Firmware 5447: Iterate() bug or feature?
02-13-2014, 05:40 PM (This post was last modified: 02-13-2014 05:54 PM by ColinJDenman.)
Post: #1
 ColinJDenman Member Posts: 102 Joined: Feb 2014
Firmware 5447: Iterate() bug or feature?
In performing ITERATE(z^2+.37-.14*i,Z,0,n) for n=1001, I get
Error: Invalid input
This is true for n>1000, but doesn't appear to be noted in the
calculator help or the pdf user guide.

I note that it is possible to nest ITERATE to get beyond 1000, but it gets unpicturesque.

Is there any good reason for the limit? The manual uses the notation '#times' for n,
is that significant somehow?

I note that the form above for n=1000 times at 0.078s, compared to a nested 2000 loop
of 0.146s. A simple program JULIA using variables, runs a 1000 iteration at 0.136s, half the speed and can do 500000 in 53.485s
02-13-2014, 06:52 PM (This post was last modified: 02-13-2014 06:52 PM by debrouxl.)
Post: #2
 debrouxl Member Posts: 249 Joined: Dec 2013
RE: Firmware 5447: Iterate() bug or feature?
There's a reason for such an arbitrary limit in a number of functions, but it's not a good one
These caps on the size of some operations were inherited from the 39gII, and chances are that they will disappear from the Prime at some point.
02-13-2014, 07:02 PM
Post: #3
 eried Senior Member Posts: 741 Joined: Dec 2013
RE: Firmware 5447: Iterate() bug or feature?
Happens in the emulator too? because I am able to use n=5000001 without issues (takes about 10s)

My website: erwin.ried.cl
02-13-2014, 07:11 PM
Post: #4
 ColinJDenman Member Posts: 102 Joined: Feb 2014
RE: Firmware 5447: Iterate() bug or feature?
[quote='eried' pid='4888' dateline='1392318149']
Happens in the emulator too? because I am able to use n=5000001 without issues (takes about 10s)

Interesting. No I haven't tried it in the emulator, just in my updated hardware.
02-13-2014, 07:17 PM
Post: #5
 ColinJDenman Member Posts: 102 Joined: Feb 2014
RE: Firmware 5447: Iterate() bug or feature?
(02-13-2014 06:52 PM)debrouxl Wrote:  There's a reason for such an arbitrary limit in a number of functions, but it's not a good one
These caps on the size of some operations were inherited from the 39gII, and chances are that they will disappear from the Prime at some point.

Thank you for the insight. I'm afraid I've been away from HP calculators since the time of the HP-41 and HP-16, so not at all versed in the idiosyncrasies of the modern age. I hope the firmware does improve somewhat. I'm not used to a HP crashing so frequently ;-)
02-13-2014, 07:18 PM (This post was last modified: 02-13-2014 07:22 PM by Han.)
Post: #6
 Han Senior Member Posts: 1,817 Joined: Dec 2013
RE: Firmware 5447: Iterate() bug or feature?
(02-13-2014 07:02 PM)eried Wrote:  Happens in the emulator too? because I am able to use n=5000001 without issues (takes about 10s)

Looking more carefully at this, do keep in mind that you are iterating a constant as z and Z are distinct.

As for the limitations, I believe those were "removed" (or better said, raised) since the most recent update. Are you running firmware 5447?

Graph 3D | QPI | SolveSys
02-13-2014, 07:25 PM
Post: #7
 ColinJDenman Member Posts: 102 Joined: Feb 2014
RE: Firmware 5447: Iterate() bug or feature?
(02-13-2014 07:18 PM)Han Wrote:  Looking more carefully at this, do keep in mind that you are iterating a constant as z and Z are distinct.

As for the limitations, I believe those were "removed" (or better said, raised) since the most recent update. Are you running firmware 5447?

Yes, just loaded yesterday. I saw the readme file noted improvements to iterate.
02-13-2014, 07:59 PM
Post: #8
 eried Senior Member Posts: 741 Joined: Dec 2013
RE: Firmware 5447: Iterate() bug or feature?
(02-13-2014 07:18 PM)Han Wrote:
(02-13-2014 07:02 PM)eried Wrote:  Happens in the emulator too? because I am able to use n=5000001 without issues (takes about 10s)

Looking more carefully at this, do keep in mind that you are iterating a constant as z and Z are distinct.

As for the limitations, I believe those were "removed" (or better said, raised) since the most recent update. Are you running firmware 5447?

Oh, you are right, I just copy and pasted! In any case, no limitations either way.

My website: erwin.ried.cl
02-13-2014, 08:24 PM
Post: #9
 ColinJDenman Member Posts: 102 Joined: Feb 2014
RE: Firmware 5447: Iterate() bug or feature?
(02-13-2014 07:59 PM)eried Wrote:
(02-13-2014 07:18 PM)Han Wrote:  Looking more carefully at this, do keep in mind that you are iterating a constant as z and Z are distinct.

As for the limitations, I believe those were "removed" (or better said, raised) since the most recent update. Are you running firmware 5447?

Oh, you are right, I just copy and pasted! In any case, no limitations either way.

I see my typo in the original post. Apologies. No typo on calculator, limit is genuine.
02-13-2014, 08:54 PM (This post was last modified: 02-13-2014 08:56 PM by Han.)
Post: #10
 Han Senior Member Posts: 1,817 Joined: Dec 2013
RE: Firmware 5447: Iterate() bug or feature?
I'm testing on the actual hardware and I am able to run the command just fine (in CAS view) using 10001 as the number of iterations (much larger than 1001).

ITERATE(z^2+.37-.14*i,z,0,10001) returns .360559337343-.737717172019*i

However, in the Home view, I get Error: Invalid input for more than 1000 iterations. So it looks like a this and probably more commands need to be updated, still.

Graph 3D | QPI | SolveSys
02-13-2014, 09:36 PM
Post: #11
 ColinJDenman Member Posts: 102 Joined: Feb 2014
RE: Firmware 5447: Iterate() bug or feature?
(02-13-2014 08:54 PM)Han Wrote:  I'm testing on the actual hardware and I am able to run the command just fine (in CAS view) using 10001 as the number of iterations (much larger than 1001).

ITERATE(z^2+.37-.14*i,z,0,10001) returns .360559337343-.737717172019*i

However, in the Home view, I get Error: Invalid input for more than 1000 iterations. So it looks like a this and probably more commands need to be updated, still.

My goodness. I didn't even consider that possibility. The Prime is a truly schitzophrenic beast. Thankyou for verifying the issue for me. I do hope HP works towards a far more consistent behaviour for this device. At the moment it does seem to be two products (three if you count the partial RPN) in one. Can't imagine how much code duplication exists in the firmware. Hewlett & Packard forgive them.
 « Next Oldest | Next Newest »

User(s) browsing this thread: 1 Guest(s)