7 of the Best Free Linux Calculators

An anonymous reader writes "One of the basic utilities supplied with any operating system is a desktop calculator. These are often simple utilities that are perfectly adequate for basic use. They typically include trigonometric functions, logarithms, factorials, parentheses and a memory function. However, the calculators featured in this article are significantly more sophisticated with the ability to process difficult mathematical functions, to plot graphs in 2D and 3D, and much more. Occasionally, the calculator tool provided with an operating system did not engender any confidence. The classic example being the calculator shipped with Windows 3.1 which could not even reliably subtract two numbers. Rest assured, the calculators listed below are of precision quality."

Comments

Useless.

[SanityInAnarchy]

No maxima? How about kmplot?

Re:Useless.

[Yvan256]

No maxima, but it can do altima and sentra.

Re:Useless.

[Annymouse Cowherd]

No dc? What kind of Linux is this guy smoking?

Re:Useless.

[maxume]

Desktop weed. The calculators in the article all have clicky-button interfaces.

Re:Useless.

[xororand]

One of the mentioned calculators has a pretty usable CLI though: "Qalculate!" - a great calculator for dealing with units, especially currencies.

Example:

> sphere(2 furlong) * (1.293 g/m^3) to kilogram
approx. 352739.273 kg

Re:Useless.

[raddan]

I am a huge fan of Landon Curt Noll's calc program (known as "apcalc" in Debian/Ubuntu), and it's usually one of the very first things I install on my machine. It's CLI-only, and having it available to my via SSH means that I have a great calculator available to me anytime via my mobile phone. I don't care for RPN; calc lets me write an expression just like how I would on paper, so it is very intuitive. It has a large scientific library, too.

When I need to graph, I use gnuplot.

Re:Useless.

[X0563511]

how do you write functions in a line entry system just like you do on paper? You've never used the rational symbol? How do you handle roots?

Re:Useless.

[insufflate10mg]

Just because there's no rational/root key on the keyboard, that doesn't mean there's not a keyword that can be typed to denote the function or operator. It's definitely possible. Annoying? Yes, but definitely possible.

Re:Useless.

[Joce640k]

Is there a Windows version? Not everybody uses Linux you know...

Re:Useless.

[AleBaba]

Actually if clicky-button-people install the package gnome-genius (in Debian) they won't be very pleased.

Re:Useless.

[Anonymous Coward]

Yeah, I might consider switching one day, but right now the interface is unintuitive, there's no consistency in the way to install software, and it just doesn't have the apps I need.

Windows, that is.

Re:Useless.

[jirka]

you probably have not looked at any of them then. Just because something has a clicky interface doesn't mean it doesn't also have a cli interface.

Re:Useless.

[maxume]

Are you intentionally misunderstanding my comment? The feature that they share is that they have graphical interfaces, I didn't say anything about them being limited to the graphical interface.

And I thought to myself, I wonder which one of them he wrote to be so defensive, and I was right (but I do have to admit, yours certainly is not well described by 'clicky-button interface', but the rest are).

Re:Useless.

[Anonymous Coward]

Cause they're meant for people to actually use.

Re:Useless.

[Anonymous Coward]

What about python?

Re:Useless.

[Anonymous Coward]

awful! most of them have simulated calculator clickety interface! eeeeeew.

and octave is missing.

Re:Useless.

[X0563511]

I didn't say it wasn't possible, but there's a big departure from how you write them on paper than how they fit onto a line-entry system.

Imagine that you wanted to root something that looks like the quadratic equation. Imagine how that looks on paper, and imagine how that looks on the display. That is what I mean by "like you do on paper"

They left out RPL/2

[jdb2]

While technically not a "calculator", unless you run it in interactive mode, RPL/2 is one of the oldest and most mature of any HP28/48/49/50 style UserRPL interpreters.

What makes it so awesome is its ability to interface with the OS via POSIX compliant commands -- it's almost like using your HP48 as a scripting tool for Unix.Too bad they didn't mention it.

jdb2

echo is good enough for me.

[girlintraining]

#!/bin/bash
echo `$[2+2]`

Re:echo is good enough for me.

[Gerald]

My favorite Linux calculator:

function math
{
        echo "scale=2 ; $*" | sed -e "s:x:*:g" | sed -e "s:,::g" | bc
}

$ math 12,147.2 x 3
36441.6

Re:echo is good enough for me.

[pintpusher]

bc++

Re:echo is good enough for me.

[Anonymous Coward]

Nice, but a couple recommendations:

1. sed can take multiple -e expressions on the command line, so you can turn those two statements into one.

2. You might want to run "bc -l" which also adds a bunch of extra functions, such as trig, natrual log, etc. it also defaults to scale=20, but you're overriding that yourself.

Re:echo is good enough for me.

[eliphas_levy]

Still can be simplified with s/echo/sed/ if you cheat with an input file :)

IIRC I use something that could be translated to:
$ function math { sed "{s|^.*|scale=2; $*|; s|x|*|g; s|,||g;q;}" /proc/uptime | bc -l; }

As always, there is a bunch of ways of skin the cat with shell commands.

Missing

[bcmm]

$ dc

Re:Missing

[John Hasler]

> $ dc

And bc.

Stick with the classics

[Denis Lemire]

'dc' is the only calculator you'll ever need!

Re:Stick with the classics

[ascari]

Desktop calculators are for losers, and here's why:

Everybody knows that all real engineers use a slide rule.

In those rare cases where a slide rule doesn't quite cut it you have a perfectly good excuse to go and ask the hot intern three cubicles down for help. Chicks dig guys who can admit a shortcoming, and who dare ask for help when they can't quite figure it out. It's an almost idiot proof way to score, second only to having pictures of nephews/nieces/puppies on your desk. So get yourself a slide rule and the rest is up to you: What has your desktop calculator done for you lately?

Re:Stick with the classics

[hedwards]

Bah, real engineers use an abacus. Quite a bit quicker once you get the hang of it too.

Re:Stick with the classics

[MichaelSmith]

Slide rule thanks. I have two. A good one which my dad saved from days past; and a cheap one from (believe it or not) a gift shop, about 10cm long.

Re:Stick with the classics

[quickOnTheUptake]

we all know expr is infinitely faster once you learn how to use it.

Re:Stick with the classics

[yuri benjamin]

You had abacuses? In my day all we had was little piles of pebbles. And we were thankful for them! Now get off my lawn!

Re:Stick with the classics

[Z00L00K]

I prefer 'bc'. Maybe because I like the "Before Christ" association.

Re:Stick with the classics

[The_Wilschon]

All we had was sets. And if we complained, von Neumann came and beat us with his curly braces!

For most people ...

[tomhudson]

anything more complicated than adding a few numbers, it's easier to open a spreadsheet than to learn how any particular calculator functions.

Re:For most people ...

[bcmm]

anything more complicated than adding a few numbers, it's easier to open a spreadsheet than to learn how any particular calculator functions.

Or "anything more complicated than adding a few numbers, it's easier to open a calculator than to learn how any particular spreadsheet functions".

That's really just a fancy way of saying that you are familiar with a spreadsheet, and not with a calculator program.

Re:For most people ...

[retchdog]

Or for those who already use the R package, that's also nice. When dc doesn't cut it, I fire up R.

Also, this never clicked for me, but it seems like a good idea: http://dirk.eddelbuettel.com/code/littler.html

Re:For most people ...

[tomhudson]

no, it's saying that the interface to most of the calculators is a royal pain in the ass, and most people won't be sure that they've gotten the right answer, because they ARE a royal PITA to use - no two are alike beyond the basics.

This is a problem in the "real world" as well - nobody uses most of the functions on their fancy non-virtual calculators because it's a real PITA to figure them out.

Re:For most people ...

[drinkypoo]

That's really just a fancy way of saying that you are familiar with a spreadsheet, and not with a calculator program.

Maybe it's just a way of saying that he thinks visually instead of mathematically, and finds it more useful to see an entire spreadsheet with notes and information, where he can look at parts of the formula at once.

I'm math-challenged, and find the spreadsheet easier to use, too. It's not because I know Excel better than bc. I don't really know either worth a shit. It's because I think visually, and not numerically. My hat's off to you if you can think numerically, but I haven't learned how yet.

Don't be so sure you know everything.

Re:For most people ...

[Evil Shabazz]

Perhaps at first blush, your reaction might seems right. But it only takes the a slight bit of thought to realize that most calculator interfaces, regardless of OS, mimic those of a physical world calculator - that is to say they don't take advantage of the extended output options of a desktop display and require you interact with numbers in the same archaic way as the physical models. A spreadsheet, on the other hand, allows you to keep all kinds of numbers all over the place for quick reference in addition to allowing you to quickly apply similar functions across different sets of numbers or more easily write more complex mathematical equations. Even when you calculator app allows that kind of thing, it's almost always more complicated to learn than its worth, and is way more easily done by just popping open a spreadsheet. As the parent said, anything more complicated than just adding a few numbers; I'm inclined to agree.

Re:For most people ...

[HouseOfMisterE]

anything more complicated than adding a few numbers, it's easier to open a spreadsheet than to learn how any particular calculator functions.

Not if you use the Google Apps spreadsheet program. It doesn't round by default and gives weird results to some of the simplest math. Play around with it a bit and you'll see.

Re:For most people ...

[nacturation]

They said the same thing about the abacus.

Re:For most people ...

[Anonymous Coward]

This article just proved I am addicted to the internet. I'm reading about linux calculators and I don't even run linux.

Re:For most people ...

[tomhudson]

anything more complicated than adding a few numbers, it's easier to open a spreadsheet than to learn how any particular calculator functions.

Not if you use the Google Apps spreadsheet program. It doesn't round by default and gives weird results to some of the simplest math. Play around with it a bit and you'll see.

No can do. I've black-holed anything and everything from google because of their CEO's very public, and very stupid, stand against privacy. Besides, I wouldn't trust my private data to any advertising company. Who would be stupid enough to do that? (I know, there are businesses doing exactly that - which just goes to show how continued exposure to PowerPoint makes you stupid)..

Re:For most people ...

[crazybit]

My hat's off to you if you can think numerically, but I haven't learned how yet.

