Processing speed
21 views
Skip to first unread message
Joe
Nov 14, 2009, 12:41:02 PM11/14/09
to tinspire
To integrate the expression t^2*sin(t^3+t^4)*sqrt(4t^2+9t^4+16t^6)
from zero to five, where sqrt is the square root, and with the exact/
approx mode in auto, using a TI-voyage200 took nineteen minutes and
twelve seconds. To do the same problem using the tinspire cas
software on my netbook computer took only 1 1/4 seconds. Would
someone please see how long that problem takes using a tinspire cas
handheld? I'm guessing it will take somewhere around one or two
minutes. Thanks. BTW the answer is 15.0074
Joseph
from zero to five, where sqrt is the square root, and with the exact/
approx mode in auto, using a TI-voyage200 took nineteen minutes and
twelve seconds. To do the same problem using the tinspire cas
software on my netbook computer took only 1 1/4 seconds. Would
someone please see how long that problem takes using a tinspire cas
handheld? I'm guessing it will take somewhere around one or two
minutes. Thanks. BTW the answer is 15.0074
Joseph
Andy Kemp
Nov 14, 2009, 1:11:24 PM11/14/09
to tins...@googlegroups.com
Just did it on my CAS handheld and it took 47 seconds.
Andy
Nelson Sousa
Nov 14, 2009, 1:11:46 PM11/14/09
to tins...@googlegroups.com
roughly 40 seconds.
But I don't see the usefulness of such a benchmark. To show hat a 1.2
GHz CPU with 1 or 2 Gb of ram is faster than a 100 MHz CPU with 32 Mb
of ram, which in turn is faster than a 15 MHz CPU with 256 kb of RAM?
Well, I bet my 2.4 GHz CPU with 4gb of ram can be even faster! ;)
You're not comparing software performance, you're just comparing sheer
CPU power an RAM availability between three completely different
platforms. It reminds me of a race I once read about in a car
magazine: they had a F16 fighter, a 1100 cc bike and a Ferrari car on
a drag race to see which one was faster. Ok, there are valid results
there (the bike almost won, being overtaken by the fighter but only
after it was airborne), but I don't think they'll be usefull for
people that want to compare before buying; I don't see the air force
switching from F16 fighters to 1100cc bikes just because the bike can
accelerate faster. Nor will potential Ferrari buyers think twice and
choose to buy a bike or a F16 fighter instead.
If you want a useful benchmark, you may compare the response time of
Nspire CAS against a Casio Classpad and HP50G calulators and see how
does the Nspire handheld matches the competition.
Cheers,
Nelson
But I don't see the usefulness of such a benchmark. To show hat a 1.2
GHz CPU with 1 or 2 Gb of ram is faster than a 100 MHz CPU with 32 Mb
of ram, which in turn is faster than a 15 MHz CPU with 256 kb of RAM?
Well, I bet my 2.4 GHz CPU with 4gb of ram can be even faster! ;)
You're not comparing software performance, you're just comparing sheer
CPU power an RAM availability between three completely different
platforms. It reminds me of a race I once read about in a car
magazine: they had a F16 fighter, a 1100 cc bike and a Ferrari car on
a drag race to see which one was faster. Ok, there are valid results
there (the bike almost won, being overtaken by the fighter but only
after it was airborne), but I don't think they'll be usefull for
people that want to compare before buying; I don't see the air force
switching from F16 fighters to 1100cc bikes just because the bike can
accelerate faster. Nor will potential Ferrari buyers think twice and
choose to buy a bike or a F16 fighter instead.
If you want a useful benchmark, you may compare the response time of
Nspire CAS against a Casio Classpad and HP50G calulators and see how
does the Nspire handheld matches the competition.
Cheers,
Nelson
Ross
Nov 14, 2009, 3:43:18 PM11/14/09
to tinspire
Wow, I didn't realize that the TI-Nspire CAS was so fast. going from
almost 18 minutes to 40 seconds is quite impressive, giving yet
another reason to get the Nspire CAS over the 89/92/200
almost 18 minutes to 40 seconds is quite impressive, giving yet
another reason to get the Nspire CAS over the 89/92/200
Nelson Sousa
Nov 14, 2009, 4:26:01 PM11/14/09
to tins...@googlegroups.com
Try to factor 2^67-1 on both V200/89T and Nspire and you'll also see a
20x increase in speed.
