The Week In Review Oct 23

Ye Olde Day Job occupied way more of my waking hours this week than it ever should.

I did manage to get a little bit done:

• Added a new driver to the linux kernel for configuring and driving the GPIO pins needed to  interface to the pinball machine.  Removed a couple of the test drivers that come with the prebuilt kernel that conflict with the GPIO usage I need.  The pinball driver is not complete, but it does allow me to reconfigure the GPIO registers to use the pins exactly how I want.  Hopefully I will get to test it this weekend.
• I have been doing a lot of scratch pad calculations on how many GPIO pins I actually need.  For certain I will use 6 for the DMD and 11 for the switch matrix (3 for selecting and 8 inputs).  For the lamp matrix I’m pretty sold on using an LED driver which would only require 3 pins.  The final decision to make is exactly how to drive the solenoids and flash lamps.  I’m thinking, if 3 pins loading and running the lamp matrix is fast enough, would a similar driver also work to drive the coils?  At this point it will all be theoretical until I actually wire something up and test it.

That’s it for this week, but under the circumstances at the “day job”, I’ll take it.

The Week in Review Oct 16 (er 19) 2009

Once again, a small list of small accomplishments from last week.  Keeping the project crawling along:

• Finalized timer system and gave it a good testing.  Should create a unit test for it just to make sure all corners are covered.
• Reworked some of the simulated DMD code.  I had frame rate calculations happening in the simulator code side which didn’t make much sense.  Started moving the frame rate calculations to the driver so that it will work for both simulator and hardware.
• In a flashback homage to my late 1970’s basic programming days, I hand coded a 5 x 7 dot matrix font for use on the DMD.  Of course this time it is done in C.
• Worked up a good prototype for the switch matrix driver.  Next step is to create some kind of simulator input for the switches.  As with all of my code, I may end up discarding this and rewriting it a few times.
• Started spec’ing out how to program the various modes for the game.  The current debate is whether to code directly in C, or to define a language for coding the game play.  As one does not preclude the other, I will probably be coding a few modes directly in C for early testing while the debate rages on.  I’ll write this up in detail at some point as it is exceedingly important.
• Started the process of wiring up some test circuits and getting the 2440 hooked up to the protoboard.  Came to the realization that I really should fabricate a nice project box for this instead of housing the key components in an old cardboard box.  I envision a trip to TAP plastics in my future.  I hear the place is fantastic.

The week in review Oct 9 2009

I’m trying hard to do at least a little bit of work on this project every day.  Most of the progress is pretty mundane.  Some simple coding, some schematic wrangling, some data sheet spelunking.   The simple truth is that not everything going on is very exciting, so I don’t feel compelled to post about it.

It did however dawn on me while reading one of the many project blogs I follow that these little advances made every day are what keep the project alive.  If you skip too many days, you have to go back and remind/retrain/relearn things that should be second nature if you keep at it on a regular basis.

To that end here is a dump of the little things that kept me engaged this week:

• Wrote the basics of the swtich matrix code.  Still a lot of work to do including getting some kind of simulator switch matrix working.
• Wrote, re-wrote, then threw out and wrote again the basics of the timer system.  I’m still not quite happy with it.
• Bought the correct 2mm pitch cables from ebay to wire the GPIO port to the protoboard  (very, very exciting)
• Identified the locations in the kernel drivers that need to be modified to use the available GPIO pins.  Came to a fairly good understanding of how the GPIO pin settings work.
• Spent a few hours understanding exactly what the Stern/Data East switch matrix hardware is doing.  Made of list of components to order (probably from Digikey) to wire up a test circuit of my own.

Hopefully these will turn into some full and more interesting posts in the near future.

That’s odd

I have been staring at the schematics for the CPU board for a long time now.  Months, if not years.  I just noticed that there is a typo on page 3 of 3.  They label both the send and return for the switch matrix as SWITCH ROWS.

The new version of the CPU board for Stern games (CPU Sound Board II)  has a completely new design and properly labels the connectors.

This has absolutely no bearing on anything, but it just struck me as odd.

For the record, the top connector driven by the the swstb (switch strobe) signal should be SWITCH COLUMNS.

In other general rambling news, my procrastination with cables has paid off.  I located the exact 2×20 2mm pitch connector with ribbon cable for the sbc2440 GPIO port on ebay!   This brings me much joy.