One of the first video games makes a comeback

12/15/2010 9:05 PM EST

Wow! Really amazing from the video clip. Many "How did they do that?" kinda questions really pop into my mind

12/16/2010 2:30 PM EST

What would you like to know?
My dad built it (yes, at BNL, back then).
I played it (ok, I was only 8 at the time).

12/16/2010 3:33 PM EST

I remember putting together a game using the AY3-8500 from General Instruments in the late '70s. It was a project in Elektor and it was my first project with a "huge" 40-pin chip....but this is 20 years before that...amazing....

With digital it's all easy...but to get the ball AND the net AND the baseline all in analogue must have taken some doing at the time... presumably on a dual-trace scope (don't think they had 4-input scopes in those days??)

12/17/2010 11:22 AM EST

@David: yes, it is a dual-trace scope, but only 1 channel is connected in the picture! The 2nd channel input is the BNC in the bottom left corner.

12/18/2010 5:20 AM EST

@Antiquus: thanks. I have been thinking a lot about how I would do this - not being a hot shot designer makes that a bit difficult. The ball I thought would be easy enough, an LC or gyrator circuit with active full wave rectifier. And for the horizontal motion, a triangle wave generator which can be reversed at any point. Easy enough.

But then I watched the video again. And sometimes the "ball" goes high and slow, and sometimes low and fast just over the net. So depending how it is hit, not only the vertical velocity but the horizontal velocity changes. Bit of an advance on my AY3-8500!! I guess you could consider the hit as a vector and take the horizontal and vertical components (depending on the pot setting) and apply them accordingly.

My conclusion: these guys certainly knew what they were doing!! Are the schematics available anywhere?? I love analogue stuff like this...

12/20/2010 8:18 PM EST

Great! This is a very interesting article. To be a part of something that has been around for half-century is excellent. That means that the designed team did an excellent job. It shows that during the early days it was more about during it right and having fun. Today it is about just trying to make a profit. Happy that they are still around to share this wonderful story.

12/21/2010 9:55 AM EST

OK, I can understand that this seems neat. BUT this is typical US Govt lab waste of labor... These guys had (and still have) too much time on their hands to design these toys. You don't see this kind of abuse in private industry. And if you do, these private firms aren't making a profit and wont be around long.
As a U.S. taxpayer, I have had enough of government waste and abuse. I've seen it at Sandia and now I see it at Brookhaven. I am sick of paying high taxes to support this non-sense. Lets get on doing real science at US Govt labs and enough of the toys for personal amusement.

12/26/2010 10:00 AM EST

Yeah I too agree that such projects should not be interpreted as wastage of tax-payers money. The fact that these have been in existence would have forced many a minds to think something analogous or create something new & challenging...

12/26/2010 6:46 PM EST

It looks like the analog computing element is actually computing the ball behaviour using a true kinetic model. The plot doesn't reverse the ball direction; the scope displays the analog ball XY position as it is computed over the time period between raquet strikes.
The relays probably multiplexed the trace between the net and ball displays, as they were cheaper than tubes.
The court display looks like a separate Leftwards ramp pulse, a vertical blip pulse and a rightwards ramp pulse; a couple of relay monostables would suffice. The hand controller pots could use these ramps to set the player's distance from the net, and also the 'strike' power using the time the switch is held down for.
The ball computing model would generate XY values, and be given new dX/dt direction and dY/dt velocity starting parameters with each raquet-strike, these having been set by the hand controllers; X also has a height term as charge on a capacitor having a constant -X bleed to simulate gravity; and Y position being obtained by integrating ball velocity over time, with a negative drag coefficient.
Note: assuming the computing element works in two quadrants for Y, there is no need to flip the display left-right for alternate striks, because the controllers 'hit' the ball in opposite directions, and the compute function works equally well with negative and positive ball velocity inputs.
I like the 'bounce'; if the ball hits the ground (one more comparator) the sign of the X velocity component is reversed.

12/29/2010 4:04 AM EST

Not sure how they do it....seems like heavy and too much real state.

1/2/2011 2:41 PM EST

Colin,

I would appreciate more reference information on this early effort. Names, Schematics and links to SIG boards. I enjoy reading more about what drove this early work.
I am a great fan on the origins and motives driving this research.

1/4/2011 2:31 PM EST

Electrical engineer was lot more fun than today. Engineer worked hard to save a few gates here and there. Now days, who care about those?

1/19/2011 9:42 AM EST

Question for Blater.

How come you have "extra" time to leave a comment?

Others aren't allowed to "waste" time, but you can?

Don't you think this is hypocritical?

1/19/2011 9:49 AM EST

It is valuable to balance work with fun.
It amazes me how many times I have used some tricks and techniques that I originally developed "off the radar screen" i.e. not immediately work related. But later found those techniques (or skills I learned) useful later on my paid job. For a designer, inventor, or developer to grow and expand his skills sometimes a little exploration is very valuable.
It amazes me how often an extra-curricular excursion that is not immediately "justifiable" turns out to be very valuable for regular (paid) work. Developers to need develop themselves and sometimes this is only done when they are allowed to explore a little here and there. The trick is to have a good balance.