K but it was a good step in the right direction and vastly more affordable. Hence why it was the best-selling Amiga and the A1K "did OK" but didn't set any sales records. "Not a single technical improvement" is ridiculous.

Yes, the A500's main "innovation" was cost reduction.

Well they released the CD32 within a year or two of the PS1 launching. It was a patent lawsuit that took them down but an AGA 68020 console with a library of A500 ports was never going to make it either.

Had a number of things gone differently Commodore might have been Apple. Both the C64 and the Amiga were way ahead of their time both in terms of raw performance and, for the C64, price/performance at least when it first came out. I learned to program on a C64 and still fondly remember it as an amazing gateway machine into computing. Was great for games too, better than most consoles of the day.

Unfortunately the 64, like all those 8-bit machines, was a technical dead end, and by the time the Amiga got momentum PC clones were eating the entire industry. PC clones killed everything but Apple, which barely clung to life through the 90s, and some Unix workstations in the high end market. It just wasn’t possible to compete with the price cuts and CPU performance gains that came with volume and scaling.

(I remember in the early 90s a lot of doubts about whether x86 could be made as fast as Sparc or Alpha or other things, but Intel and later AMD did it… especially when it came to price/performance.)

In retrospect Amiga might have competed there had it gone higher end and been a Unix-like OS underneath.

> (I remember in the early 90s a lot of doubts about whether x86 could be made as fast as Sparc or Alpha or other things, but Intel and later AMD did it… especially when it came to price/performance.)

And now Apple is in the process of beating x86 to a pulp with ARM.

Going full circle using a team with Alpha heritage.

The thing that made x86 take over the world was the immense scale and volume it got from the PC clone and small PC-based server (NT and early Linux) market. It allowed huge R&D investments and let x86 eat more expensive workstation and server CPUs from the bottom up, eventually conquering the server and then cloud.

ARM got the same thing from the mobile phone market, allowing it to eat x86 from the bottom up in exactly the same way. It went from phones and IoT, to laptops and small servers, and now to large cloud servers.

ARM does have a small advantage from its ISA being much easier to decode, especially with a wide decoder, but x86 engineers have shown that they can stay competitive.

> Unfortunately the 64, like all those 8-bit machines, was a technical dead end

I suspect that most of them were dead-ends for business reasons rather than technical reasons. More often than not, they were in a race to the bottom to try to sell to home users. Compatibility between generations, when it did exist, was often an afterthought. Commodore owning MOS probably didn't help either. A 16-bit successor to the 6502 was eventually introduced, in the same year as the 386, and the 65816 didn't even come from Commodore/MOS. I'm not sure what the story behind the other popular 8-bit processor is, i.e. the Z80. It has a 16-bit successor in 1979, but I've rarely heard of it's use.

> Had a number of things gone differently Commodore might have been Apple.

As a die-hard Amiga owner and fan, I've pondered this for many years but have come to the sad conclusion the mythical "handful of things" which, if done differently, would have meaningfully changed the fate of the Amiga simply never existed. The key reason is that many of the same things which made the Amiga so extraordinary in 1985, like custom graphics co-processors and designing the system around analog video standards also made the Amiga uniquely problematic to migrate to a RISC architecture like Power PC in the 90s. At the end, Commodore's unannounced (but in progress) plan was to migrate the Amiga platform to an HP RISC CPU, abandon analog video timing for 1280 x 960 digital RGB displays and run old Amiga software via emulation. I'm highly skeptical any such emulation would have worked sufficiently well to be considered "an Amiga." Even today, the authors of the incredibly advanced WinUAE Amiga emulator are still discovering subtle timing issues after 20 years of development. If the "new Amiga" platform couldn't run at least 90% of 68K Amiga games at 100% speed then it's not really an Amiga and instead a new and different computer platform. And if it's pretty much a new platform anyway, why not consider the PC or Mac?

