HP Forums

Full Version: my new old toy
You're currently viewing a stripped down version of our content. View the full version with proper formatting.

i just received a privileg sr 54 nc from someone who i mail calculators back and forth with. it is sort of a german hp 45; rpn and deadly accurate. in the calculator forensics thread a while ago it fits in about second. i am very impressed. here is how it fits in:

saturn chip: 1.35733

privileg: 3.85748

ti 58/59 series: 4.661314

hp 41 & 11: 417.403

hp 67: 4076.64

hp 35: 4076.9

ti 30: 177087.1

casio fx-101: 785883.0

i still prefer the 41 to any calc/comp ever made, but there is something to be said for that kind of accuracy - and in the 70's too.

How do the numbers relate to accuracy? Explanation or link please?

It was posted in another thread some days ago. I think it's from a Mike Sebastian web page titled "Calculator forensic".
You put the calculator in degrees mode and calculate the sinus of 9, the cosinus of the result and the tangent of this last and then you undo the path, arctangent, arccosinus and arcsinus. In theory you should end with 9. In a HP41 you have 9.000417403, in pioneers and up 8.99999864267. Then you make the diference with 9 and multiply by one million.

The comparison method was developed by Mike Sebastion and he calls it Calculator Forensics. You can read about it at his website:

http://calcinfo.com/

Mike prefers to compare calculators by the number that comes back from the asin operation. He has some suggestions on how to extract the non-displayed guard digits that involve multiplying by 1,000,000 or so. The idea of subtracting 9 and multiplying by 1,000,000 is my variation, which I started doing before I looked at Mike's website. I first saw the basic algorithm here.