Re: High-end calculators in the WSJ Message #5 Posted by Allen on 9 Sept 2009, 9:10 p.m., in response to message #2 by hecube
Quote:
The way to the future is Mathematica on Touch/iPhone devices with a simplified "basic user" overlay.
No debate here..
Quote:
All calculators should have a back-lighted color display that doesn't have to be as big as the iPhone screen. Something nice to look at at any rate.
Respectfully, I disagree. Backlights are fun for gameboy's but almost useless for a calculator. While I agree it would be nice to have, the cost in $$ and battery life will not nearly make up for the 2x per year I would use the function- and I use my calculators alot. With a gameboy there is no inputs and outputs to the system. you respond to things on the screen by pressing buttons, which changes the display, which causes you to respond... etc...
With a calculator, you almost always have numbers or other data to put in, and something to write down when you're done, and both those things require light or at lest a pen with radioactive ink. Did they ever make a glow-in-the-dark slide rule? <idea>
I also respectfully disagree on the Color LCD. not worth the suck on battery power, and the added color will probably not add any information to the screen whatsoever. I reference Tufte and his many books about Data Ink. If it is not required to display data, get rid of it and Use the ROM/display space/battery life/development costs etc.... for something that does.
Quote:
Whatever Steve Jobs says, there's nothing like tactile feedback for the keys.
Could not agree more. :)
Quote:
To me, the biggest mistake HP ever made starting with the HP-28C is to get rid of stepped programming and treating programs like data. From a practical point-of-view, it becomes a big obstacle to access the computational power of the calculator.
There are strengths and weaknesses to both. The power of the 3rd generation calculators 28,48,50g is that you can do more complex functions- for example parallel list processing, multiple data types, longer strings, etc.. Some things are easier on the RPL, some on RPN. I would not trust a 'one is better than the other' statement from someone who has not mastered BOTH- and even then I would suggest there is a large middle ground where either are good. There is also the 71b BASIC/FORTH language which also has strengths and weaknesses.
Quote:
Let's be honest here, no one is ever going to do artificial intelligence research on a pocket calculator.
Never say never... that really depends on the heuristics of the program. I've found the decision algorithms in HPChess to be quite advanced for a hand-held calculator. It's not "Deep Blue", but that's beside the point. FYI, in some ways the HP 50G is well beyond the early 1976 Cray-1 Supercomputer. The beauty of it is that a 15 year-old student can do on a 50g what took scores of PHD's to do on a computer in the 1960s- if it was possible. Feynman helped design nuclear weapons and quantum electrodynamics with far less sophisticated machines.
Quote:
The calculator's main reason to exist is the "immediate result". If I can't get that, it's dead in the water.
May I disagree again? There are many uses for calculators.. Fast and accurate results are one (although computers are better at both). Calculators "boot" in a fraction of a second- even today it takes my computer several minutes to turn on. So I think convenience takes front seat over the 'immediate result'. The other issue is portability. Until recently, you could not really cram a supercomputer into your pocket. Today with a smartphone connected to the internet, you don't need to....you can call up wolfram|Aplha or other site.
Quote:
I never got around to program my HP-48 but I could program the HP-41 within the first five minutes when I got it.
I have 6 of each.. try the 48, you might like it.
Quote:
"trapezoidal"
Please click here :)
edit: removed link to Feynman interview
Edited: 9 Sept 2009, 9:26 p.m.
|