Apple only barely managed to survive the 90s transition and the Mac was nearly a 'plain vanilla' platform (no custom co-processors, no brilliant analog timing hacks, etc) so it was much easier to emulate on a RISC platform than the Amiga would have been. In 1993 a 68020-based Amiga could emulate a 68000-based Mac entirely in software and run PageMaker at nearly 100% speed, whereas a 1992 68020 Mac emulating an Amiga in software would have been unimaginable.

So, we're left with a Catch 22. If we go back in time to the 80s and change the things about the Amiga platform necessary for it to survive the 90s industry transitions, we're also undoing the things that made the Amiga so special when it launched. Surviving the apocalypse of the 90s required surviving three separate 'extinction-level' asteroid impacts: 1. CISC -> RISC, 2. Analog video -> Digital video, and 3. Vertically integrated computer company -> Ecosystem of component vendors (mid-90s PCs had multiple vendors competing to be the CPU, operating system, graphics card and sound card). Neither Commodore nor Atari could have survived all three of those 'planet killers'. Even Apple WITH Steve Jobs, more money, a bigger, better brand and an emergency bankruptcy-preventing $150M loan from Bill Gates - barely survived. The first asteroid also took Motorola themselves out of CPUs (even Intel almost didn't get RISC-emulating-x86 CISC microcode working in time) and the third asteroid changed the entire playing field, causing even IBM to retreat from desktop. (and some would argue Apple only existed on life-support with the candy iMacs keeping the lights on until the iPod/iPhone really put the company back on its feet.)

The way Amiga was build, graphic subsystem being its own thing with CPU clued on the side, swapping CPUs would be the easiest task. Even the slowest available RISC CPU would emulate <1mips 68000-7 above real time speed with ease. PA-RISC werent sold on open market, but Amd 29K, Sun Spark, MIPS R3000 and Motorola 88100 were. 1991 Q1 25MHz CPU prices:

    Am29K $135 ~15-20 mips only with expensive SRAM/caches
    88100 $82 ~15 mips but supposedly its a TERRIBLE CPU according to Apple engineers
    R3000 $150 25 mips? almost Playstation 1 :)
    SPARC $100 20 mips

for comparison Motorola 68030 9mips $160, 68040 supposedly ~25mips !!$640!! with limited supply, Intel 486DX2 66MHz ~25mips $400-700 available late 1992. But Commodore was more interested in waiting for 1992 cost reduced 4mips $24 68EC020-16 crap comparable to $40 386SX-16.

> $150M loan from Bill Gates

There was no loan. Microsoft lost lawsuit over theft of Apple Quicktime code https://www.theregister.com/1998/10/29/microsoft_paid_apple_...

> There was no loan.

https://www.cnbc.com/2017/08/29/steve-jobs-and-bill-gates-wh...

What would you believe is true, a court documents or a PR piece on the cover of TIME? No doubt that quote for the press was part of the settlement, just like Microsoft pledging IE/Office support for 5 years https://en.wikipedia.org/wiki/Apple_Computer,_Inc._v._Micros...

Now imagine Jobs not settling and waiting one year to 1998 when DOJ decided to rip Gates ass wide open https://www.justice.gov/archive/atr/public/press_releases/19... Microsoft barely survived splitting in two, Court confirmed Microsoft an illegal monopoly and initial judgment was catastrophic for MS https://en.wikipedia.org/wiki/United_States_v._Microsoft_Cor... , a couple more arguments from a strong witness and it might have happened.

I didn't see anything in the linked document saying the transfer of $150M pictured in Time Magazine (and reported independently by many other news outlets) was quid pro quo settlement as part of that lawsuit. That's an alternate hypothesis you're proposing which I have no opinion on. You said "There was no loan" in response to my citing an event that's overwhelmingly accepted as an historical fact by every mainstream source. You're welcome to have a different opinion but my citing an accepted historical fact was not incorrect.

