This would have been around 1988-1989, and the Mac had been out for a while; this seemed like a quaint old computer at the time. Since the Lisa wasn't quite Mac compatible, there wasn't really anything you could do with it...the software it had (early variants of stuff like Mac Write and Paint and such, called, I think Lisa Write, etc.) was what you got.
When I unboxed everything I found the receipts from when the original owner bought it all new. He'd spent something like $20,000+ on the whole setup. Computer was $10k, and each of the hard disks was several thousand dollars.
I sold the whole setup for about twice what I paid for it (I seem to recall about $300, but it's been a long time, and it wasn't super memorable...I bought and sold a lot of weird old computers back then), after cleaning it up and testing everything and tinkering with it until I was bored (I was a Commodore kid with a C128D and saving up for my first Amiga...Apple stuff was just a curiosity, not anything I wanted for myself).
Though I wasn't super into it at the time, it's one of the few things I kinda wish I still had all these years later. It has real historical significance that I didn't really appreciate at the time.
Now, buying a new computer results in marginal productivity gains because likely it's just replacing another computer and maybe improving things or maybe just adding a touch-bar.
There are a few things you can do by adding expensive hardware, but they're generally limited to advanced media processing, like video editing and raytracing. Maybe a few medical tricks.
(Gaming does not count, it scales to generally available hardware.)
Apple Lisa 1983 $9,995 ($25,143 in 2018 dollars)
Apple Macintosh 128k 1/24/84 $2,495 ($6,000 in 2018 dollars)
Macintosh II 3/2/87 $5,498 ($12,125 in 2018)
Next Cube 9/18/90 $10,000 ($19,177 in 2018 dollars)
In hindsight I should have bought Apple stock. Oh well.
Today, a Rasbpi costs $5.
Holy freakin' cow.
When I was in college in the mid 2000s my senior project needed a single board computer. We spent $500 on a small PC104 board with a 100MHz National Instruments Geode (x86 compatible) process. That didn't include any RAM or storage. I really wish the Raspberry PI would have been available back then. Especially the Raspberry PI ecosystem, makes interfacing with stuff so much easier. We interfaced a small LCD I found at a surplus shop and it took us weeks to get it working. No you can buy a touch screen LCD for $30 and there are libraries to get it working in about 20 minutes.
I'd say it's now $3,750 for the whole setup.
$3,000 for the tower/laptop plus docking station [1] [2], $650 for dual monitors [3] [4], and $100 for peripherals [5] [6] [7].
And that's with Amazon + non-Mac.
[1] https://www.amazon.com/ThinkPad-20ER002KUS-Windows-Professio...
[2] https://www.amazon.com/Lenovo-ThinkPad-Docking-40AJ0135US-Ge...
[3] https://www.amazon.com/Dell-27-Inch-LED-Lit-Monitor-S2719DGF...
[4] https://www.amazon.com/WALI-Standing-Adjustable-Capacity-MF0...
[5] https://www.amazon.com/CORSAIR-K55-RGB-Gaming-Keyboard/dp/B0...
[6] https://www.amazon.com/Logitech-Lag-Free-Wireless-Gaming-Mou...
[7] https://www.amazon.com/SteelSeries-QcK-Cloth-Gaming-Mouse/dp...
Windows is fantastically supported: Microsoft Office, Adobe suite, AAA games, Active Directory, databases, browser and browser plugins, and a thousand others.
Don't get the most expensive RAM or CPU, you lose 10% real performance but you halve your budget.
By the way, it turns out people think "automation is replacing jobs in manufacturing" because they've misread the data showing them just how much cheaper computers have gotten since then, and therefore how much more computer one factory employee can make.
Your comparison is hardly apt.
The software side also helped. Windows (¿mostly?) ran a fixed palette, while the Palette Manager on the Mac made it easy for applications to pick whatever color (out of 16M) they wanted.
On the hardware side, most PCs still had monochrome monitors, and even those with color monitors didn’t typically have ones as good as the Trinitrons that Macintosh IIs often had.
You're right though on the color palette. I never understood why Windows nor its applications didn't change it to something less garish than the default CGA palette.
If you saw a PC with a color monitor, chances were that monitor was moved over from a DOS machine, and dated from the 320x240 era.
When I think about that investment as a ratio to my disposable income at the time, I was crazy! But, as a retired IT Exec, things did work out in-the-end.
Everyone loved the C64 though, so fond memories are worth something.
I grew up with the ZX Spectrum, like so many others, and it is what started me programming/developing/being interested in computers.
You're probably looking at it from a biased point of view though. When I was growing up everyone had a Snes. Of course it was just all teenage boys that had one.
But Lisa's pricing made the Mac more appealing. It wasn't exactly affordable, but you were getting maybe a third of a Lisa - including the new GUI, which was obviously the future [tm] compared to the Apple II - at less than a third of the price.
Even at the prices being charged, the Mac and PC were competitive with previous standards, and affordable enough to have a sizeable market among affluent middle class users.
They offered more than the old S-100 boxes did, for the same or less money. And they were much cheaper and more "personal" than industrial minis like the PDP-11 and the VAX.
Regarding the price, mind that the Lisa came with an integrated office suite (long before the success of MS Office) and was targeted at offices and professionals (think dentists). Which rendered it a somewhat curious "workstation for office work", at least from todays perspective, where eventually office machines became the epitome of cheap, bare-bones boxes.
The MRD mentions as potential users secretaries, managers, and executives (of Furtune 1500 businesses) as well as bookkeepers in general.
[1] https://archive.org/details/Apple_Lisa_MRD_Marketing_Require...
I did buy a dual floppy PC clone in about 1983. I don't remember how much it was--wish I still had my receipts from that far back--but it was a big purchase for me at the time. [ADDED: I probably dithered over it for something like a year, during which time it became obvious that PC clones were the future rather than the S100 etc. systems running CP/M.] Based on ads from the time, it was probably about $2500 but a printer and software would have added to that.
And when I went to business school about a year later, I was one of very few people in my class who had their own computer. (There was a small computer lab in the school--that actually had a Lisa among other things as I recall. At some point when I was there they added a bigger lab with a bunch of IBM AT clones (80286s)).
I remember one of my friends had a Tandy the 386 ran rings around it (it supported vga graphics for one - I can remember playing the original Civilization game and Prince of Persia on the 386).
As a kid I took it all for granted, but we never would have been able to afford such hardware on my dad's engineering salary at the time. The price for a new one adjusted for inflation today is jaw dropping.
According to the article Lisa was a $50 million gamble.
I just read somewhere about the 2000 engineers Google took on for the Pixel phone from HTC. Keeping the lights on for the buildings they are in is a $50 million gamble, that's without paying them or allowing for inflation. But you get the idea, $50 million was cheap for the product compared to what hardware tech costs to develop today, particularly if it has an operating system to write from the ground up.
Incidentally, the threat from IBM mentioned in the article. Peanuts turned out to be the ill-fated PCjr that had good graphics and a bad price point for the home market. Popcorn turned out to be the first PC based lug-gable computer from IBM.
And a really lousy "chiclet" keyboard. I don't remember all the details of the PCjr flop but, as I recall, it was still quite a bit of money for a system that had a whole big bunch of compromises.