My hat's off if you can write an spreadsheet formula without thinking numerically. As fas as I know you can't define functions geometrically in any of the available spreadsheets.
Having a spreadsheet implies a graphic way of organizing/presenting the information, but reasonably numerical skills are always necessary for writing functions.

Also, there is a widely used (some people consider is somewhat of a standard) calculator language called RPL, supported by many handheld calculators and computer software (including GPL alternatives).

Re:For most people ...

[drinkypoo]

My hat's off if you can write an spreadsheet formula without thinking numerically. As fas as I know you can't define functions geometrically in any of the available spreadsheets.

I sincerely hope you're being deliberately obtuse. Being able to break a mathematical problem up into [optionally labeled] pieces makes the whole thing a lot more understandable to those of us who don't drink, eat, and breathe mathematics by choice. Is that really so hard for you to understand that I have to spell it out explicitly?

Having a spreadsheet implies a graphic way of organizing/presenting the information, but reasonably numerical skills are always necessary for writing functions.

But making sense of them again later is a hell of a lot easier if you've got it broken up into bits in a spreadsheet.

Also, there is a widely used (some people consider is somewhat of a standard) calculator language called RPL, supported by many handheld calculators and computer software (including GPL alternatives).

Oh good, suggest that math be made more baroque, that will help.

Re:For most people ...

[Anonymous Coward]

i saw your sig -- i'm curious where you fall on the gender spectrum? i'm a regular /. user, but i'd rather not post my username, as there are links to lots of places where my real name can be found, and potential employers might look me up on google and be a bit wary of hiring me (sucks, but that's how it goes...)

email? i'm erinxsquid@gmail.com =)

Re:For most people ...

[crazybit]

The theory of multiple intelligences says about the "Visual-Spatial" intelligence: "There appears to be a high correlation between spatial and mathematical abilities, which seems to indicate that these two intelligences are not independent. Since solving a mathematical problem involves manipulating symbols and numbers, spatial intelligence is involved.

You might think you are bad at math, and you can find spreadsheets easier, but it's not because you are a "visual person" - there are others visual-spatial's that will choose RPL more than a speadsheet.

Re:For most people ...

[insufflate10mg]

God, I thought I was the only one...

...though I took it more in a geeky sense. Like "wow, I haven't the slightest use for any of these, but they're calculators... on Slashdot.. so I'll read."

Re:For most people ...

[Garridan]

Join the fun! You can use a Linux calculator from your browser -- no Linux required!

http://sagenb.org/

Re:For most people ...

[Anonymous Coward]

And yet you used a gmail address.

Impressive.

Re:For most people ...

[honkycat]

Have a look at SpeedCalc then. It was my favorite when I was using Linux and since it looks like it may be available for Mac I'm going to check it out again. It had a clicky calculator interface, but could also be interacted with using typed keywords. Assigning variables was intuitive (just type x=... or whatever) and I thought really bridged the gap between a desk calculator and a full programming environment quite well.

I also recommend just using Python (for me usually via ipython). The data entry is not as obvious as a spreadsheet (though if you go via a text file and use loadtxt (either matplotlib or numpy function, I don't recall which) it's more repeatable. The tradeoff is it's easier to see the manipulations in code than it is in the spreadsheet (to me) versus easier to see/modify the data in the spreadsheet.

Re:For most people ...

[honkycat]

er, I meant SpeedCrunch. Serve me right for posting after 2am.

Re:For most people ...

[Haeleth]

But making sense of them again later is a hell of a lot easier if you've got it broken up into bits in a spreadsheet.

If your spreadsheets are logically organised and present your calculations in an easy-to-follow manner, then congratulations: you are unique.

I have had to maintain other people's spreadsheets in the past. They are not easy to make sense of. It's like the worst spaghetti code ever, but in 3D, and without even a linear flow of control, let alone any semblance of structure.

Convenient for one-off calculations and graphing? Yes, slightly more convenient that $SCRIPTING_LANGUAGE and gnuplot. Maintainable? Har har hardy har.

Re:For most people ...

[Anonymous Coward]

Thanks for missing the point.

Fact is that some(/most?) people have great difficulty in chopping up a calculation into parts. If you have to do this numerically, you could lose track of what you're doing (some calculators have history, but still). Being able to spatially separate these parts helps a lot of people tremendously, because you can go back anytime, to any part, and change what you want. Doing so with a regular linear numerical calculator is more difficult.

Re:For most people ...

[jonadab]

> I'm math-challenged, and find the
> spreadsheet easier to use, too.

I majored in math, and I find the spreadsheet easier to use for a lot of things too. Although, the calculator app opens faster, so I tend to use that when I just want to do a simple addition or such like. But for anything complicated, yeah, I'm generally going to go with the spreadsheet.

I think it mostly has to do with whether you already know (the basics of) how to use spreadsheets or not. If you don't happen to already know how to use spreadsheets, a calculator program is going to be easier to learn, because everyone under the age of fifty has used one of the physical real-world devices on which it's modeled.

Re:For most people ...

[bcmm]

nobody uses most of the functions on their fancy non-virtual calculators because it's a real PITA to figure them out.

For a certain value of nobody - looks like you just aren't the user such devices target. Try studying physics or maths. :-)

I have Casio fx-991ES. It's not a graphing calculator, but it does most things that one would expect from a scientific calculator (trig, powers, logs, etc. etc.), as well as definite differentiation and integration, vector and matrix maths and equation solving. I have genuinely used (not just played with) every button and the majority of the feature (decimal to octal conversion springs to mind as a feature I don't actually need).

Now, each to his own, but I have always found it easier to find and press the button with a square root symbol on it than to consult a spreadsheet's help system to find out what they've called their square root function (sqrt()? root()? squarert()?).

I'm not arguing against spreadsheets; they most certainly have their place, and I use them, but that place is things like doing stats on medium-small data sets, (or doing any maths repetitively on a medium-small set of numbers) keeping track of money (of course), and the like.

Between problems you solve with spreadsheets and ones that require the use of proper computer algebra systems, there is a large space which is still best served by calculator-style interfaces. For example, suppose you want do something simple like use the quadratic formula on something with too many fractions to be done mentally (at least by me). It's going to look very nasty in a spreadsheet, but can be done in less time than it take Maple to start up (yeah, extreme example). I'd use my calculator even if my computer was already booted, unless I already had a CAS running.

My calculator also has the advantages that it will actually lay it out pn-screen (roughly) as if on paper, which makes mistakes more obvious, than they would be in the mess of brackets a spreadsheet formula would have (multiple brackets with no syntax-highlighting - hell), and will give me the answers as exact fractions or surds if possible.

Re:For most people ...

[TheRaven64]

He might not be unique, he might just be one of the people that followed the Improv-clone path instead of the VisiCalc-clone path. I've never seen anyone badly screw up a spreadsheet in an Improv clone, but most of the world seems to have decided that VisiCalc clones are the way to go.

Re:For most people ...

[TheRaven64]

The Google Apps spreadsheet does all of the calculation on the client. That means it is using JavaScript which requires all numbers to have the same range and precision as 64-bit IEEE floating point numbers. While most spreadsheets have various ways of representing numbers, including binary coded decimal and arbitrary-precision, everything in Google's one is a double. If you use it for anything important, you are an idiot, but it's quite handy for simple things.

Re:For most people ...

[The_Wilschon]

You might also prefer LabView, for visual arrangement of calculation and even programming flow. I didn't particularly like it for some specific technical reasons, but I can see where someone as visually oriented as you say you are might really benefit from it.

Re:For most people ...

[tomhudson]

Just go to the site :-)

Where's DC/BC?

[Bananatree3]

DC or BC are more than adequate, are already in 99% of the distros out there and are chock full of features!

Re:Where's DC/BC?

[drinkypoo]

bc is crippled by the past in classic Unix style. Why does scale default to 0? Because otherwise you break things.

Re:Where's DC/BC?

[John Hasler]

> DC or BC are more than adequate...

But they (shudder, moan, recoil in fear) involve the *COMMAND* *LINE*!!!

Re:Where's DC/BC?

[Cow Jones]

bc is crippled by the past in classic Unix style. Why does scale default to 0? Because otherwise you break things.

alias bc='bc -ql'

Re:Where's DC/BC?

[michaelmuffin]

hoc, the successor to bc, does not have this particular annoyance

Re:Where's DC/BC?

[amRadioHed]

What does it matter what the default is? Put your preferred scale in your .bcrc file and be done with it.

Re:Where's DC/BC?

[Anonymous Coward]

I agree and use it all the time, but orpie is so much nicer.

Re:Where's DC/BC?

[Cow Jones]

From your comment you apparently think I am a fucktard.

I don't know how you arrived at that conclusion. FWIW, I agree with you that the bc defaults aren't what people usually want. All I did was point out a simple way to make bc more useful; I've had that line in my .bashrc for ages. If you'd rather complain than use an alias, that's fine by me.

CJ

Re:Where's DC/BC?

[arth1]

The scale defaults to 0 because it's the sensible thing to do, considering that bc is usually called (as part of a pipe) in a shell script that only groks integers.

If you need a different scale, it's not that hard to set it. Even permanently for your user, or the entire system.

Re:Where's DC/BC?

[Anonymous Coward]

You can also set the environment variable BC_ENV_ARGS to include -ql, and thus not be dependent on a particular shell setup.

Re:Where's DC/BC?

[Anonymous Coward]

bc -l

There. Problem solved.

digg

[larry bagina]

why no digg button? Is it because they already posted this story and have that dupe thing figured out?

Re:digg

[Anonymous Coward]

No, it's because Slashdot has a policy of only posting stories at least a week after they've been posted to Digg. So there's no need to share on Digg, because Digg will already have covered any Slashdot story AGES ago.

Read Slashdot for the comments (especially because the stories are often flat-out wrong), read Digg for current news.

hp48

[tantrum]

I'm amazed they left out the hp48 emulator. It was an amazing calculator, and the emulator does exactly what it it is supposed to do - everything.

It did everything a calculator is supposed to do, and it was _almost_ able to boil my coffee.

After my 10 years working with programming, this is still the environment i love the most. Actually it is probably the only thing i still know the exact location of at all times.

