Posted: 4/24/2008 12:15:04 PM EDT
|
I have spent almost 2 months working on a program. It uses a massive amount of RAM and does a massive amount of crunching. I'm at the point where its pretty much done and I'm nailing down the remaining bugs. Unfortunately, the last (and one of the most vital) parts of the program takes 2 hours to crunch through depending on input size. It sucks to wait through it and find out if you made a mistake or not. Before and after the long crunch I can create and load memory dumps to speed things up and test the remaining parts. This sums it up nicely...  -Foxxz |
|
Computer frustration...I had a big load of it today. I've got a customer that needs an old two-way radio reprogrammed. It's a nice radio, but it's almost 20 years old. I have the programming software and equipment, but it runs on a.... 486 class machine. And not a faster one.  I don't have one, so I go out and get one. There's an e-recycler not far away. I find the ideal platform for the job, an old but running Compaq 486DX66, with the PC integrated into the monitor chassis. I get it back to the shop, throw in a small but good hard drive with my software (and DOS) already on it, get set up...and spend most of the day banging into assorted problems, which include but aren't limited to program bugs, difficulties in upgrading to the newer version of the programming software, not one but TWO defective and/or miswired programming cables, power troubles...the list is long and irritating. But, the computer worked perfectly once the software configuration was set up right. That's a good thing, because it's not so easy to pick up a 486 class machine on short notice these days. I now have the ability to easily reprogram a lot of older radios that are still plentiful out there in service, too, so I can offer that service and make a few more bucks on it, now that the learning curve has straightened out and things WORK. It was fun...not. But at least I got through it OK and the customer's happy. CJ |
|
Well, it crunched through OK and the output it generated looks good. I'm really happy because a major bug that was introduced yesterday was fixed today after I couldn't figure out what was going on. That aside, I went onto a more productive use of my times and recoded another entire section that I was initially given poor specs on. Thats working. I've added some optimizations to try to speed things up and cut down on redundant output. Its running again. I'm going to be off work before it finishes!  -Foxxz |