Both Jobs and Gates discussed the loan publicly in various interviews together and separately. The loan was secured by equity. It was on both corporation's audited financial statements. Later reporting said Microsoft eventually made money on the transaction which would also have been reflected in their audited financial statements. If you disagree with the widely accepted historical record, feel free to deep dive the financial statements and other public records to overturn the consensus on what happened. In the meantime, I'm not interested in debating your alternate hypothesis about an accepted historical fact which I cited as a minor supporting element in a long list of things supporting a sub-point.

Entirely separate from the accepted historical record surrounding the loan, I've always suspected that one reason Gates was willing to give Apple a loan secured by equity in an almost bankrupt company was that it was useful to Microsoft for Apple to survive. The existence of plausible competitive alternatives is one point any company defending an anti-trust action would make. Whereas the DOJ would certainly use the lack of any viable competitive alternatives as another element supporting their anti-trust case. Neither point would have been definitive nor conclusive, just one supporting element out of many. As far as I'm aware, neither Gates nor anyone else involved has ever said this was a reason motivating the loan nor is there any evidence of that but it makes sense given the context. I thought that at the time and, although it was unproven (and unprovable), I think the context was so blindingly obvious, everyone in the industry just accepted that it was probably a key driver motivating the loan. It was simply the smart move for MSFT to make - and to whatever extent Jobs may have alluded to the anti-trust context in his discussions with Gates (if he did), it was savvy of him to realize the DOJ's anti-trust actions created an incentive for MSFT to help ensure Apple's near-term survival and leverage that context to Apple's benefit. That MSFT eventually profited on the transaction was just icing all around.

>I didn't see anything in the linked document saying the transfer of $150M pictured in Time Magazine was quid pro quo settlement

That part was in court documents from DOJ probing Microsoft:

> handwritten note by Fred Anderson, Apple's CFO, in which Anderson wrote that "the [QuickTime] patent dispute was resolved with cross-licence and significant payment to Apple." The payment was $150 million

"handwritten note by Apple's CFO" uncovered in 1998 pretrial discovery

So the handwritten note literally said $150M? Did it specifically state August 1997? Did it discuss the amount of equity in Apple that MSFT would be getting and the terms for triggering redemption, conversion or repayment of the loan? No? Okay.

Huge tech frenemies like MSFT and Apple tend to have one or more lawsuits between them at any given time and that doesn't stop them doing other cooperative deals and transactions. If Steve said to Bill "Hey, I need you to invest in Apple or we're not gonna survive. I'm gonna launch these awesome candy colored Macs and you'll make a shit ton of money. Oh, and BTW, it might also be a good look for you to still have Apple around if that DOJ investigation turns into an anti-trust suit" and then Bill said "Okay Steve, I'll do it but I want a good chunk of equity so let's make it a loan secured by equity in the form of convertible note... oh, and if there's any other legal spats going on between us, let's bundle settling those in too." That would be completely normal and unsurprising. I'd be more surprised if an "all outstanding disputes" clause wasn't included.

The transaction was still reflected in their respective audited financials as a LOAN. If the loan T & Cs also included settling any existing lawsuits AND Steve washing Bill's car every other Saturday - that doesn't make it not a LOAN. It's just routine everyday business between two huge companies. Of course the loan had lots of other provisions and conditions and since Apple was near death, any other terms would have been highly favorable to MSFT. However, the existence or favorability of any such conditions have nothing to do with my point - which was about the Amiga computer not surviving the tectonic industry-wide apocalypse of the 90s no matter what Commodore did. The well-documented fact Apple barely survived the 90s and was close to bankruptcy, surviving only by extraordinary external intervention, was only a minor tangential aside. It's irrelevant to the point I was making what that extraordinary external intervention involved other than it being extraordinary. All that matters is that Apple's near-death survival was definitely NOT a plan Commodore could also have used to keep the Amiga platform alive to see Jan 1, 2000. I'm quite done with this off-topic digression and won't be responding further.