I love beeing a geek :)

Re:hp48

[Overzeetop]

There was a disturbance in the force when Carly the Horrible 86'd the calculator line at HP, and collective sigh when they came back. I keep a sheet of 1/8" aluminum in the front of my HP48Gx soft case to protect the LCD when it's in there. I can't imagine engineering without it. I fully expect to have it buried with me when I die.

RPN FTW!

Re:hp48

[dasqua]

My hp48gx has been my calculator of choice ever since I first got it. Still works fine.

The benefit of this:
    quicker to use, 1 second startup
    portable, its a physical device
    easily upload results to the PC when needed

I also use python/SAGE...

Re:hp48

[cruff]

Here is the X48 emulator home page. I fire this up when I don't have my real 48SX with me.

Re:hp48

[dslbrian]

I've used x48 on Solaris and HPUX on many occasions, however I've found it to be somewhat unstable on recent linux releases (I suspect it's the GUI code, not the emulator itself). In any event I've found a more portable solution (aside from my actual HP50G). Emu48 skinned and running the roms for a HP50G makes for a very nice Win32 desktop calc (all the necessary bits are here, here, here, and refer to this first).

Simply placed on a USB drive the Emu48 install becomes portable. With a PortableApps install, and with a small bit of config editing on the PA side, the Emu48 directory can be dropped into the PortableApps directory and will integrate into the PA menu. Configured like that, you get portable, nice startup/shutdown, and it retains its memory between machines.

Re:hp48

[that this is not und]

I bought an HP50G, but I still can't give up my HP11C because there are just too many things hidden on the new key layout. swap x and y for instance should be right there as one of the single keystrokes. I'd love somebody to tell me 'you dummy, use this key' and have a fix for the problem. Every time I get out the calculator and scan to see what keystroke might work, I get angry. Then I get out the 11C and just use it. But the HP11C is so old that I worry. They're not irreplacable but they cost a fortune on eBay now.

Re:hp48

[ducomputergeek]

I have the 48G. Got it in middle school and still use it. And it spent 10 years being beat to death in my backpack. I remember in high school and college everyone had Ti, pretty much everyone. I remember people would ask to borrow it, couldn't figure out how to add 1+1, then give it back and never ask to borrow it again! It got me through every class in college except for finance (couldn't do Modified Internal rate of return and I went out and bought a HP business calculator that is in a box somewhere.

It survived 10 years of being in a backpack and still works to this day and I still use it. It's in my shachel. (It's not a man purse. Indiana Jones has one).

Re:hp48

[jeffreymsmith]

You dummy, use the right cursor key to swap the two most recent entries in the stack.

(Works on my 49g+; I would assume it's the same for the 50g.)

Re:hp48

[IAN]

You dummy, use the right cursor key to swap the two most recent entries in the stack.

That works only if you have already entered the number (expression, &c) on the stack (i.e., the command line is not active.) True x<>y works even if you're still in the command line, and it's unambiguously labeled on the keyboard. I dearly love my 48SX, but for pure number crunching any old RPN calculator (like the 11C mentioned above) is faster to use. Btw, "right cursor for swap" started with the 48SX, and on its keyboard it's quite obvious why: SWAP is right above the right arrow, left-shifted.

Re:hp48

[Anonymous Coward]

Thats beacuse RPN isnt the logical way to do things, no matter how much you and your shills scream it is. How can people think its easier to encode your math in RPN, when you can just type it in, and see the problem as it looks on paper, I value that way more then any stack, especially considering how easy it is to use your own "stack" via the variables. My TI-89 has lasted me years as well. Id bet money i can solve anything you can with my 89 and faster.

Re:hp48

[dslbrian]

I don't know why they didn't label the arrow keys on the 50G. I went through college on a 48SX, so I knew about some of the arrow key functionality, but I think only veterans of the 48 know such things. Likewise they failed to ship a full manual with the thing - I didn't even know until a couple years ago that there existed an advanced user guide on the 50G (normal guide, advanced guide - also CAS documentation). There is a ridiculous amount of hidden capability in the 50G, but they don't ship manuals for it, so nobody knows about it. And for anyone who has the 50G dropping keys on them, set the KEYTIME, another great documentation failure by HP.

As far as the swap, I'm not certain if what you call true x-y swap exists on the 50G, but if it does exist in the catalog you can probably map a soft or hard key to it. I would have to check the manual on how to do that, but I have done it before. I had to map a soft key to get back the '->Q' function that was a hard key on my 48SX (left-shift EVAL - converts a fractional number to a ratio of two integers, ex. .9225 ->Q gives 369/400).

Re:hp48

[Agripa]

I bought an HP50g recently to replace my aging HP48g and have no significant complaints. The only change I made was to shorten the keyboard de-bounce time. I guess the earlier run of HP50g calculators had marginal keyboards but mine performs very similarly to my HP48g.

Re:hp48

[texwtf]

The previous version of the "post-48" series was the hp49g+.

Absolutely godawful keys. Really horrible and also not great.

The hp50 is _much_ better, and I agree with other posters (on other forums) that it is the first calculator to be "better" than the hp48gx.

But it's not much better, and some things are less intuitive.

If you get one the bible (aka advanced user/programmers guide) from hpcalc.org is mandatory.

Re:hp48

[texwtf]

I am taking a circuits class with 20 other students, all of whom are using ti-89s.

doing AC steady state circuit analysis is loads faster with the hp.

for example, find the parallel equivalent for a 20, 30, and 40 ohm resistor:

HP:
20 inv 30 inv 40 inv ++ inv

ti:
1/(1/20+1/30+1/40))

11 keypresses vs. 19

When you get to complex numbers (inductors and capacitors) it's not even close. I finish calculations in well less than half the time it takes the TI users. And not because I'm some sort of superwhiz with the calculator- it just works better.

The TI is a good calculator, but you can't really appreciate the speed of RPN until you've taken the time to get practiced with it.

The HP also has an algebraic mode, fwiw, if you want the "how it looks on paper" effect. It can be helpful if you are working with a complex equation and want to make sure you have done it correctly.

So.. RPN isn't god's gift to calculation for everything, but it can be very handy in many situations.

Re:hp48

[texwtf]

"me too".

The documentation bites huge donkey balls. Even the "advanced" guides could use some help. It's a good command reference, but it doesn't help much if you don't know the command to do specific things.

The hp48 guides were light years better.

for example, enter in polar coordinates:

( 6 40 )

the (angle) is right shift alpha 6. Not labeled on the keyboard. ARGH!

Too many ARGH moments with the 50. Where's the ROT key? buried in some menu.

I'll probably like it more after I figure out how to customize the crap out of it.

Re:hp48

[Agripa]

Luckily for me my HP48g lasted until the HP50g became available. It still works but the keyboard has become somewhat unreliable.

The things that I like about the HP50g include the nicer display, SD support, USB support, 4 x AAA instead of 3 (I'm using low discharge NiMH cells without problems), and user replaceable lithium coin cell memory backup. My Yaesu FT-530 shares that last feature and it has really worked out. The keyboard arrangement has some differences which took me a few days to get used to but is acceptable. Any deficiencies there can probably be made up for with some customization which I have not gotten around to yet. I would have preferred full documentation in book form.

How about a symbolic calculator?

[Garble Snarky]

I still use the TI89 that I've had for almost 10 years, because to this day I have yet to find a desktop symbolic calculator that satisfies me.

I use matlab for work, and its command line interface to maple is decent. What I really want, though, is to somehow combine a command line interface with a nice typeset display - visually parsing the results is so much faster that way. Does such a thing exist?

Re:How about a symbolic calculator?

[Anonymous Coward]

Same here; although I've only had my Ti89 for a few years. What I really want in fact, is a good calculator application for my N900. You'd think there is some money in it, I paid a couple hundred dollars for my Ti89 and it's paid for itself many times over. And I'd gladly pay $100 for an app that allowed me to carry around my Ti89; and bonus points if it's even better (a nicer UI wouldn't be hard, and wolfram alfa style working out would be awesome)

Re:How about a symbolic calculator?

[Anonymous Coward]

A pricey but uber powerful version of this is mathcad, it runs on the same base as maple but has an excellent interface.

Re:How about a symbolic calculator?

[hairyfeet]

Don't know how good it will work, as I don't run Linux or have an N900, but here is a Ti emulator for Ti89-92+ that has Linux, Win32, and Mac ports. Maybe you can give it a spin and tell us how well it works? It wants GTK on Windows and I have no desire to install GTK.

Re:How about a symbolic calculator?

[friedmud]

Try using actual Maple instead of using it through Matlab. It has a pretty awesome GUI interface that allows you to enter equations either visually (like put down an integral symbol and fill out the fields) or using the command syntax...

  But either way it beautifully renders your math so it looks just like it does on paper. This is the reason I always reach for Maple over any other calculator.

Re:How about a symbolic calculator?

[Anonymous Coward]

It's called Mathematica. Kind of curious that you missed it when you mentioned the other two "big" math programs.

Re:How about a symbolic calculator?

[j_sp_r]

True. Also Matlab removed Maple from it's latest version and replaced it with their own software (Breaking a lot of things in the process)

Re:How about a symbolic calculator?

[Anonymous Coward]

I still use the TI89 that I've had for almost 10 years, because to this day I have yet to find a desktop symbolic calculator that satisfies me.

http://lpg.ticalc.org/prj_tiemu/

Re:How about a symbolic calculator?

[Garble Snarky]

I know its stupid, but I can't stand the camel-case naming conventions and square-bracket function calls in Mathematica. I can't get used to it. Clearly, a solution to this problem is not THAT important to me.

Re:How about a symbolic calculator?

[Anonymous Coward]

... combine a command line interface with a nice typeset display ... Does such a thing exist?

TeXmacs will connect to maxima and do exactly that, and much more. Best kept secret in free software.

Re:How about a symbolic calculator?

[stereoroid]

I've tried two on Linux: wxmaxima and GIAC/Xcas. The latter is a little more powerful, especially the CAS, but not as polished. Both are scriptable, and have a "copy LaTeX" option, so there's your typeset display right there. 8)

Re:How about a symbolic calculator?

[sam_nead]

Yes - it is called Sage http://www.sagemath.org/

Re:How about a symbolic calculator?

