Math students, you are getting such a raw deal.
May 28, 2008 7:23 PM Subscribe
Why doesn't Moore's Law apply to graphing calculators?
When I was a freshman in college (1994), my parents bought me a TI-85 graphing calculator (6Mhz, 32k RAM) for (I think) around $130. 14 years later, for about the same price, I can get a calculator with about 4 times the speed and 4 times the RAM.
Shouldn't today's calculator be vastly more powerful (or at least much cheaper) -- in line with advances in home computers?
When I was a freshman in college (1994), my parents bought me a TI-85 graphing calculator (6Mhz, 32k RAM) for (I think) around $130. 14 years later, for about the same price, I can get a calculator with about 4 times the speed and 4 times the RAM.
Shouldn't today's calculator be vastly more powerful (or at least much cheaper) -- in line with advances in home computers?
My pet theories as a math teacher:
Every time they put out a significantly more powerful graphing calculator (with extra features to take advantage of that extra power), it gets banned from use on standardized tests (SATs, AP exams, etc.), so not enough people will buy them. The price doesn't come down because the creators have to convince the schools to use the calculators, not the individual consumers, who will often buy whatever the standard for their class or whatever they were taught to use in high school.
posted by ErWenn at 7:40 PM on May 28, 2008 [1 favorite]
Every time they put out a significantly more powerful graphing calculator (with extra features to take advantage of that extra power), it gets banned from use on standardized tests (SATs, AP exams, etc.), so not enough people will buy them. The price doesn't come down because the creators have to convince the schools to use the calculators, not the individual consumers, who will often buy whatever the standard for their class or whatever they were taught to use in high school.
posted by ErWenn at 7:40 PM on May 28, 2008 [1 favorite]
There are several reasons.
Complacency: TI is very entrenched with math instruction in high schools and colleges and thus enjoys guaranteed yearly sales of their standard graphing calculators. They simply don't have much incentive to increase the specifications of their calculators or lower prices.
Restrictions: Both TI and other manufacturers like HP have rather powerful calculators with ARM processors, but standardized tests restrict powerful calculators. A student wants to be able to use the same calculator that he or she has spent hours getting familiarized with on the big test at the end of the year.
posted by zsazsa at 7:45 PM on May 28, 2008
Complacency: TI is very entrenched with math instruction in high schools and colleges and thus enjoys guaranteed yearly sales of their standard graphing calculators. They simply don't have much incentive to increase the specifications of their calculators or lower prices.
Restrictions: Both TI and other manufacturers like HP have rather powerful calculators with ARM processors, but standardized tests restrict powerful calculators. A student wants to be able to use the same calculator that he or she has spent hours getting familiarized with on the big test at the end of the year.
posted by zsazsa at 7:45 PM on May 28, 2008
The new one probably uses less power, too. It's likely that graphing calculator capabilities are improving exponentially just as PC capabilities are, and I'd argue that that's the key idea behind Moore's Law, not the specific doubling time (which might be around seven years, according to your observations).
posted by Mapes at 7:52 PM on May 28, 2008
posted by Mapes at 7:52 PM on May 28, 2008
Probably because the interface / form-factor sucks. What's the point of upping the power when the screen and keypad are tiny? You may as well use your computer with Mathematica, Maple, Matlab, etc on it. Especially if you are doing Real Work®, rather than some trivial problem out of a textbook. If it's trivial enough for a calculator, a seasoned professional will just make an estimate (e.g. sin θ = θ when θ is small) 90% of the time.
posted by randomstriker at 8:05 PM on May 28, 2008
posted by randomstriker at 8:05 PM on May 28, 2008
Is there a website that provides a Dummies-style guide to the TI-85? I bought one in 1994, and now my daughter is interested in it.
posted by Wild_Eep at 8:17 PM on May 28, 2008
posted by Wild_Eep at 8:17 PM on May 28, 2008
I think the bottom line is that the calculator companies have hit a wall with traditional graphing calculators on two fronts of demand. In education they hit a wall of educators not wanting the calculator to be able to do more and in industry its being beaten by real computers not only because they will always have better computing power but also because the information is easier to use(email, print, analyze).
The price seems to remain high because they can get away with it.
