HP 35s keyboard weirdness

[Hiawatha Wheelwright]

I am a new owner of an HP-35s. I've skimmed over the list of known bugs, and believe I may have a bit more info on one of them.

There have been a handful of times when I have entered a calculation without looking at the screen, only to see a result that was egregiously large and obviously wrong. I assumed that this was the result of the known "dropped keystroke" bug, where occasional binary operations would be missed by the calculator.

But it just happened again and this time I saw what happened. I had entered a single digit, pressed +/- to change the sign, and then pressed the multiply key. The multiplication was not performed, and after a noticeable delay (maybe a bit less than one second) the digit "4" appeared next to the existing number.

ie

x-register display          keystroke

32                             2

2  (32 now in Y)              +/-

-2                             *

-24  (32 still in Y)

At this point, entering another digit bumped the -24 up into Y and the 32 into Z. Has anyone come across anything like this? Seems like very odd behaviour but I will certainly be keeping an eye open to see how this manifests again in the future. I am currently theorizing that the binary keys in the right column will under some circumstances be processed as the corresponding digit in the second column, but the fact that the next digit did a preceding push doesn't quite fit that. Almost as if the * got processed as a 4 followed by some sort of no-op.

Edited: 10 Jan 2010, 10:33 p.m.

[Gene Wright]

and thanks, Hiawatha for making the note on amazon.com, but the review you pointed out that came from datafile was put there my me. That's my amazon.com alias it was posted under. Just an FYI. ;-)

[David Hayden]

To track this down and make it reproducible, I'd try focusing on the delay. Does the 35s do garbage collection? Based on the tiny amount I know about it, I'd suspect not, but who knows.

Or maybe something wacky in the keyboard software.

[John B. Smitherman]

Hi Hiawatha. I believe that others have experienced erratic behavior from the 35s as discussed here:

this thread

I'm hoping that HP finds the resources to tweak the 35s someday.

Regards,

John

[Eddie W. Shore]

I followed the steps presented and I got the correct answer of -64. Was there a keystroke after the 32?

[Hiawatha Wheelwright]

Just to add to the list: Today during a couple of calculations, I got some dropped digits. ie, I'd hit a numeral, get a keyclick, yet the digit didn't show up in the display. Second time would be fine.

[Bart (UK)]

That could be due to the lack of a keyboard buffer. However, I have not yet been able to reproduce the error you listed in the first post yet.