[Anonymous Coward]

wxmaxima? mathematica? you could also try and use texmacs as a frontend for maxima and others.

Re:How about a symbolic calculator?

[treeves]

I liked that about MathCad back in the day. Did Maple buy Mathcad?

Christ

[Blakey Rat]

Half the people on this site probably weren't even alive when Windows 3.1 came out... could you guys give the pointless Microsoft bashing a rest? Just once, ever?

Could we maybe just get over it instead of posting another "LOLZ Microsoft BOB is bad guyz!!"

This shit pisses me off.

Re:Christ

[the eric conspiracy]

Crikey. I had to use that cr*p. There is no way any company that put out a product like Windows 3.1 could ever be bashed enough.

Re:Christ

[dunkelfalke]

It wasn't that bad for the time. Worked even on low end hardware. OS/2 surely was much better, but also much more memory-hungry.

Re:Christ

[girlintraining]

Half the people on this site probably weren't even alive when Windows 3.1 came out... could you guys give the pointless Microsoft bashing a rest? Just once, ever?

Windows 3.1 was released in, what, 1993? 17 years ago. So the majority of slashdot posters are less than 17 years of age, by your reasoning. I find that unlikely as the median age in the US is 36.7 years of age. That is where most slashdot posters are located...

Re:Christ

[Ethanol-fueled]

You kinda make a good point. In my opinion, the difference between a calculator and a program is a GUI interface with buttons for numbers and functions. Almost everything mentioned in the discussion(but not the article) are command-line calculators and computer algebra systems.

Gnome's calculator is excellent for basic stuff. Compared to Windows, Linux is still severely lacking with the usability of Computer Algebra Systems - many of which are still commmand-line( in before 'Get off my lawn'), though some have very crappy GUI wrappers which open plots in new windows. That's one of my biggest pet peeves as a student and recent Linux convert. Needing 2 or three open windows(terminal, gui, plots) is too cluttered for those of us who are spoiled with things like Maple*. On the other end of the usability spectrum, there's the labyrinthine Sage, which requires running in Mozilla if you want a GUI! That caused usability problems with NoScript even before I started to use the damn thing, and now it won't even start again without tinkering. I might even make room for a Windows partition for Maple or Matlab, or run them in VMs.

* Yes, I know they make 'em for *NIX. They just might be the first Linux programs I'll actually buy.

Re:Christ

[dunkelfalke]

You forget that Slashdot has got a very specific target audience. There are lots of young folks here. When I started posting here more than a decade ago I wasn't even 20 myself (and back then Microsoft bashing even made some sense).

BTW Windows 3.1 was released in 1992.

Re:Christ

[Runaway1956]

Just get off the lawn, kid. Or, mow it. When we want your opinion, we'll offer you an ice cream.

Re:Christ

[westyvw]

You are right. Microsoft has gone on to design much more then faulty calculators. Lets move on. Now they design faulty spreadsheets.

Re:Christ

[uassholes]

Around 1985 I bought my fourth computer: an 8086 based system running Xenix, Microsoft's Unix.

Windows 3.1 was vastly inferior. They're back sliders.

Re:Christ

[the pickle]

Half the people on this site probably weren't even alive when Windows 3.1 came out

In which case, now would seem a very appropriate time to quote Santayana: "Those who cannot remember the past are condemned to repeat it."

Perhaps it would be instructive to have a link in TFS to something explaining exactly *why* the Windows 3.1 calculator was so deficient in its abilities. You know, for the kiddies who weren't alive back then.

p

Re:Christ

[Anonymous Coward]

Wasn't that the joke at the time? "What's the difference between Windows 3.11 and Windows 3.1?"

There is no difference. 3.11 - 3.1 = 0

Re:Christ

[Nimey]

1992. It's easy for me to remember that because that's the same year Wolfenstein 3D was released.

Also, you're making unwarranted assumptions about the demographics of Slashdot versus the general US population.

Re:Christ

[karnal]

I thought creepy old men offered kids popsicles, not ice cream.

That's what Family Guy taught me.

Re:Christ

[digitalhermit]

could you guys give the pointless Microsoft bashing a rest? Just once, ever?

No. Users of Microsoft product have a shared experience. It's no different from that knowing wink when I talk to other Florida residents about Hurricane Andrew. Or comisserate with a Cubs fan. Or talk about the most recent inanities of the most recent PHB with another cubicle dweller. Windows is our shared hell, our Inferno. We could no more stop talking about its pains than we could stop complaining about taxes or the latest government screwup. Indeed, I could holler over my cubicle wall, "Remember Code Red?" or "Just like Nimda" and four people will join in a collective groan of agreement.

Re:Christ

[the eric conspiracy]

It was terrible for it's time. It was released in 1992 and had no IP stack. Its competitors were things like Apple's System 7.

 

Re:Christ

[Blakey Rat]

Since System 7 didn't have a TCP/IP stack, you picked something of a bad example. :) MacTCP wasn't standard equipment until much later, either 7.5 or 7.6.

I think you have rose-colored glasses on, frankly. Every OS sucked back then, to pretty much the same degree.

Re:Christ

[Anonymous Coward]

Careful. We'll take away your Linux and force you to use Xenix the rest of your life

Re:Christ

[CAIMLAS]

I don't think the realization has sunk in that Microsoft may have turned over a new leaf has sunk in for many people yet.

I didn't say they had turned over a new leaf, just that they may have. The vote is still out: will Windows 8 (or whatever) suck, or are they actually going to start to consistently make good hardware?

Windows 3.1 through XP all sucked, horribly. The few improvements they made were long available elsewhere, and there were many problems with each of them which they never bothered to fix. They've got two decades of shit products to live down; I think a healthy dose of skepticism is necessary.

Re:Christ

[l3v1]

Half the people on this site probably weren't even alive when Windows 3.1 came out... could you guys give the pointless Microsoft bashing a rest? Just once, ever?

Age is not an excuse for ignorance. What's crap it's crap, even time can't change that, also, short memory isn't always a good thing.

Re:Christ

[Anonymous Coward]

Every OS sucked back then, to pretty much the same degree.

Shhhh! Don't let the OS/2 or SunOS zealots hear you say that!

Re:Christ

[Garridan]

True, you need a browser to use Sage's GUI, but you certainly mustn't run Mozilla. It works in every reasonably modern browser, including Chrome (despite my opposition to supporting beta browsers) -- hell -- Willam Stein even uses from his iPhone. If you can't get it to run under NoScript... that's because it uses lots of javascript, and you've disabled it. It's like complaining that your car gets bad gas mileage after you let the air out of the tires!

However, I must disagree with your description of Sage as "labyrinthine". It's got gobs of documentation that's easy to find and navigate, and it's a Python, which is ridiculously easy to learn. Furthermore, the community is large, and it's easy to get help, both on irc and over the mailing lists.

(disclaimer -- I'm a Sage developer, and I love the notebook)

Re:Christ

[lordlod]

I don't think the realization has sunk in that Microsoft may have turned over a new leaf has sunk in for many people yet.

How many times can a leaf be turned over?

Since it's a leaf analogy surely you should realize after it's been turned over once that any future leaf turnings and just going to result in the same old shit.

Re:Christ

[Blakey Rat]

Windows 3.1 was released in, what, 1993? 17 years ago. So the majority of slashdot posters are less than 17 years of age, by your reasoning. I find that unlikely as the median age in the US is 36.7 years of age. That is where most slashdot posters are located...

Yes. Obviously I was posting a well-researched and indisputable fact and not at all engaging in hyperbole.

By the way, you might (correctly) initially read that last sentence as sarcasm... but considering your ignorance of literary devices, I doubt it.

Re:Christ

[that this is not und]

I remember Code Red. I worked at a place that for some reason was a PepsiCo test market. So we had Code Red in the vending machine very early on. I drank a ton of it back then.

The place I worked also had a competent (very competent, the source code for implantable medical devices races around on that network) IT staff. Never heard of whatever 'code red' you are talking about.

Re:Christ

[Anonymous Coward]

Half the people on this site probably weren't even alive when Windows 3.1 came out...
Several of us were teenagers when Amstrad released zx80. Too bad you can't remember the 48K limit of the Z80 CPU (due to 16K Basic ROM).
Microsofts products didn't really bacame better over time you know, Microsoft is just hiding the limits behind better hardware.

Re:Christ

[MrMr]

So the majority of slashdot posters are less than 17
I think the parent was referring to mental age. So below 17 should be spot on.

Re:Christ

[TheRaven64]

I was alive when Windows 3 came out and I remember the calculator well, because it didn't fit in with the rest of the environment. Unlike every other app, which had been rewritten to use the new beveled buttons, the calculator used the flat ones from Windows 2 that just flashed black when you clicked on them, rather than having a push effect. Oddly enough, it was not the same calculator that shipped with Windows 2, so someone had tweaked the UI but not thought to upgrade it to the newer controls.

Windows 3.1 and NT 3.x came with the same calculator and it wasn't until Windows 95 that MS updated the UI. The OS X calculator was pretty useless until they added the programmer mode around 10.4 or 10.5. In this mode it will display any number in binary under the main display and let you toggle individual bits by clicking on them. This is amazingly useful when debugging. It also supports RPN.

Re:Christ

[Blakey Rat]

Indeed, I could holler over my cubicle wall, "Remember Code Red?" or "Just like Nimda" and four people will join in a collective groan of agreement.

Yeah, remember when we were hit by these viruses using flaws that had been patched months before because our IT department is full of incompetent morons? Remember how a simple firewall would have completely neutralized the threat to our SQL servers?

Seriously, that says nothing about Microsoft. It does say something about your IT department.

Re:Christ

[dunkelfalke]

OS/2 also lacked a TCP/IP stack out of the box back then. And frankly, while I was a fan of OS/2, it had its share of problems.

Re:Christ

[Anonymous Coward]

Users of AmigaOS would disagree with you.

Re:Christ

[mvdwege]

No. Those flaws do say something about Microsoft. Just about any flaw exposed in their products is of the same type: functions accepting outside input without validation. The fact that so many flaws have turned up over the years says outright that you guys are sloppy coders.

Mart

Re:Christ

[Luyseyal]

"Remember Code Red?"