As an aside the math capabilities included in the current (TI-89) is also quite a bit better; basically it added the some nice symbolic manipulation.
posted by humanawho at 8:26 PM on May 28, 2008
The price seems to remain high because they can get away with it.
As an aside the math capabilities included in the current (TI-89) is also quite a bit better; basically it added the some nice symbolic manipulation.
posted by humanawho at 8:26 PM on May 28, 2008
I've personally always wondered why they never made the leap to a color screen. I would buy one.
posted by DMan at 8:32 PM on May 28, 2008
posted by DMan at 8:32 PM on May 28, 2008
I'm sure it would use a lot more power and drain the battery more, DMan...
posted by ph00dz at 8:38 PM on May 28, 2008
posted by ph00dz at 8:38 PM on May 28, 2008
On one side, EasyCalc is free and runs on a now sub-$50 Palm III.
On the other, Here's the Voyage 200 PLT, for $110.
As everybody else said, they're definitely not for tests. I have a TI-86 I bought for $5 at a yard sale for that.
posted by Orb2069 at 8:42 PM on May 28, 2008
On the other, Here's the Voyage 200 PLT, for $110.
As everybody else said, they're definitely not for tests. I have a TI-86 I bought for $5 at a yard sale for that.
posted by Orb2069 at 8:42 PM on May 28, 2008
Best answer: I think the major issue is the lack of competition. TI basically has the highschool market locked up, and for most people that's when they buy the one-and-only graphing calculator they'll ever use. HP has a big slice of the college and professional markets, but there you're talking about a much smaller market. Casio is a distant, distant third in the U.S.
And calculators aren't like cellphones or PCs; they're not typically something that consumers replace every few years in order to get the latest-and-greatest. I'd wager that most people barely scratch the surface of the graphing calculators that they have right now, so they're not exactly clamoring for more stuff. This means it's tough to grab customers from a competitor: once you've lost the initial sale, you're pretty much done.
So, combine an entrenched market leader (at best a duopoly in the pro/engineering market), with very low repeat sales, and a product that satisfies most of what customers want from it ... and there's not a lot to drive change.
Also unlike computers, there's not a lot of third-party software development on graphing calculators (there's some on the TI-89 and some of the HPs, but it's not much), so you don't get into the hardware/software development spiral that drove the "upgrade treadmill" for PCs through the 80s and 90s until today.
The only feature I suspect you'll see graphing calculator manufacturers add in the near future -- if they haven't already, I haven't been keeping up -- are iPod-style color TFT screens. But in terms of core features, it's a tough sell to get people to buy more than what's already in the TI-83 or (at most) 89; you just don't need much more for high-school level math as its taught in the U.S., and that's the big market for them.
At some point, it will probably stop being cost-effective for TI to continue manufacturing or sourcing the parts it uses in its current lineup, and they'll probably move to something that's more like a cellphone or an iPod on the inside (in other words, a commodity platform). In fact, at some point it might become cheaper to eliminate all those membrane switches and just go with a bit touchscreen, a la the new iPods and iPhones. But it'll be a while, and in the meantime I don't expect them to change all that much. And even if and when they eventually do modernize the hardware, most of the functionality will just be the same, plus some eye candy.
posted by Kadin2048 at 10:35 PM on May 28, 2008 [1 favorite]
And calculators aren't like cellphones or PCs; they're not typically something that consumers replace every few years in order to get the latest-and-greatest. I'd wager that most people barely scratch the surface of the graphing calculators that they have right now, so they're not exactly clamoring for more stuff. This means it's tough to grab customers from a competitor: once you've lost the initial sale, you're pretty much done.
So, combine an entrenched market leader (at best a duopoly in the pro/engineering market), with very low repeat sales, and a product that satisfies most of what customers want from it ... and there's not a lot to drive change.
Also unlike computers, there's not a lot of third-party software development on graphing calculators (there's some on the TI-89 and some of the HPs, but it's not much), so you don't get into the hardware/software development spiral that drove the "upgrade treadmill" for PCs through the 80s and 90s until today.
The only feature I suspect you'll see graphing calculator manufacturers add in the near future -- if they haven't already, I haven't been keeping up -- are iPod-style color TFT screens. But in terms of core features, it's a tough sell to get people to buy more than what's already in the TI-83 or (at most) 89; you just don't need much more for high-school level math as its taught in the U.S., and that's the big market for them.