In symbolic manipulation is "only" about 10x faster, though.
Nelson
20x increase in speed.
In symbolic manipulation is "only" about 10x faster, though.
Nelson
Joe
Nov 14, 2009, 9:57:41 PM11/14/09
to tinspire
To factor 2^67-1 on the ti-v200 took 2 minutes and 24 seconds. On the
netbook computer with nspire cas for windows it was instantaneous.
The Hp50g gave up after 1 minute and 45 seconds and returned the
integer value of 2^67-1 unfactored. It sounds like it would take
about 6 or 7 seconds on the nspire cas handheld. I wonder how long it
would take on a Casio Classpad?
joe
> >> > Joseph- Hide quoted text -
>
> - Show quoted text -
netbook computer with nspire cas for windows it was instantaneous.
The Hp50g gave up after 1 minute and 45 seconds and returned the
integer value of 2^67-1 unfactored. It sounds like it would take
about 6 or 7 seconds on the nspire cas handheld. I wonder how long it
would take on a Casio Classpad?
joe
>
> - Show quoted text -
Andy Kemp
Nov 15, 2009, 5:56:14 AM11/15/09
to tins...@googlegroups.com
Just did this on the CAS handheld and it was closer to 2 seconds...
Joe
Nov 15, 2009, 11:53:17 AM11/15/09
to tinspire
Thanks Andy. That is even faster than a 20X increase over the v200
performance which is quite an improvement. I sure hope someone will
give us some times for the ClassPad for these problems. I'm very
curious and don't have any experience at all with the Casio device.
Perhaps Nelson can shed some light on this. BTW, the integration
problem that I used in my initial post is a textbook problem that came
from the Calculus text by Stewart, where it is labeled as a CAS
exercise. I think it is worth pointing out to the student that it
takes about 20 minutes to solve this problem on the 89 or v200, so it
is very important to enter the problem correctly. Then use a lot of
patience, or turn to a faster alternative such as the nspire cas, or
nspire cas software on a pc, or maple, or mathematica if you have one
of these available.
> > > - Show quoted text -- Hide quoted text -
performance which is quite an improvement. I sure hope someone will
give us some times for the ClassPad for these problems. I'm very
curious and don't have any experience at all with the Casio device.
Perhaps Nelson can shed some light on this. BTW, the integration
problem that I used in my initial post is a textbook problem that came
from the Calculus text by Stewart, where it is labeled as a CAS
exercise. I think it is worth pointing out to the student that it
takes about 20 minutes to solve this problem on the 89 or v200, so it
is very important to enter the problem correctly. Then use a lot of
patience, or turn to a faster alternative such as the nspire cas, or
nspire cas software on a pc, or maple, or mathematica if you have one
of these available.
Nelson Sousa
Nov 15, 2009, 3:25:46 PM11/15/09
to tins...@googlegroups.com
Sorry, can't help you with that. I don't have a Classpad. I'll keep
an eye for people that may have them and I'll give it a try if I get
the chance.
Cheers,
Nelson
an eye for people that may have them and I'll give it a try if I get
the chance.
Cheers,
Nelson
jfmarques
Nov 15, 2009, 4:26:20 PM11/15/09
to tinspire
On a old computer (pentium 4, 1.8 MHz, 1 GB RAM) casio classpad
computer software runs a little more than 3 minutes to factorize
2^67-1.
answer: 193707721.761838257287
My brother has my classpad. I'll do in the casio classpad handheld.
Joaquim
computer software runs a little more than 3 minutes to factorize
2^67-1.
answer: 193707721.761838257287
My brother has my classpad. I'll do in the casio classpad handheld.
Joaquim
Joe
Nov 16, 2009, 1:42:52 AM11/16/09
to tinspire
Hi Joaquim, it would be great if you could provide the time for the
casio classpad handheld to work that problem. I've always been
curious about the classpad but I don't know anyone that has one and
have never seen one my self.
Thanks,
joe
> > > - Show quoted text -- Hide quoted text -
casio classpad handheld to work that problem. I've always been
curious about the classpad but I don't know anyone that has one and
have never seen one my self.
Thanks,
joe
Reply all
Reply to author
Forward
0 new messages