Ugh, that was the reason I ended up ditching Time Warner for Ma Bell. Roadrunner decided to block inbound port 80 for, like, forever due to Code Red. I complained a number of times. At first all I heard was "Huh, what's that?" Eventually, I got to people who said that was to protect people from Code Red. After awhile, they said it was an "enterprise service" and I could get a commercial account if I wanted to run a webserver from my home.

So I said "screw you" and got my 5 static IPs from SBC.

I have since heard that inbound 80 is no longer blocked on residential cable. However, they made me so angry I have decided to leave them behind. Uverse seems to get the job done, anyway.

-l

Re:Christ

[Urkki]

Users of AmigaOS would disagree with you.

To be more accurate: Those users of AmigaOS who enjoy meditation and are interested in virology would disagree...

Re:Christ

[the eric conspiracy]

MacTCP was introduced with System 6 (1988) and was replaced by Open Transport in System 7.5.

Of course by modern standards everything in the past sucked. But System 7 came with 32 bit memory addressing, a TCP/IP stack and UI that actually was actually reasonably useful.

Re:Christ

[Blakey Rat]

Yes, but MacTCP never shipped *with* the OS. You could get it from Apple's FTP site, or from your local friendly retailer, but the fact remains that System 7 didn't have a TCP/IP stack. (7.5 did, in the form of Open Transport, as you say.)

What's the deal with commenting on posts right before comments are closed? Are you trying to make sure nobody else can get the last word?

Another vote for BC

[russotto]

Yep, it's "bc". If bc isn't sufficient, it's "bc -l". If even that won't do it, I move to sage.

And if you're younger than Windows 3.1... GET OFF MY LAWN.

I use bc and like better than any GUI

[ls671]

I use bc and I like better than any GUI based calculator. Compiled with readline functionality, it just rocks in my humble opinion:

~$ bc
bc 1.06
Copyright 1991-1994, 1997, 1998, 2000 Free Software Foundation, Inc.
This is free software with ABSOLUTELY NO WARRANTY.
For details type `warranty'.
scale=5
(2*80/3.333)^3
110625.18091
((2*80/3.333)^3)/21
5267.86575

man bc for details

Re:I use bc and like better than any GUI

[ls671]

I forgot to mention that I use it in scripts too:

~$ (echo scale=5 ; echo 22/7) | bc
3.14285
~$

Calculators: useless; Languages: useful

[MacTO]

Calculators are useful as handheld devices, but you may as well use an interpreted programming language if you're on a computer. That is particularly true if you consider yourself a Unix user. So my favourites are:

bc: fast to use, arbitrary precision, and it seems to be universally available
awk: faster to use when you are performing the same calculation many times over
python: has a richer library of mathematics functions

Re:Calculators: useless; Languages: useful

[maxume]

The only issue with Python is that it is awkward to use decimals. Sure, floats are good enough most of the time, but not all of the time.

I use frink when I don't want to think about representation problems:

http://futureboy.us/frinkdocs/

The support for units isn't nearly as cheesy a feature as you would think.

Python

[petscii]

I gave up on all the software calculators a few years back. Now I use python almost exclusively. In fact if I could get a Ti-82 form factor that ran nothing but Python I'd be happy.

This was a natural progression, not something that I forced myself to do. I really like how you can create the rules of the universe and make your own python modules and re-use them. I've done this for my DC and AC theory classes. As well as microprocessor.

The only downside to this is when I need to give the data to someone else to look at. Now I can export to XLS via python as well.

Re:Python

[hoytak]

I'll second that! Python makes calculating fun again.

Re:Python

[spiffmastercow]

I remember in college needing to plot parametric equations to print out, and not knowing how to do so in any available software. so I wrote my own plotter in Python in roughly an hour.. Powerful language.

Re:Python

[DEmmons]

yeah, i'll admit that python is the only calculator i use anymore, and i have gotten lazy enough to use it for a lot of things any really smart person would be able to compute mentally in a second or two. python is fun and makes sense to me.

Nonpareil

[thecross]

http://nonpareil.brouhaha.com/ Calculator sim. I think one of the older versions can emulate the HP 15C.

Re:Nonpareil

[metamatic]

Mentioned in TFA.

Emacs Calc

[macshit]

Emacs Calc, i.e. "M-x calc" in Emacs is by far the best calculator I've ever seen.

Here's the blurb from the manual:

"Calc" is an advanced calculator and mathematical tool that runs as part of the GNU Emacs environment. Very roughly based on the HP-28/48 series of calculators, its many features include:

• Choice of algebraic or RPN (stack-based) entry of calculations.
• Arbitrary precision integers and floating-point numbers.
• Arithmetic on rational numbers, complex numbers (rectangular and polar), error forms with standard deviations, open and closed intervals, vectors and matrices, dates and times, infinities, sets, quantities with units, and algebraic formulas.
• Mathematical operations such as logarithms and trigonometric functions.
• Programmer's features (bitwise operations, non-decimal numbers).
• Financial functions such as future value and internal rate of return.
• Number theoretical features such as prime factorization and arithmetic modulo M for any M.
• Algebraic manipulation features, including symbolic calculus.
• Moving data to and from regular editing buffers.
• Embedded mode for manipulating Calc formulas and data directly inside any editing buffer.
• Graphics using GNUPLOT, a versatile (and free) plotting program.
• Easy programming using keyboard macros, algebraic formulas, algebraic rewrite rules, or extended Emacs Lisp.

That list gives you a bit of an idea, but doesn't really capture how just darn cool Calc is; it just seems to do everything.... (The things I particularly value are the vector/matrix operations and the symbolic manipulation operators.)

It's (default) model is HP-style RPN, except of course with a much larger visible stack, and multi-level undo.

[You have to be careful tho because recent releases of Emacs come with two calculators -- a "simple" one, which you get with "M-x calculator", and the super incredible one you get with "M-x calc"... (yes it's kind of silly, but as usual with Emacs, there are historical reasons...]

Re:Emacs Calc

[godrik]

TFA was talking about the Linux operating system and not the emacs operating systems...

Re:Emacs Calc

[chelberg]

There's a difference?

Re:Emacs Calc

[chelberg]

I just tried calculating log(500!) and it crashed emacs! Arbitrary precision up to its internal limits.

Re:Emacs Calc

[kinthalas]

<'log(500!)>

Computation got stuck or ran too long.  Type `M' to increase the limit

<M>

max-lisp-eval-depth is now 2000

<'log(500!)>

2611.33045846

Re:Emacs Calc

[siride]

That's always true. It's not infinite precision. There's no such thing.

Re:Emacs Calc

[JustOK]

You're exactly right.

Re:Emacs Calc

[chelberg]

Crashed on my mac :) In any case, when emacs crashed for me it was well below the available memory limits on my machine. Not sure why it crashed though.

Re:Emacs Calc

[chelberg]

Perhaps I need to get an iPad to calculate this -- lol

Re:Emacs Calc

[that this is not und]

There's a big difference.

Linux is now a bloatware OS. You won't run it productively on a 486 with 16 megs like you could back in 1998. You've got Kay Dee Eee, and as much layering and obfuscation as the egos of a 'Distro creator' can cram in there. It isn't like Slackware anymore.

Wait... emacs is a bloatware editor.

The difference being, of course, that emacs still runs really great on a 486 with 16 megs.

Re:Emacs Calc

[TheRaven64]

Linux is one of the bootloaders that some people use to launch the EMACS operating system. You can run EMACS without Linux, by using a different loader. You can also run Linux without EMACS, but it's not particularly useful.

Re:Emacs Calc

[jirka]

genius> log(500!)
= 2611.33045846
genius> log(5000!)
= 37591.1435089
genius> log(50000!)
= 490995.24305
genius> log(500000!)
= 6061189.16882
genius>

Sig figs?

[GrouchoMarx]

the calculators listed below are of precision quality

To how many significant digits? We need to know what level of precision we're working with.

Re:Sig figs?

[Anonymous Coward]

And what the hell does "precision quality" even mean?

Re:Sig figs?

[John Hasler]

> We need to know what level of precision we're working with.

With bc and dc, infinite.

Re:Sig figs?

[MichaelSmith]

> We need to know what level of precision we're working with.

With bc and dc, infinite.

Thats some computer you have there.

Re:Sig figs?

[MrMr]

bc, dc and the like are arbitrary precision calculators:
scale=1000
sqrt(3)
1.732050807568877293527446341505872366942805253810380628055806979451\ 93301690880003708114618675724857567562614141540670302996994509499895\ 24788116555120943736485280932319023055820679748201010846749232650153\ 12343266903322886650672254668921837971227047131660367861588019049986\ 53737985938946765034750657605075661834812960610094760218719032508314\ 58295239598329977898245082887144638329173472241639845878553976679580\ 63818353666110843173780894378316102088305524901670023520711144288695\ 99095636579708716849807289949329648428302078640860398873869753758231\ 73178313959929830078387028770539133695633121037072640192491067682311\ 99288375641141422016742752102372994270831059898459475987664288897796\ 14783795839022885485290357603385280806438197234466105968972287286526\ 41538226646984200211954841552784411812865345070351916500166892944154\ 80846071277143999762926834629577438361895110127148638746976545982451\ 78855097537901388066496191196222295711055524292372319219773826256163\ 14688420328537166829386496119170497388363954959381

If you're a mac user

[moosesocks]

Slightly OT, but if you're a Mac user, I highly recommend the PEMDAS Widget, which is just that tiny bit more powerful than a typical desktop calculator to make it 100x as useful.

HP42S Simulator

[Anonymous Coward]

My vote is for Thomas Okken's HP42S simulator. My favorite calculator from university days is resurrected!

Use Google

[chipschap]

Google's quick calculator does a lot more than I ever realized before I started using it for quick one-liners.

Re:Use Google

[benjamindees]

Totally agree. Google calculator is one of the most awesome things ever.

Google

[Anonymous Coward]

Just type your equation into Google. Works on Windows, Linux, Mac, iPad, BSD, etc.

Re:Google

[F452]

Google's calculator is just a front-end to "bc". :-)

Re:Google

[Anonymous Coward]

I've never known bc to handle something like this:

46 cubic kilometers per atmosphere in cubic rods per bar

Re:Google

[Haeleth]

But only for a limited range of equations. For example, it can only calculate factorials up to 170! -- after that it stops recognising them and just does a normal search instead.