At some point, it will probably stop being cost-effective for TI to continue manufacturing or sourcing the parts it uses in its current lineup, and they'll probably move to something that's more like a cellphone or an iPod on the inside (in other words, a commodity platform). In fact, at some point it might become cheaper to eliminate all those membrane switches and just go with a bit touchscreen, a la the new iPods and iPhones. But it'll be a while, and in the meantime I don't expect them to change all that much. And even if and when they eventually do modernize the hardware, most of the functionality will just be the same, plus some eye candy.
posted by Kadin2048 at 10:35 PM on May 28, 2008 [1 favorite]
Technologically, they are already powerful enough for everything they'll do. Heck, the TI-89 can run Super Mario 3 and Doom (well, kinda). Solving very hard equations and more symbolic manipulation tricks would not be worth the power/programming effort required.
I would guess manufacturers just want to keep the platforms stable (TI-83/84, TI-89) for ease of deployment (schools, SATs) and development. TI also has the advantage that they are the source for nearly all the parts their calculators use. I don't think they make much flash memory however ("ti flash memory" doesn't come up with much), so that is slowly bending to Moore's Law.
posted by easyasy3k at 1:33 AM on May 29, 2008
I would guess manufacturers just want to keep the platforms stable (TI-83/84, TI-89) for ease of deployment (schools, SATs) and development. TI also has the advantage that they are the source for nearly all the parts their calculators use. I don't think they make much flash memory however ("ti flash memory" doesn't come up with much), so that is slowly bending to Moore's Law.
posted by easyasy3k at 1:33 AM on May 29, 2008
For what it's worth, the Ti-89 is based on an aging design with the original version being released in 1998. At the time of release, a 16MHz M68000 was reasonably powerful for a calculator.
There's also the fact that a large proportion of their market has moved onto PC systems such as computer algebra, numerical analysis and statistical packages for most "real work" making the most significant proportion of those buying it schools or students where additional raw processing power isn't so much of a concern compared to having a calculator permitted in examinations. As such they probably see no justification in upgrading it as they don't shift units based on their processing power, they are the market leader in a relatively niche market so they can artificially keep prices high, they make a large margin on each one sold that doesn't give them the incentive to upgrade it or investing in giving a tried-and-trusted design a major refresh.
In contrast, more recent designs such as the HP 50g, which has a 75mhz ARM9 processor, or the TI-Nspire, which has a 150mhz ARM9 processor, are comparable to the hardware on the Nintendo DS.
posted by HaloMan at 6:25 AM on May 29, 2008
There's also the fact that a large proportion of their market has moved onto PC systems such as computer algebra, numerical analysis and statistical packages for most "real work" making the most significant proportion of those buying it schools or students where additional raw processing power isn't so much of a concern compared to having a calculator permitted in examinations. As such they probably see no justification in upgrading it as they don't shift units based on their processing power, they are the market leader in a relatively niche market so they can artificially keep prices high, they make a large margin on each one sold that doesn't give them the incentive to upgrade it or investing in giving a tried-and-trusted design a major refresh.
In contrast, more recent designs such as the HP 50g, which has a 75mhz ARM9 processor, or the TI-Nspire, which has a 150mhz ARM9 processor, are comparable to the hardware on the Nintendo DS.
posted by HaloMan at 6:25 AM on May 29, 2008
randomstriker's view is pretty much the opposite view to that taken by me most of my colleagues, and in the physics department of a university. We all reach for a calculator several times a day precisely because the interface is so minimal and perfect.
In answer to the question, the processor is not more powerful because it doesn't need to be any more powerful for what they continue to be used for. It's almost the same as asking why home refrigerators not getting any colder, despite the fact that physicists are now within milliKelvin of absolute zero.
posted by caek at 7:23 AM on May 29, 2008
In answer to the question, the processor is not more powerful because it doesn't need to be any more powerful for what they continue to be used for. It's almost the same as asking why home refrigerators not getting any colder, despite the fact that physicists are now within milliKelvin of absolute zero.
posted by caek at 7:23 AM on May 29, 2008
"by me and most of my colleagues in the physics department of a university", that should be.
