In 50 years the most I’ve ever done with floating point was coding some >builtin functions for the Iron Spring PL/I compiler.
I got to do all of the financial functions which involved a whole lot of floating point arithmetic. Calculating the Internal Rate of Return for
a series of N payments was finding the root of an Nth degree polynomial.
Bond yields are based on IRR with soem extra confusion for partial time periods. I managed to get them to match the official answers but it was tricky.
On Tue, 24 Dec 2024 04:20:30 -0000 (UTC), John Levine wrote:
I got to do all of the financial functions which involved a whole lot of
floating point arithmetic. Calculating the Internal Rate of Return for
a series of N payments was finding the root of an Nth degree polynomial.
Bond yields are based on IRR with soem extra confusion for partial time
periods. I managed to get them to match the official answers but it was
tricky.
In the early '80s I worked on a handheld pH meter that used an 8049. Unfortunately I can't remember how I got from the A/D measuring the
voltage on a Ross electrode to pH. Another programmer worked in parallel
for an ion concentration meter using the same hardware but a different
custom LCD display. I do remember we couldn't shoehorn both into the same device.
The Z-80 based bench models definitely were using FP. Much of the work
I've done in the past 25 years has involved GIS, often using doubles
rather that floats to get the necessary precision.
Horses for courses.
Sysop: | Tetrazocine |
---|---|
Location: | Melbourne, VIC, Australia |
Users: | 4 |
Nodes: | 8 (0 / 8) |
Uptime: | 214:56:30 |
Calls: | 73 |
Calls today: | 1 |
Files: | 21,500 |
Messages: | 73,905 |