New Type 30 display emulator!

[Bill E]

I've done a completely new replacement for p7sim that is a lot lighter weight, much lower cpu loading, and faster. I'm just wrapping up packaging, will put it in my repo soon. It will also work with the original pidp-1 branch, so you can build just it if you want, it's one C file and one Makefile.

Here's my writeup if you want to see what it's all about.

Bill

[Bill E]

This is what snowflake looks like with my display. 

Bill

[Bill E]

Minor update to t30dpy to improve scaling when the window is not 1024x1024, minor performance tweak, docs updated with new tests.

I added type340stress as a test, this is brutal sending over 500K points/sec to the display (yes, that is what a real t340 would have done, very impressive for the time!).

That means t30dpy is rendering about 20 million display pixels/sec. Original p7simES fails miserably on this, t30dpy doesn't drop a single point. (pats self on back). There are still occasional issues with SDL pixel scaling for some window sizes < 1024x1024, seems there is no good solution. But, many subsizes are fine.

Bill

[Bill E]

More updates to add a config file, tweak settings, add gamma correction, and much fiddling to get SDL's scaling looking good with all screen sizes. Well, as good as it can do.

Several of the key parameters can be adjusted in the config file, especially the gamma setting. Docs updated, but see the code for full details and also the info in the sample config file.

I hope someone appreciates all the effort I put into documentation in both the code and the md docs. Actually, I do this partially so I know what I did when I come back to something

weeks or months later. :)

Bill

[Glenn Babecki]

Bill,

I am way behind on all things PiDP due to many other priorities, but just wanted to let you know your efforts are deeply appreciated.  Your work is so prolific that I have to learn to "drink from the firehose" all over again (if you catch my meaning). 😉

The upside (for me) of lagging behind all the updates is that many more kinks are getting rung out and new features are emerging!

Many thanks for all the hours you've put in on the PiDP-1.

--
You received this message because you are subscribed to the Google Groups "[PiDP-1]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-1+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-1/79865d03-bfe8-4250-bbf2-59422c10997an%40googlegroups.com.

[Bill E]

And as usual, I wasn't totally happy. I've refactored the code a bit once I realized that all of the rgba values can be precomputed. Now there are no floating point operations done in the display loop, at least not by my code. It uses about 25% of the cpu that p7simES uses and looks a lot better too, imho.

Bill

https://github.com/wjenh/pidp1-mods.git

[Bill E]

I came across this pic of spacewar on the Computer History Museum PDP-1 that shows what I meant about p7sim not accurately reproducing the P7 phosphor. Notice the visibly blue dots. I've tried to match the original as best as I can using the published P7 dual  phosphor colors and decay times, but it takes some hackery and magic handwaving to do it on a raster-scan, frame-based LCD. The original drew the dot in 5 microseconds with an intensity no LCD can come close to.

Bill