And, yes, there are plenty of cases where one might want to know the factorial of a larger number.

Re:Google

[Rockoon]

Still looking for a calculator with a Primorial operator.

SpeedCrunch

[Cow Jones]

I've really come to like the SpeedCrunch calculator, which is available as a Debian package, and (according to their website) also runs on Windows and Mac. It's probably not inteded for scientific calculations, and it can't display graphs, but it has a very simple interface ideal for quick calculations. The tooltip with the current result of unfinished expressions is a nice touch, as is the history of past calculations (session).

CJ

Re:SpeedCrunch

[The Wild Norseman]

I use SpeedCrunch Portable on my flashdrive (for WinXP machines) and have used it for the past several years while in IT Skool. I never had to do much in the way of graphing so the lack of graphing with SpeedCrunch never bothered me.

Other than that, I still generally find it easier to just to type in to Google whatever simple math problem I'm needing to solve.

Re:SpeedCrunch

[enter to exit]

yep. I also found SpeedCrunch to be a top notch desktop calculator.
It has the best UI for those of us with butterfingers as the history is always visible

Use your head people

[Anonymous Coward]

I mean really, use your head.

Input: typically ears or eyes, or fingers if you are blind.
Output: fingers (written) or mouth (oral)
Processor: brain

Re:Use your head people

[Garble Snarky]

Can you use your head to list the first 4 zeros of the second-order Bessel function of the first kind? Neither can I, that's why I use something like matlab or octave.

Re:Use your head people

[thegrassyknowl]

Processor: brain

My head doesn't contain a brain! It contains a highly parallel learning neural super computer.

Re:Use your head people

[MobileTatsu-NJG]

I mean really, use your head.

Input: typically ears or eyes, or fingers if you are blind.
Output: fingers (written) or mouth (oral)
Processor: brain

Then you hike up your pants, knock something expensive over, and go "Did I do thaaaat?"

Re:Use your head people

[Haeleth]

Can you use your head to list the first 4 zeros of the second-order Bessel function of the first kind?

Yes. They're 0, 0, 0, and 0.

I can factor large primes in my head, too!

Great ones

[AlexWillisson]

I don't know about those in the article (never heard of any of them), but here's what I use: Emacs M-x calc, maxima and QtOctave. Gnuplot for graphs. Often Google or python shell for quick things, if I'm too lazy to open emacs calc. I'm surprised none of those were mentioned in article.

octave in an xterm

[Jeremy Erwin]

what more could you possibly want?

Re:octave in an xterm

[Anonymous Coward]

what more could you possibly want?

Something that's not a clone of matlab. Holy crap, I hate everything about matlab.

Re:octave in an xterm

[stereoroid]

I could want an Octave development environment, especially if it offered some of MATLAB's friendly features. I'd want something like OctaveDE, in other words.

Arbitrary precision math libraries

[spiffmastercow]

On a related note, does anyone know of a good arbitrary precision decimal math library, preferably for Mono/.NET? Everything I've tried seems to crap out on division of numbers in the range of 10^100000. So far I've had to use arbitrary precision integer libraries, then use the old fixed point math hacks that used to be common before CPUs had floating point support.

Re:Arbitrary precision math libraries

[Anonymous Coward]

Erlang has integer precision only limited by the computer's memory.

Re:Arbitrary precision math libraries

[spiffmastercow]

That would be fitting, since I'm trying to compute Erlang C values.. I was just hoping to find something that would't require a clunky integration with .NET, or require me to learn a new language.

Re:Arbitrary precision math libraries

[the eric conspiracy]

Gnumeric has Erlang functions.

Re:Arbitrary precision math libraries

[spiffmastercow]

That's a spreadsheet app though, right? I need a library function.

Re:Arbitrary precision math libraries

[The_Wilschon]

A quick googling ("libgmp .net mono") turns up the possibility that there may be a wrapper for libgmp in mono. Other than my google-fu, I know nothing of .net and mono.

Re:Arbitrary precision math libraries

[Anonymous Coward]

Well, what did you try? gmp and mpfr are some of the better maintained.

I might be biased, but...

[Enleth]

I prefer the Python interactive shell and GNU Octave (or any other Matlab-compatible environment, including Matlab itself) for numerical calculations, Asymptote for plots and other methods of data visualisation, Maxima when a CAS is in order and LaTeX to turn all the stuff generated by those packages into something readable and publishable.

Throw in some scripted links between all those tools, a few functions from Peter Acklam's Matlab Utilities, your favourite function for converting a matrix to a LaTeX table and saving it into a file in a single call, a few exec()-equivalents here and there, and you'll get a rig that auto-regenerates your report/publication/thesis/shopping list/whatever else you might have been doing, in a single run of a single program, should you spot a mistake somewhere deep in the calculations, or a typo in the input.

For one, I don't think I'll ever understand people who use spreadsheets. And copy their results to the word processor. And then spot a mistake in a formula, fix it and proceed to copy the new, correct results from scratch. And then spot a typo in the data.

Why biased? Well, I'm studying control systems and robotics. It's all about task automation. Besides, everything in this field involves using Matlab for something, and just about everyone in the academia (the technical side of it, at least) is using LaTeX, so you just kind of get used to using those two for just about anything after a while, and automating everything with scripts.

Of course, the above assumes somtheing more complicated than a few basic operations in a single line. We're talking about sophisticated calculators here. For simple tasks I'm just using Google...

Re:I might be biased, but...

[zippthorne]

Why biased? Well, I'm studying control systems and robotics. It's all about task automation

I don't see why you call that bias. A computer is a machine for performing repetitive tasks. If you find yourself doing the same thing over and over manually, you're doing something wrong.

Those spreadsheet people are silly. Especially since you have been able to link into a spreadsheet from the associated word processor in the same suite so that the changes propagate for over a decade.

Re:I might be biased, but...

[dargaud]

Talking about calculators, here's one simple thing I haven't manage to find a solution to, while it would seem a pretty obvious 'Unix philosophy' thing: how do you pipe a series of numbers from the command line and get a simple, auto-scaled graph ? For instance: cut -d" " -f1 access.log | sort | uniq -c | sort -nr | someplotcommand in order to get a histogram of the first column of numbers. I tried with gnuplot and various others, even asked on forums.

Re:I might be biased, but...

[Enleth]

An Asymptote script should be able to do that. It's both a data representation language and a programming language, similar to LaTeX in principle, but resembling C++ in syntax. It doesn't support shebangs , so you'd have to invoke the intepreter explicitly (or write some kind of a wrapper attached to the begining of a file), but it does provide a mechanism for reading from standard input, and any other file, for that matter. Take a look at the histogram.asy example file provided with Asmptote.

Exercise

[zogger]

OK linux calculator and math geeks, here's a question I have wondered about before. This is just for fun, show off your leet skillz. Start with the first released linux kernel, get the size, look at some major releases, etc, do your magic as of today's sized kernel, and give us the best guess in your graph or projection when the kernel will reach or exceed one gigabyte in size, the release date as close as possible.

Sage?

[selven]

Why hasn't anyone mentioned sage yet? It is quite bloated for a calculator (it's intended to rival Mathematica, not MS Calc), but it does plain old arithmetic, calculus, equation solving, factoring and plotting (2d, 3d, 2d/3d implicit, complex, complex implicit) quite well.

Re:Sage?

[Anonymous Coward]

+1 for sage

it's been cleaned up a lot lately. it's incredibly powerful, with much better syntax than any other computer algebra system (python) and access to other computer algebra systems for what they're good at- calculus through maxima, group theory through gap, and that's just the parts i know about

Re:Sage?

[phatsphere]

+1 to Sage http://www.sagemath.org/ from me, too. You are always limited by these tiny utilites and you are not able to program with them easily like with Sage - since it is based on Python.

Huh?

[symbolset]

Are we worried now about the math results of people who can't figure out a calculator? Why?

Re:Huh?

[Filip22012005]

Because they run our economy.

Re:Huh?

[Toonol]

The usefulness does depend on the problem. Anything using sets or arrays larger than a few number is going to be much easier and faster (and less prone to error) on a spreadsheet. A spreadsheet is probably more useful for experimentation, as well. For typical evaluation of expressions, though, a calculator is probably much faster. The smartest thing to do is use whichever is more efficient for the task, where the efficiency is based on both your own skillset and the tool.

Re:Huh?

[Coryoth]

The usefulness does depend on the problem. Anything using sets or arrays larger than a few number is going to be much easier and faster (and less prone to error) on a spreadsheet. A spreadsheet is probably more useful for experimentation, as well.

I really do think it is matter of what one is used to. For me manipulating lists, arrays, sets, matrices, tensors etc. seems easiest and most natural in something like ipython (with numpy), mathemematica or matlab. Of course those are the sorts of tools I use every day, so it seems most obvious to me how to quickly make them do what I want. I rarely if ever use spreadsheets, and unsurprisingly I would find one a rather clunky experience. I imagine you use spreadsheets far more than say mathematica, and so you will instantly know how to do whatever it is you need in spreadsheet (while I will certainly not), but find using mathematica a clunky experience.

Sage?

[Anonymous Coward]

Try Sage (sagemath.org). It can be finicky to install, but is a great CAS! It also will integrate with Octave, Maxima, R, GAP, Python, and several other programs. The typesetting is great and the graphs look good too.

How about a construction calculator?

[3seas]

Anyone know of such a thing on linux?

Re:How about a construction calculator?

[Anonymous Coward]

Yes please! I used to sell a great but expensive hardware calc in the 80s that made conversion between metric and imperial painless. I'd love to have that in software.

Re:How about a construction calculator?

[John Hasler]

What is a construction calculator?

Re:How about a construction calculator?

[Anonymous Coward]

A godsend for people in the building trades. Google around a bit for a better idea. For one thing it makes dealing with feet-inches-fractions painless, handles metric conversion /easily/, and usually has a handful of common formulas.

Re:How about a construction calculator?

[maxume]

It has some notion of the units present on a tape measure, and functions helpful for a builder:

http://www.amazon.com/dp/B0008GO6A6/ref=asc_df_B0008GO6A61018580?smid=ATVPDKIKX0DER

It's all about the tape!

