This was the hardest part. If you wire the wrong things together you can brick your Pi. I took some time to decide on which pins to use, and after assembly and testing everything worked. I’ll explain which pins I picked, but note that this information will not work with other hardware configurations.Continue reading →
I have DCS World, and whenever I get a new module I like to try and “just fly the thing” with minimal cockpit familiarity. I always expect hilarity to ensue but it’s really not that exciting. Like with the Ka-50 I just crashed with in-air starts, and could never get it off the ground with a cold start (until I read the operating manual). But helicopters are hard, so that’s not a surprise. Planes are easy. After getting the MiG-21BIS module though, I did have a few solid chuckles.
Here’s my in my first ever flight (started in flight at about 500 meters). I attempt to find the gear and flap controls and land. Gear was easy; flaps, not so much but I ended up ‘cheating’ and using a keyboard shortcut for flaps as I feared I was coming in way too fast. Which I still was. Skip to 4:50 if you just want to see the ‘bouncy’ landing.
And then I fumbled around trying to open the cockpit and extinguish the fire (which was a entire-plane-on-fire, not engine fire), but my fire extinguisher system was off.
Got a new oxygen sensor installed, tuned it a bit, and went to an autocross in my 280Z a couple weeks ago. I was third in my class! Out of three, though. So last. Still, the real victory is that my car made it there, ran, started and restarted fine, and then made it back home.
I also got to try my slicks out for the first time, and man, they make the car much more predictable. Lift-off oversteer is still there but manageable and much more forgiving (which probably doesn’t help that I have a bad habit of inducing it). Traction is not as greatly improved as I thought it would be, but I was just running them on the rear.
Anyway, here are a couple runs. A fun run (no time, just learning the course), and my fastest run.
I added an update to msqur today. Just a small update to the landing page since it was kind of blank. Moved the settings icon to the view page instead of every page. Also fixed the readme and license files for formatting.
I get a lot of hits from Russia on that site. Actually, I get a lot of hits from Russia on all my sites. I imagine most of them are bots probing for weaknesses. A bunch are trying to manipulate GET requests, but there’s not much you can do. Surprisingly little abuse to the upload function, which I thought would have been the first thing to be attacked. Msqur receives quite more hits than my other sites. Nothing’s happened so far, but I make backups so I’m not too worried.
Anyway, some thoughts for future updates are browse filters (by engine size, compression, etc.), a search function (not sure how to distinguish it from browsing since there isn’t much plain text to search for at the moment), and pagination to the browse index. But again, it’s on the back burner so not a huge priority for me.
It’s been a while since I’ve worked on my Z. I’ve been working on rebuilding my wife’s 300ZX, which is progressing slowly. My Z had a dead oxygen sensor and was not tuned correctly. It also had a fuel leak. But last week I fixed both of those issues and now it runs like a champ.
I then leaned out the fuel table by about 25% (I had richened it up when the sensor died to be safe). I was able to use autotune to then fix up the fuel map quite a bit. I really think it’s time to get it tuned on a dyno, but haven’t committed yet.
Sent off my oil temperature gauge to get the range changed from 100-300 degrees F or so to 60-260.
I’d like to use the slicks I got in at least one race this season, but I need to remove my bump steer spacers for the 15″ wheels they are on to fit.
So I picked up a Bosch 044 fuel pump for a song, and decided to get rid of my Fuelab Prodigy pump. I almost wrote about the Fuelab install, but since I intended on replacing it, it wasn’t worth it. It was just too much for my vehicle, even at my target boost levels.
The Fuelab pump was brushless, and so was very picky about voltage. It was unusable during cranking due to a large voltage drop because of my small battery. So I had to add in a check valve, which helped, but I still needed to do a ten-fold increase in my cranking pulsewidth calculations. Then there was the speed control wiring. It was too much ‘because racecar’, so I decided to swap to the Bosch when offered.
To my surprise it came with a check valve integrated in to a type of banjo fitting. But, since I already added a check valve closer to the fuel rail, this would be redundant. Also, banjo fittings flow poorly.
A couple of new fittings and moving the mounting brackets over a bit was all it took to get it installed. Note all the holes for all the difference bracket configurations on my mounting plate. Heh.
I produced some videos in 60 FPS, but couldn’t remember which ones. Since, of course, I didn’t label their FPS at the time, I needed to find them in vast sea of 30 FPS videos. Rather than just use a simple grep with ffmpeg to find which ones, I thought it might be more useful to find all videos that have a certain FPS threshold.
Usage: ./60fps.sh [directory]
FPS=$(ffmpeg -i "$1" 2>&1 | egrep -o '([0-9]*[.])?[0-9]+ fps' | egrep -o '([0-9]*[.])?[0-9]+')
if [ $? -eq 0 ]
TEST=$(echo "$FPS"'>'50 | bc -l)
if [ "$TEST" -eq 1 ]
echo "$FPS: $1"
# echo "$FPS: $1"
# echo "Could not get fps from: $1" >&2
for FILE in $(find "$1" -type f -name '*')
# echo "trying: $FILE..."
If I thought I’d use it more, I’d move the hardcoded FPS test value to an optional argument for the script. Instead, I’ll leave that as an exercise for the reader.
As mentioned previously, I wanted to open the source to msqur up and put development on the backburner. I have added an appropriate license for it (although I do need to add library licenses/notices in there as well). It is now GPL licensed, and I have made the repository public.
Some things I added:
Added basic charts for the 2D tables
Added table header text to tables
Added reingest script to allow for easier MSQ cache updates
Fixed some of the documentation up
Known issues, mostly the same as before:
MS3 file support is poor
Constants are just blurted out without any organization
INI Parsing of formulas and directives not implemented yet
API documentation is still in infancy
I want to move on to other projects, but I may come back from time to time to update things. I intend for the near future, at least, to leave msqur.com up and keep it updated.