HP calc. Customer Service
in [HP48]
|
Prev: HP50g vs. Voyage 200
Next: HPUserEdit crash at launch
From: Michael Kuyumcu on 1 Sep 2006 14:19 Well - my FIRST impression of the nspire is + it is really quite fast for a pocket calculator. You can move around complete coordinate systems with plotted function graphs almost instantaneously. And I remember another topic on this list ("my HP won't factor 2^67-1", while the TI need only 4 seconds) + the screen makes a nice impression (large, clear, fine) - programming abilities are "pre-optimal" (to put it nicely): no assembler programming no *programs*, only functions. While this appears desirable to many purists, it is not at all to me. I think that disabling programs, for no obvious reasons, is treating the users in a patronizing way. - graphics options missing: no 3D plotting. And this, for me, is a serious thing. I am supposed to teach analytic geometry using the nspire. Thus I would need 3D coordinate systems, lines, planes, spheres and so on and a way to make them intersect etc. In this state, with only 2D graphics, it's half-useless for this curriculum topic. If I were to use this device for analytic geometry, I would have to program functions myself that display 3D objects. I'd have to program a basic 3D engine. I am afraid that I simply don't have the time for that. + sufficient analysis and calculus capabilities. It seems to have a lot less functions that the HP, and this is partly why I call the ROM programming of the HP superior in principle, but not in praxis (ARM). ? vector calculating capabilities + full-blown spreadsheet. But I have yet to find out how good it is really. + sufficient number of statistical functions. But are there good ways to visualize the results? That's it for now. Regards, Michael Kuyumcu JB schrieb: > Yao Konan wrote: > > What make you think that the TI-Nspire runs at over 100 Mhz ? > > According to the education.ti site the NSpire CPU has over 6 times the > > speed of the TI89TI CPU,so i would say that it should run at 75 Mhz > > like the HP50G CPU and close to the ClassPad CPU. > > Anyway,i think that it is incredible because the NSpire certainly uses > > the same algorithms as the TI68k and BCD,yet it seems from 16 to almost > > 20 times faster than my TI92+ for BCD floating operations. > > So this means that TI has done quite a good job on the TI-Nspire > > software. > > > > Jean-Yves Avenard a écrit : > > > > > Michael Kuyumcu wrote: > > > > > > >> That would mean that the TI-NSpire is more than 10 times faster than my > > > >> TI92+ on this problem.This is simply incredible ! > > > > > > you're comparing a 16 bits processor running at 12Mhz vs a 32bits RISC > > > processor running at over 100Mhz ... > > > It will likely be more than 10 times faster... > > > > > > JY > > HI Yao Konan, Now that you have had a peek at the tinspire, aside from > the disturbing lack of 3d plotting, how do you rate it? I'd rather > have the 3-d capability and more math capability such as Laplace > transforms than the extra speed. Of course having it all would be even > better. - JB
From: Michael Kuyumcu on 1 Sep 2006 14:21 I did not write that. Michael Kuyumcu Jean-Yves Avenard schrieb: > Michael Kuyumcu wrote: > > >> That would mean that the TI-NSpire is more than 10 times faster than my > >> TI92+ on this problem.This is simply incredible ! > > you're comparing a 16 bits processor running at 12Mhz vs a 32bits RISC > processor running at over 100Mhz ... > It will likely be more than 10 times faster... > > JY
From: Michael Kuyumcu on 1 Sep 2006 14:24 No speed tests here, sorry. I think I'm not interested in them. I want to see if the device meets the curricular requirements for my math classes. But if you post a few representative timing challenges here, I will be glad to have them executed :-) Regards, Michael Kuyumcu Yao Konan schrieb: > Wow again over 10 times faster the TI92+. > If the TI-Nspire is as relativelly as fast for all operations than it > is definitvelly a quantum leap from the TI89Ti/Voyage 200 in term of > performances. > It is too bad it doesn't have 3D plotting because with such a speed and > the grayscale levels high resolution screen,real time 3D rotation of > surfaces would be wonderful. > Btw have you down some speed tests on your own ? > > Michael Kuyumcu a écrit : > > > That's 15 seconds > > Michael Kuyumcu > > > > > > > > Yao Konan schrieb: > > > > > > The on-calc system info says otherwise: it treats the whole memory as > > > > one block. The system software takes up about 8 or so MB, and the rest > > > > is free for you, the user. At least the system info does not > > > > differentiate any ram or rom areas visibly. > > > > > > Well,i think that those 32 MB are Flash ROM as the TI site claims that > > > the TI-NSpire has close to 20 MB of storage memory thus Flash ROM. > > > The TI site also claims 16 MB of computing memory so iassume that this > > > memory is not SRAM and can't be used to store datas. > > > > > > > Interesting: fib(10000) produces an overflow, even in Exact mode, while > > > > the CAS seems to be capable of handling multiple-precision numbers > > > > otherwise with no problems.fib(1000) works, and does so within 2 > > > > seconds. > > > > > > Surprising. > > > That would mean that the TI-NSpire is more than 10 times faster than my > > > TI92+ on this problem.This is simply incredible ! > > > To have a better idea of the relative speed of this tool could you > > > clock this operation which takes almost 4 min on a TI92+ ? > > > approx(randmat(20,20)^-1) > > > It doesn't really suprise me that fib(10000) returns an overflow. > > > > > > > Regards, > > > > Michael Kuyumcu > > > > > > > > > > > >
From: Harold A. Climer on 1 Sep 2006 14:37 On 29 Aug 2006 05:55:37 -0700, "Michael Kuyumcu" <info(a)noemanetz.de> wrote: SNIP SNIP >Strangely, HP does not seem interested in aquainting new generations of >students with their - conceptually superior - calculators. As TI, >Casio, and Sciface give the schools these 120 free units, in return, >the schools bind themselves to equip the following three years of >students with the same devices - at the school's cost! So these >companies gain a first firm standing in the schools, every student will >know their products, and will probably continue to buy and use them >also when school is over and they attend university. When many teachers in the US have to spend part of their own meager salaries for paper and pencils etc for their students, I don think this method of getting calculators out in the schools would work because of the costs involved. HP is going to have to donate calculators to some schools systems My dad, now passed away ,in his twenty five years of teaching spent his own money quite a few times on textbooks. workbooks, paper, pencils, etc for his classes ,because the books provided by the schools system were almost 10 years old. . Harold A. Climer Dept.Of Physics,Geology,and Astronomy University of Tennessee at Chattanooga Chattanooga TN USA 37403
From: Yao Konan on 1 Sep 2006 14:47
> HI Yao Konan, Now that you have had a peek at the tinspire, aside from > the disturbing lack of 3d plotting, how do you rate it? I'd rather > have the 3-d capability and more math capability such as Laplace > transforms than the extra speed. Of course having it all would be even > better. - JB As TW has said i am just reacting to what is reported by Michael who is testing the TI-Nspire.Actually i am quite impressed by the speed of the NSpire even if i am disapointed by the lack of 3D plotting,advanced programming and of more advanced maths functions. For the programming i hope that there will be a way to program it in C++ and a SDK to enable applications developpement. However i am sure that TI is serious about replacing Derive by TI-NSpire thus either there will be 2 versions of the TI-NSpire or TI plans to strongly improve the software in the future. In fact i don't understand why TI has used AMS 3.10 as the base for the CAS when it would have been better to use Derive instead. |