Fred's firmware development diary

[Fred]

Yesterday (I think?) I got a binary repository setup for maven deployments. It's not 100% finished yet, but it's 75% finished, I guess.

Today I've been doing tuner work of various kinds among other things. Tomorrow, I feel more tuning work coming on, and some website stuff too.

More news as it comes to hand :-)

Fred.

[Fred]

More tuner work done yesterday! More different tuner work done tonight, by proxy. Much website work done earlier today and yesterday too. Aiming to release a little surprise on Monday morning next week, NZ time. Fingers crossed that I can get everything that I need done by then. Onwards and upwards.

Tomorrow will be website stuff and Andy stuff, yes, that's right, it's time to put some time into the firmware again.

Fred.

[Fred]

I didn't get to Andy's stuff yesterday, though I did do some of the web stuff and I did do a bunch of tuner work. I got an application to talk FreeEMS speak, so it's just a matter of cleaning it up and we're in business, at least in a simple way :-) More completely later.

Unrelated to what I did is what Mike has done:



Not bad for a few short weeks from scratch! :-)

Thread here: viewtopic.php?f=41&t=1741

It's growing fast, so expect more news/screenies from it in future.

It looks to me as though there will be no less than 7 applications talking current FreeEMS soon :-) Pretty cool.

Fred.

[Fred]

And tonight, nothing works! :-( I might have to quit out of X or maybe even reboot the box. Bizarre. I was working with Mike testing stuff and one moment it was working, the next it was not. Perhaps it's time to do some testing on this machine instead. Not happy. Reloaded the entire driver stack for USB, no luck. Weird.

[Fred]

Works on mac, at least, it builds/runs/doesn't crash. Will try it with a device sooner or later, too.

[Fred]

Had a play around. Managed to stream data from the Puma to cutecom on the mac. But no signs of life from the venerable TA card, regardless of which machine, and no signs of serial regardless of which adapter, on this box. I think I'm cursed. Today looks like it's going to be be other chores. Sorry Andy! :-(

[Fred]

I've been cleaning up my git repo for the firmware, and others' too. I deleted 21 branches on my private server (automatic uploads of temp stuff) and pushed a few new more meaningful labels too. Cleaned up some local stuff too. Much better!

Still need/have in progress:

• Reflow oven (local and private) - In flux (ha ha)
• <currentSensorCommit> (local and private) - Maybe tomorrow?
• <allOfSeansBranches> (local and public) - Sean's stuff will be a while...
• Preston's fixed cranking hack (local and private) (2: one he's running, one may be helpful for a permanent fix)
• DIS experimentation (local and private and redundant) - Need to do some other work before I can complete this. This branch just holds some config, nothing else of value.
• notes on one of the issues (local and private) - Should paste these into the issue and delete.
• partially finished dual sync rate commit (for sim especially) (local and private) - Other priorities, but good work done here.

Sim is also on what was dev, now. So I pushed it to master as good. That doesn't mean all configs are good. Just that the base is good, and the decoders are likely good. The config was transferred from one format to another, thus I could have made a mistake. I hope everyone upgrades when they get the chance, it'll give me warm fuzzies. This code is now 2 months old and has only 4 users on it? Hopefully we can do better in future. What I might do is this:

Set of commits pushed
Set of testing required
No further firmware work until testing completed
Repeat

I can work on other important stuff while people get around to using the existing work, then.

Fred.

[Fred]

The code shown in this video is 500 commits and more than one year old! Imagine how good it is now! :-)

http://youtu.be/6l-KyVG-teM

It wasn't me, I didn't do it.

:-)

[Fred]

Effort brings reward! :-)

http://youtu.be/2Ycjg8DFmZk

Mmmmm, ice cream.... mmmmm.

Fred.

[Fred]

First start of the scheduler on which all current code is based!

http://youtu.be/Fzr1TXJWQwE

A great moment :-)

Fred.