Williams games, from the DEFENDER and ROBOTRON era played one sound at a time.

This code reproduces the Williams sound board, which ran a Moto 6800 at .85xx Mhz. It took arguments sent to it and or commands to play sound X via a parallel input. The classic Williams sound came from parameters input to this board. They would literally type them in and write down good ones.

Jarvis spoke about a sound priority system to give good players the right, most important sound at the right time. I hardly noticed! Read that and went to listen and defender actually interrupts sounds as needed! Which can actually make a new sound to the player. He has a GDC talk about all this.

I liked this piece. It is well written and sets a nice tone while putting some nice history out there many will find of general interest.

It ends... well, yeah. The next entry in this series should be good.

Yes, the decision to match the GS screen width to the Apple 2c was a poor one when it comes to NTSC composite video.

The result was the GS pixel clock does not align with the NTSC colorburst period. It made artifact color useless, and that was sad because the GS could have done amazing graphics on an ordinary TV.

But, that aside, the overdriving of the color signal meant most users suffered through crappy video, or ran the machine monochrome, or had to buy the Apple monitor.

Artifacts are visuals that come from contrived abuses. Or that appear despite the intent for them not to appear.

On the Apple, Woz leaned into NTSC and employed it to what I consider pretty great effect.

Dithers are contrived uses, not abuses!

The difference is one can dither on any display system and achieve visuals that convey greater color depth or more intensity granularity than the system design intent would allow for.

Again, on the Apple, artists would combine artifact colors, as well as the lack of color, say black and white, in ways that suggest other colors not intended.

The dot patterns can suggest various intensities and hues one cannot just ask for jn a COLOR = statement.

Does that help?

I will try and find an image or two to link a bit later.

Hey I missed you linking a sample image generated on the color system I was musing about.

https://twostopbits.com/item?id=5863

https://hackaday.io/project/164212/gallery#0f87e94323e101952...

I do not think it matches up. Too bad! But maybe the author will share details. I really wanted to know what they did.

Also, after I get past this appointment, maybe we can chat about dither vs artifact.

I feel the same way.

For a while, I had a SONY CD Changer in my car. You ever see the scratched up CDs often sold at flea markets for a buck each or less?

Many computers and players won't read them.

But that SONY does. And not only does it read them, the thing takes a bit longer to buffer due to all the scratches and massive error correction it has to do while buffering...

But once it does all that, it will play the disc flawlessly! Bumpy roads, too cold, too hot, disc too scratched?

All pretty much no problem.

I would gladly buy another. Was that good, but it won't read ANY recordable format.

Bummer.

It went with the car when I sold it. I put some of the very worst discs I had ever seen in it for demo purposes too.

Dude that bought the car loved it like I did. I did not mention CD-R largely because the player was going with the car no matter what.

Yes! I ran a very high end CAD tool on SGI computers at that time and it could output whe assemblies to VRML!

And that capability preceded other tools and file formats by a decade easy! And offered as many features, and a few we still don't see today given things like Siemens JT, or PTC and whatever they use now.

We would display them three ways:

Netscape was the easy way, and one got a cool view manipulation GUI that way. This performed pretty well on higher end 90's hardware. Could display most of the front of a car with full dimensions and other annotations at reasonable frame rates.

Or, use the dedicated Java viewer the program output in addition to the raw file, which is what the browser used.

Or, buy a viewer tool, and we had one. Cosmo something....

On SGI, using the Java viewer, one could view the VRML in actual 3D with shutter glasses and pull up predefined views useful to shop people.

Really cool. Very few used the real 3D capability.

And that remains true today!

I have another high end CAD tool on my laptop, and it can display 3D on my 60 inch plasma at home. Does it 60hz per eye, which is crazy for both being more than a decade old!

There are at most a handful of companies able to employ similar capability today. Space X is one who does use it however. Go Space X!

I do, and will model complex surfaces in 3D. Pure joy to use.

But, it has to be that most users either do not care, or 3D bothers them, or something...

Also, same goes for viewing files. A small number used VRML back in the day. Today, a smaller, though also a larger than back in the day number use those tools today.

The rest are on paper or static PDF.

FWIW: Open WRL will display the 90's era CAD VMRL data. Kudos to them!

Back when I ran one of those machines, I was struck by just how on point the display was!

That resolution was dead on point for the CRT Apple used. It was capable of a bit more, say 700 lines, maybe even 800 horizontally. Vertically, maybe 400 to 480 would be pushing it.

Say they used 640x480. The user would have been happy with a pixel in the horizontal direction, and maybe less happy with it in the vertical one. And it would have been a bit less crisp all 9bwr the screen, IMHO.

Monochrome CRTs can be over driven to a crazy amount and they just work. And often, unless the overdrive is just crazy, the user will probably see the differences as the GUI changes too. I have run 1024x768 on a 7" amber screen.

Worked, but not well.

It is much happier with about 500 vertical lines and more like 700 horizontal ones.

PAL ish monochrome 720x586 looks fantastic and is 50hz. Slow phosphors = 50hz being no big deal. NTSC ish 720x480 at 60hz looks good too, maybe a bit more crisp.

All I am saying is for that CRT, the resolution Apple chose will look great! And that is due to that particular CRT and drive circuit.

I think Apple could have pushed it to 640, lime the GS machine could do, and go 400 lines vertical and that CRT would perform almost as well. We may not even be able to tell.

Maybe they did not do that as an overall balance between what the 68K could pixel wrangle, RAM and CRT performance.

Cosmopolitan is really interesting!

I ran the APE file on Android, using termux with nothing more than a quick request to install whether.

To think that binary would also work on a PC, or mac... crazy!

I agree with the author: home grown video systems are cool. Very interesting. I want one