posted by caek at 7:24 AM on May 29, 2008
posted by caek at 7:24 AM on May 29, 2008
Your question has been answered perversely by Texas Instruments' new TI-Nspire graphing calculator, now becoming available in retail stores. I've had one for a year. Though upgrades continue, currently I feel that the TI-Nspire is the splash after TI jumped the shark. It has as many keys as a computer keyboard in 1/6 the area. The TI-Nspire has windows and scrollbars, but it's not quite intuitive enough. If classrooms have to spend more time explaining how to find and use some app, rather than the math d'jour, that's a killer. It chews through batteries much faster than its predecessors. People have retrofitted new TI-Nspire functions allowed on AP exams to the older TI-83+, TI-84+, and TI-89, which are programmable.
Casio offers color and touch screen calculators. Casio offers equivalent functionality for half the $110 price of the dominant TI-83/TI-84 calculators. I think Casios has not succeed in US schools due to UI flaws, battery-eating, and better marketing by TI.
Some school districts have used Palm and similar handhelds, which with a little external keyboard make decent word processors and all around cool tools. HP's high end calculators have significantly more functions than TI's. As repeatedly noted above, the ed market has the functionality it wants, and people who would use more advanced functions prefer computers. Sure, tech could meld the calculator with the portable media player, cellphone, PDA, and keep all your books on it too. We could call it the Dynabook. Schools might like the reduction of paper books, but are not prepared to deal with the abusability of such technology.
TI has a lock on the US ed calculator market. That market doesn't need more functionality or speed. By Moore's law and by Casio's demonstration, TI could halve the cost of its graphing calculators and probably make 'em zippier. But why should TI do that?
posted by gregoreo at 7:24 AM on May 29, 2008
Casio offers color and touch screen calculators. Casio offers equivalent functionality for half the $110 price of the dominant TI-83/TI-84 calculators. I think Casios has not succeed in US schools due to UI flaws, battery-eating, and better marketing by TI.
Some school districts have used Palm and similar handhelds, which with a little external keyboard make decent word processors and all around cool tools. HP's high end calculators have significantly more functions than TI's. As repeatedly noted above, the ed market has the functionality it wants, and people who would use more advanced functions prefer computers. Sure, tech could meld the calculator with the portable media player, cellphone, PDA, and keep all your books on it too. We could call it the Dynabook. Schools might like the reduction of paper books, but are not prepared to deal with the abusability of such technology.
TI has a lock on the US ed calculator market. That market doesn't need more functionality or speed. By Moore's law and by Casio's demonstration, TI could halve the cost of its graphing calculators and probably make 'em zippier. But why should TI do that?
posted by gregoreo at 7:24 AM on May 29, 2008
Sounds like a job for the iPhone.
But yeah, it won't be going with you into the SAT.
posted by zpousman at 8:34 AM on May 29, 2008
But yeah, it won't be going with you into the SAT.
posted by zpousman at 8:34 AM on May 29, 2008
If such processors and smart calculators are so prevalent, what would stop a person from using it on a test? If they as likely to have on your person as pen and paper to do the math? Do you have a pencil and a TI-81, pencil (#2) and paper in your pocket, or a cell phone? Clearly traditional educations ideas on technology and even collaboration are dated. It's a redundant question anyway.
Between GraphicCalc for Windows Qalculate for Linux, Unix, and OSX, there are plenty of options. One can even use Sage via a web browser on many cellphones.
One Laptop Per Child has discussed this idea and provides a couple different graphic calculators (for different age groups) and one can install very high end math platforms on the XO laptop.
The idea that the most rural kids in Peru are going to have better math resources than US kids should give one pause. What a great selling point for the XO.
posted by sethwoodworth at 3:12 PM on May 29, 2008
Between GraphicCalc for Windows Qalculate for Linux, Unix, and OSX, there are plenty of options. One can even use Sage via a web browser on many cellphones.
One Laptop Per Child has discussed this idea and provides a couple different graphic calculators (for different age groups) and one can install very high end math platforms on the XO laptop.
The idea that the most rural kids in Peru are going to have better math resources than US kids should give one pause. What a great selling point for the XO.
posted by sethwoodworth at 3:12 PM on May 29, 2008
This thread is closed to new comments.
posted by Electrius at 7:32 PM on May 28, 2008