[markdavis]

The Linux calculator we use at work is gtapecalc: http://gtapecalc.sourceforge.net/

It is oldler, but a great business calculator. The best feature it has is the ability to emulate a calculator WITH A PRINTOUT TAPE! So you can see everything you did, edit those numbers, add comments, even print the "tape".

Re:It's all about the tape!

[innocent_white_lamb]

You can find a gtapecalc srpm (and binary) that will compile cleanly on newer Centos and Fedora versions here: http://www.melvilletheatre.com/articles/el5

Re:It's all about the tape!

[cyberpunkrocker]

A few years back I was very surprised to find that there wasn't any 'Tape Calculator Emulators' for Linux. I seem to have missed this; thanks for bringing this to attention! (Although it seems to be more-or-less abandoned project... I wonder whether it compiles on modern Gentoo?).

The tape calculator I have used for some time is SuperbCalc: http://www.mariottini.net/roberto/superbcalc/. It is java-based, so it will run on all OS'es.

Re:It's all about the tape!

[Anonymous Coward]

Awsome. I have been looking for a calculator or calc program that will do this for years. Why the fancy HP-48/49/50 etc with their unlimited stack dont have a mode for this is beyond me. Wonderful for error checking, to have that tape you can refer back to. I don't see any accountant/bookeeper style person doing without it.

Re:It's all about the tape!

[Tzarius]

I've found Console Calculator to be quite handy in the office environment, although it is (currently) windows only. The "virtual tape", minimize-to-tray, responsiveness, and absence of the space-devouring visual number pad help are great when you're working on another business app. I will be trying gtapecalc though!
http://www.zoesoft.com/console-calculator/

orpie

[cynyr]

i found orpie http://pessimization.com/software/orpie/ to be very useful during intro to AC circuits back at university. it handles input in both polar and rectangular notation and will output the answer in either. a fast RPN calculator. The only thing i haven't found is a calculator to do symbolic algebra, as in solve things in terms of sqrt(2)'s and such for me.

Not much of a review

[physburn]

And if i want a reverse polish calculator, i would have installed FORTH. I would guess from the text, that extcalc was the best. They could have spared more than one line for each of 7 programs.

---

Linux Feed @ Feed Distiller

Re:Not much of a review

[funwithBSD]

I miss my hp41cv, lost in collage dorms. *cry*

Re:Not much of a review

[Anonymous Coward]

Free42 is a complete emulation of the HP42S, which is essentially a (thinner) portless HP41CV. Or try the i41CX+ for your iPhone/iPod Touch - it's non-free, but well worth the cost.

Python+Numpy+Scipy+Matplotlib

[acheron12]

What more could you need? (Acceptable answer: Sage?)

Re:Python+Numpy+Scipy+Matplotlib

[Anonymous Coward]

Sympy, for symbolic mathematics.

RPN Better than algebraic?

[mrcaseyj]

I've heard that RPN calculators save keystrokes, but my precalculus book listed RPN and algebraic keystrokes for various problems, and it seemed that the savings was only due to unnecessary keystrokes with the algebraic version. Looking at several examples in that book, I couldn't find one where an efficient user of an algebraic order calculator couldn't do the problems in just as few keystrokes. Could someone give an example of a problem where RPN uses fewer strokes than an algebraic order calculator (including what strokes are needed on the RPN)?

Re:RPN Better than algebraic?

[jdb2]

Could someone give an example of a problem where RPN uses fewer strokes than an algebraic order calculator (including what strokes are needed on the RPN)?

One big example is continued fractions. For example : 2 INV 2 + INV 2 + INV 2 + INV 2 +...... approximates the square root of 2.

The algebraic method would involve this unweidly and ugly expression : 1 + 1/(2+1/(2+1/(2+1/(2+1/(......)))))

jdb2

Re:RPN Better than algebraic?

[adonoman]

Any time where the order of operations that you want doesn't coincide with the "natural" algebraic order:

Algebraic: (2 + 4) * (5 + 6)
11 keystrokes

RPN: 2 4 + 5 6 + *
7 keystrokes

Re:RPN Better than algebraic?

[mrcaseyj]

Algebraic: (2 + 4) * (5 + 6)
11 keystrokes

RPN: 2 4 + 5 6 + *
7 keystrokes

By my count:
Algebraic: 2 + 4 = * ( 5 + 6 =
10 keystrokes

RPN: 2 Enter 4 + 5 Enter 6 + *
9 keystrokes

Although it is only a single stroke on this problem, there does seem to be an advantage in keystrokes. I think there would be an additional keystroke saved on each additional sum in parenthesis that you tacked on to the product.

Re:RPN Better than algebraic?

[Schraegstrichpunkt]

For me, it's not so much saving keystrokes as it is minimizing errors and not having to count parentheses. Also, RPN maps better to how I think about a problem than algebraic entry does.

Re:RPN Better than algebraic?

[mobets]

Missed a few, but RPN is still fewer keystrokes.

Algebraic: (2 + 4) * (5 + 6) [ENTER]
12 keystrokes

RPN: 2[ENTER] 4 + 5[ENTER] 6 + *
9 keystrokes

Re:RPN Better than algebraic?

[that this is not und]

One of the biggest problems with Alg calculators is that the = key is also a 'clear everything' key. Basically there's no stack for intermediate results. Unless you take the extra steps of storing to a 'memory' location.

Re:RPN Better than algebraic?

[Rich0]

One or two others have hinted at this, but I find that RPN just seems much more natural. The fact that you have a stack means that you can attack a problem in almost any order, without really any sacrifice of keystrokes or hacks like the "Ans->" key or memories or whatever.

Want to start on the outside of a big equation and work your way in? No problem - although you'll have to keep track of a few values on the stack (usually not a big deal unless the expression is very unwieldy). Want to start on the inside and work your way out? That is trivial.

When I see people working with normal calculators and they need to capture intermediate values I often see lots of rounding and re-entry. With an RPN calculator I can see the intermediate values, and yet keep them at full precision without any need for memories/etc.

Seeing all the intermediate values often is useful in real-world situations, as well. For example, often you'll run into equations in real life where some value of interest is a sum or product of lots of components (each of which is the result of a short calc). With an RPN calc you just perform each short calc and end up with a stack full of values, and then you hit + or * 10 times or whatever to sum/product the whole list. However, before you do that you can easily look at the list (or just watch the stack collapse and see the values as you use them). This gives you a quick idea of how the various values contribute to the whole.

When I use infix calculators I find myself having to plan ahead a lot more to do a calculation, or writing out big long strings of math. RPN just fits how I think better.

Re:RPN Better than algebraic?

[maxume]

Cheap Casio scientific calculators that I used did have a stack.

press 5 = and 5 would be in the stack.

Press + 5 = and you would get 10.

Press = again and you get 15. Press = again and you get 20.

I'm not familiar enough with RPN to make any great comparison, but I imagine the Casio wasn't quite as powerful.

Re:RPN Better than algebraic?

[TheRaven64]

I almost find RPN more natural, but it's too destructive. I'd like to have a calculator that stores the operations tree and lets me modify any part of it. But for everyday use, who uses a calculator app? I just use a system service that evaluates the selected text and appends the result.

Re:RPN Better than algebraic?

[Anonymous Coward]

Err, what's wrong with '2' followed by the square root button?

Re:RPN Better than algebraic?

[4181]

RPN maps better to how I think about a problem than algebraic entry does.

I've always assumed that to be one of the biggest sources of resistance to RPN. Those who are uncomfortable with math can use algebraic entry without thinking about the meaning of the problem.

Re:RPN Better than algebraic?

[Beat The Odds]

Which is why I still program in Forth..... http://en.wikipedia.org/wiki/Forth_(programming_language)

Re:RPN Better than algebraic?

[Anonymous Coward]

With an HP50g and the like you can type in the entire expression, evaluate it and re-edit it all you want.

If I don't have the calculator handy I just use python. What with the command line history it's just a matter of doing cursor-up and re-editing the line.

Re:RPN Better than algebraic?

[El_Isma]

Most modern calculators let you do something like this:
2 (ENTER)
2+1/ANS (ENTER as many times you want, and you get your result)
You require a calculator that lets you edit the expression for this to work.

Re:RPN Better than algebraic?

[pugugly]

Feh - I'm no mathematician by any stretch, but I scored 730 in Math. RPN is just entirely counter-intuitive - algebraic notation simply is better at helping (Me at least) formulate the problem.

RPN? - stick verbs at the end of sentence I can, but stupid people that do that look. Elitist kiss-asses they are.

Pug

Re:RPN Better than algebraic?

[Schraegstrichpunkt]

I should clarify: I find algebraic notation easier to read, but RPN easier to type. The HP graphing calculators are nice because they display in algebraic notation, but you can enter your equations interactively using RPN. So, as soon as I type "X ENTER Y +", it shows "X+Y". Then, if I type "2 /", it will show me "(X+Y)/2" (actually, it shows it like you would see in a textbook, with "X+Y" on top of "2" with a horizontal line in between them).

I looked on Youtube for some videos of the HP calculators, but they were long and boring. Instead, here's a nice video of an iPhone RPN calculator app. Of course, the HP calculators are much more sophisticated, but that video shows the general idea: You have a stack, and you perform operations on the stack.

However, I still tend to use algebraic notation on a full-blown computer, because I haven't found any good, free RPN-enter interfaces for Linux. dc and rpncalc are a bit of a pain.

Re:RPN Better than algebraic?

[4181]

I looked on Youtube for some videos of the HP calculators, ...

For a good video of the philosophy behind RPN, see Rocket Girls Ep.2:

What are the values of delta v1 and delta v2?

But some of those numbers have seven digits...

You have a calculator, don't you?

Right, right! ... the heck is this? There's no equals key!

Of course not! Don't you know reverse Polish calculation?

Of course not!

Oh, really? In that case ... I'll take five minutes to cram its operation into your head. When I'm done with you, you'll never be able to use a regular calculator again. Prepare yourself.

Re:RPN Better than algebraic?

[Schraegstrichpunkt]

Awesome.

Qalculate!!

