Saturday, we went to walk the Airborne march -- my uncle had presented us with tickets. We chose to walk the 15km, because that's basically 3 hours of walking, and when you're spending a day on the town you walk around that much too, right?
Except it was very hot, very busy at times, we didn't have enough water with us, and I was wearing the wrong socks in the wrong shoes. End-result: 2 medals, klik nearly fainted, I had 5 blisters and a sore ankle, both of us extremely tired.
Sunday, I had another session of the Amber campaign. This time, the shit really hit the fan: crossbow-wielding snipers popped up like nobody's business, royalty got killed, wounded and abducted, and I messed up royally. I had to write the log -- there's a pile of 6 sheets of paper that I need to turn into a semi-coherent narrative of the session. Gonna be tough.
I have been working on the RGB-LED thing. It works wonderfully on PORTB, but when I try it on PORTA, all sorts of weird shit happens. I just don't know how to fix it, not even after four days of poking at it. This frustrates me to no end -- especially as I had planned to have the whole circuit built, programmed and packaged by this friday. That's just so not going to happen.
I don't recall when it was, but one morning I woke up with all sorts of ideas for cool electronic art installations. klik bought me a nice notebook, so that I can write my ideas down, to get them out of my system. Which is good, because those half-assed ideas keep bubbling around and frustrate me even further.
Talk about frustration. Yesterday, I put an extra 256MB of memory in klik's computer: she runs a lot of Photoshop on WinXP, and with only 128MB of RAM, that is far from ideal. After some fiddling with the panel connectors of the motherboard, the system booted normally. Photoshop is working pretty fast for her now, so that's good.
Boldened by this success, I decided to equip my own system with that spiffy Zalman CNPS7000A-AlCu CPU cooler that I bought. My computer has been running on the boxed Intel cooler -- it makes too much noise for my tastes. The Zalman solution is much more efficient, and can get away with much lower rpm ratings -- add a second ball-bearing and a fancontroller and you get a really whisper-quiet cooling solution.
I've removed and re-installed the CPU cooler a few times now, it's nothing special. So I did it again: twiddle with a screwdriver to remove the cooler, take out the CPU, wipe away the old thermal paste, apply new thermal grease, spread over the heatspreader with a (bank)card, install the new heatsink/fan, connect to the header via a fancontroller, and Bob's your uncle.
Unless, of course, you do all this and your system tells you (via a crappy Winbond speech processor) "System failed CPU test". So you inspect the CPU, reinstall it, start debugging. And it keeps on telling you that the system failed its CPU test. This takes all of your evening. Then, at your wits' end, you re-install the crappy boxed HSF again -- and the system boots without a hitch.
I am... dissapointed. I got some good suggestions to my cry for help, though -- we'll see whether I can get it to work after all.
There's so much I want to do, and I have too little time for it. I also have to answer more interview questions and I need to write a review of Avenger...
: The 'voice' you get from the soundcard is barely recognizable. I wonder what the people at ASUS thought: "Oh, cool! Now you can customise your motherboard error messages, and we have someone tell you what's wrong!"
The only correct response to that is, of course: "Dude! I don't want to listen to error messages, because that means something went wrong. So I don't want to customise those messages either! And what's wrong with a series of beeps, just like everyone else uses? With the booklet in front of me, I can determine what's wrong, instead of having to hook up a speaker, crank it up and try to decipher the seeminly random stream of clicks and hisses into something coherent that might be an error message!"