And may Gail Wellington rest in peace. Every time I met her back in the day, she was a pleasure to be around and nothing but helpful and kind.

> swapping CPUs would be the easiest task.

My point was not that PA RISC wouldn't have the horsepower to emulate a 68K computer. My point was that Commodore fully emulating the Amiga architecture well enough to play highly-optimized games (which heavily leveraged the complex interactions of the Amiga's co-processors) would be extremely challenging (not due to speed but complexity) and that I was skeptical they would have managed to do it (especially given their depleted resources and software bench). I've discussed this with Amiga engineers who worked at Commodore and they were skeptical of that plan at the time, so I don't think my skepticism is unjustified.

> It just wasn’t possible to compete with the price cuts and CPU performance gains that came with volume and scaling.

Commodore could if it tried, sadly Commodore was all about milking it. They were buying 68000 CPUs from Motorola at something ridiculous like $2 in mid eighties! Very slim margin over manufacturing cost thanks to good negotiations (owning foundry means you know how much it really cost to make chips, multiple 3rd party sources available). By 1991 Commodore was selling $100 BOM A500 for $300 retail. Amiga was printing money just like C64 before it, all with close to zero R&D cost. And then suddenly it stopped in 1992 https://cdn.arstechnica.net/wp-content/uploads/2017/01/commo...

1992 was the year of market being flooded with cheap 386/VGA clones running Win3.1 thanks to Intel lost AMD Am386 lawsuit forcing wide price drops at the low end. i386SX-25 went from Q1 1990 $184 to Q4 1992 $59, AM386DX-40 Q2 1991 $231 to Q1 1993 $51. Brand name 386@40 + cache, 4MB ram, 170MB HDD, dual floppy, SVGA + Monitor, Win 3.1, keyboard + mouse systems were ~$1500 while 386SX mom&pop builds started below $1000.

Turns out there are limits to how long you can sell same low end computer by just dropping price. C64 lasted impressive !12! years with majority of sales between 1983-1990, $600 down to $150 (afair at the end manufacturing cost was something silly like $30). Amiga 6 years with no upgrades $1000 to $300 + 2 more bumbling with fail half upgrades until bankruptcy.

Btw in 1990 68000 CPUs were finally $3 closing in on crazy Commodore negotiated price from 1985. Enabled SEGA to sell Genesis at $100-150 retail.

Prices according to "Semiconductor procurement worldwide" 1991-1993 by Dataquest.

Yep. Coincidentally I got my first 386 clone in 1992, and eventually started playing around with early Linux distros. I still had my Amiga, but it was obvious the writing was on the wall when you could get a decent 386SX for the price of an A1200.

I hadn’t heard of Gail Wellington before, but I read her more detailed bios that tomhow posted. I wish I could have met Gail, she has an incredible story. I can’t imagine how challenging it would have been for a woman to pursue engineering, take time off to form a family, and then return and rise like she did at Commodore. It’s evident that she possessed a lot of grit.

Our first computer was a Tandy from RadioShack. When I got really into computer animation I learned about the Amiga and The Video Toaster, which included Lightwave, which generated images for some of the best 90s television science fiction shows. It also, notably, was created by Brad Carvey who's brother Dana Carvey, who wore a Video Toaster t-shirt as Garth in Wayne's World 2.

Anyway, we finally, maybe, had enough money to buy an Amiga and drove a couple hours down to New York City to B&H Photo back when it was basically one shop, and there it was.

But the guy's at the shop said, "you don't want to buy this - Commodore is going under."

So, never did get to actually use the Amiga but, I felt like I did. Subscribed to the magazines and all. It really was a magical machine.

Just wanted to state that I cannot access the article; Cloudflare throws my Firefox into an endless catpcha cycle.

We said goodbye to Gail today. She was a true pioneer and a fierce fighter for her team and for doing the right thing. Those of us who worked with her were truly blessed.

@dang - black bar time?

yes

Reply on news.ycombinator.com