[Hurricane78]

Nothing. I repeat: Nothing beats Qalculate!! (It’s so hot, Firefox’s spell checker suggests “Ejaculate”!)
If you go any bigger, you “explode” into a math suite, and not a calculator anymore.

Ignore the silly screenshot with the “button” view. Most calculator software tries to imitate a physical calculator, with buttons and LED displays. Which is just an EPIC FAILure in UI design. This screenshot shows a real usage example of Qalculate!:
http://navid.radiantempire.com/pub/Haskell-Synth-Entwicklung.png (The BG contains a Haskell programming editor and a sound analyzer. Yes, I’m still a noob at this. ;)

Re:Qalculate!!

[bill_mcgonigle]

Yes, I'm still a noob at this

When you hit F11, you'll see a Print Screen you'll see a 'Capture Mode' pop-up menu that you can change to just grab a window under cursor after n seconds.

Re:Qalculate!!

[bill_mcgonigle]

haha, way to edit cohesively there..... you hit 'Print Screen' of course (unless you've got odd remappings like this machine... sigh).

What about...

[Rufty]

octave???

Re:What about...

[Anonymous Coward]

Free42???

x48

[Anonymous Coward]

A full fledged HP 48g(x) emulator that actually looks like an HP48. It is not supported anymore, but it works great on my system still.

Anonymous Coward

[Anonymous Coward]

Clearly the most robust is Sage:

http://www.sagemath.org/

I think we have a winner....

[voss]

for the nerdiest linux topic on slashdot for 2010.

Re:I think we have a winner....

[MobileTatsu-NJG]

Agreed.

from the im-wearing-you-down-baby dept.

calc2 from The AWK Programming Language

[messdog]

The calc2 program from page 144 of my yellowed copy of The AWK Programming Language is all I have needed for 98% of the calculations I make. If I need a spreadsheet, I'll use one, but specifying the RPN on the command line and getting the answer right there is the most convenient. Nothing else to load and unload, just type a command and get an answer. It even supports named variables! I save it as rpn.awk on the *nix or Windows systems I spend any time on. I spent about 10 minutes adding percentage and modulus operators, sometime in the last 17 years or so. It works!

Linux Calendars

[FatherDale]

Sadly, I mis-read the headline and came here expecting a photo of Miss Ubuntu February and Miss RedHat April....

Wolfram Alpha?

[DrEasy]

Wolfram Alpha is not a Linux calculator per se, but it's a calculator you can use while on Linux! :)

sage

[Anonymous Coward]

sage is the best calculator ever. it's scriptable in python, comes with a web interface and can typeset results, too.

sagemath.org

TI-85?

[Anonymous Coward]

No TI-85 emulators?

perl -e

[rrohbeck]

Seriously. OK, sometimes the expression becomes a few lines long and at some point I'll write it to a file and fire up vim, but that happens rarely.

ERm..

[MobileTatsu-NJG]

Uh.. the seven best Linux calculators...? Okay, I take it back, can we go back to the Apple news please?

Try this one!

[anwyn]

This one is so advanced most people don't know how to use it.

Great multi-platform business tape calculator

[Anonymous Coward]

SuperbCalc is great. Java. Runs on Windows, Linux, and Mac
http://www.mariottini.net/roberto/superbcalc/index.php

Spam article

[Animats]

The linked page has 24 lines of useful text. It also has 10 ads, and at least 50 links of marginal value.

In other words, it's spam.

Calc: C-style arbitrary precision calculator

[bu1137]

I prefer this one: http://www.isthe.com/chongo/tech/comp/calc/ No gui. Scriptable, with a c-style language. Very usefull.

My faith in numbers.

[Anonymous Coward]

A put down of the MS calculator and not a ONE even asks how are all the calculators mentioned vetted?

grpn is my favorite

[calidoscope]

Obviously an rpn calculator, pretty much the basic trig and exponential (in some ways less than an HP-45 - my first calculator). On the other hand, it does have a deep stack limited by screen space, complex numbers, some matrix ops and an instruction called "rip" - resistors in parallel, which also works with complex numbers.

Re:grpn is my favorite

[turing_m]

That's what I use too. It does the job well enough. If only I could figure out how to get access to the numbers on the stack higher than 2 (swap). But so far I've found it more useful than any other RPN calculator in the ubuntu repos.

These days whenever I want to calculate anything remotely non-trivial I go to a spreadsheet or database, depending on the data. I might want to do a similar calculation again, and I might want to have the calculation properly debugged rather than double check my figures with a calculator. (Yeah I know, someday I will learn R. I promise.) But for fast, once only, back of the envelope type calculations (or exams), RPN calculators still can't be beat. Bound to your calculator button, and accessible from your cell phone.

And for the latter, you want Calc (Roar Lauritzsen's).

http://midp-calc.sourceforge.net/Calc.html

ycalc is the best for programmers

[Anonymous Coward]

The best calculator for you may not be the best calculator for me. For programmers, ycalc is the best. It is the ONLY calculator that supports bitwise operations and lets you toggle bits in binary mode as well.

3.1 wasn't that bad

[Fujisawa Sensei]

Half the people on this site probably weren't even alive when Windows 3.1 came out... could you guys give the pointless Microsoft bashing a rest? Just once, ever?

Could we maybe just get over it instead of posting another "LOLZ Microsoft BOB is bad guyz!!"

This shit pisses me off.

Except every few years they remind us how much they really suck.

And 3.1 wasn't that bad. Except when you compared it with an Amiga.

But since then they've had Windows ME, and Vista to remind us how much they suck.

LOL!

[Joce640k]

I checked some of them out...

Most were considerably *worse* than Windows' built-in calculator - which I didn't think was possible.

Plain, simple RPN command line calculator

[hannorein]

I'm using kalc (http://kalc.sourceforge.net/). It's a pleasure to use. Plain and simple and fast. It's not being developed anymore but the latest stable version is almost perfect ... Hanno

Windows 7?

[aCC]

Now, I understand that Windows 7 is quite good. But I would not go as far as calling it the best calculator for linux. It's also not free....

(In best /. tradition I'm not even reading the summary and misreading the headline... what? There is an article, too???)

Semantic level

[Anonymous Coward]

Amazingly, your sequence [ 2 + 4 = * ( 5 + 6 = ] works (on Kcalc), apparently because "=" also closes parenthesis -- efectively meaning ")=".

So you have a point in that keystroke count differ by not so much. OTOH, the whole point of calling something "algebraic" is to signal direct transposition of formulas from paper to calculator -- without extra work, like the RPN method implies.

In fact, your use of an intermediary "=" works similar to pressing Enter to store a temporary result on the stack.

If one is to use such a convoluted form, I'd go with RPN which seems to have simpler rules.

Of course, that's just me.

Some

[Anonymous Coward]

~$ cat bin/=
#!/bin/bash
echo "scale=5; $1" | bc

or

octave

One stop shopping: R

[Anonymous Coward]

R.

I got a winner

[marcosdumay]

It's interesting that I come here exactly looking for a calculator that could handle units and be used from a CLI. Gnuplot integration is a nice plus.

Not very thorough

[QuietLagoon]

It looks like the review was little more than a feature checklist, with little regard to precision and accuracy of the calculators. Where in the review were the number of guard digits checked? Answer: nowhere.

.

Move on, there is little to see in TFA.

python

[Anonymous Coward]

>>> import math
>>> math.log(reduce(lambda x,y: x*y, range(1, 501), 1))
2611.330458460156
>>>

EMU48

[Anonymous Coward]

Meh, I'm still used to my old HP48 engineering calc from college, so I run EMU48 via WINE. Works like a champ! :)

SAGE

[Taxman415a]

I think a SAGE notebook is what you are looking for. It's basically multiple different open source mathematics software packages (such as Maxima) glued together with python. The notebook can typeset it automatically for you or you can output the LaTeX. Well you may like the custom system you have better, but SAGE is pretty slick since it goes into a wide range of math. I suppose any customizations you would like could be contributed to SAGE.

How about PARI

[32771]

I do think that GUI calculators needn't look like the ones you can hold in your hands. The numbers are already on the keyboard why use a mouse to access them. So a simple command line tool is fine with me.
PARI (or gp) seemed to do well for me. I like the fact that it supports arbitrary precision arithmetic and uses rational numbers as long as possible. The part that irks me is that it doesn't deal well with other number systems like hexadecimal or binary especially as far as output is concerned.

If you know something better, I'm open for suggestions. I didn't try the ones suggested in the article though.

cli is nice

[belmolis]

I generally use bc for straightforward numerical calculations and R for more complicated things. Another nice cli program is frink, which understands and tracks a huge set of units. It's free as in beer but not open source.

Texas Instruments TI-89 Titanium - Compatability?

[dogzdik]

Are there are any emulators and or interface and or "WINE" style programs that can interface with / run the said calculator or it's software? Anything else is just ungay.

RPN is stack-based

[Estragib]

The <enter> is not part of the RPN. More correctly, you don't need it exactly because you're using RPN, which works on a stack. Values get pushed on, operations pop values off the stack. 2 4 + 5 6 + * P works just fine.

You add 2 and 4 to the stack, then execute + (on TOS and TOS-1), so TOS is now 6. You then add 5 and 6 to the stack, exec + again to get a stack of 6 and 11, on which finally you execute * to get a TOS of 66.

A reverse-polish calculator stores numbers on a stack. Entering a number pushes it on the stack. Arithmetic operations pop arguments off the stack and push the results.

Manual page dc(1), ll. 22-24

% echo "2 4 + 5 6 + * P" | dc
66

Re:RPN is stack-based

[Luyseyal]

You're missing the point. You have to enter a space, carriage return, or some other differentiator for the calculator to know that you mean 2 then 4 not 24. This is not necessary in algebraic notation because it can all be run together. That is why s/he counted the ENTER press.

-l

Re:RPN is stack-based

[Estragib]

You're right, I did. So with 24+56+* we're up to 10 strokes for RPN, too.

Re:RPN is stack-based

[Luyseyal]

Yeah, I haven't seen any awesome examples of why RPN is better than algebraic other than some notion of conceptual purity. There's probably not enough of a gain to make much of a difference...

-l

/This — coming from a long-time Dvorak layout user!
//Don't switch. The research shows it's not worth it. But sadly, I'm now more used to it than qwerty...