I was bored and found a HP-27S

[jebem]

(06-27-2014 10:52 PM)Dave Frederickson Wrote:  If no one could dump the ROM then how does go27S run? Have you tried?

Do you mean the Google Play app?
No, I haven´t tried because they say we need to obtain the rom dump by ourselves, but I don't see how can it be done.
I'm using a real physical HP-27S machine.

[jebem]

(06-27-2014 10:28 PM)Katie Wasserman Wrote:  

(06-27-2014 10:18 PM)jebem Wrote:  Because I felt the need to do it, I have put together this small listing with links to HP-27S references here in the MoHPC:

Wow! Lot's of interesting stuff, all the way back to 2000.
Thank you for this collection.

You are very welcome, Katie!
I had to do it, in order to get some knowledge about this calculator, so I'm doing this research since two weeks ago, compiling and analyzing as much information as I can.
However I must confess that the way Thomas and others here take advantage of the Solve feature really puzzles me...

Meanwhile I have updated it with a few more related links, all of them from the MoHPC.

[Dave Frederickson]

Yes, Oliver De Smet's emulator. Have you been successful dumping other ROMs? Or, back to my previous question, have you tried to dump the ROM?

Dave

[jebem]

(06-27-2014 11:01 PM)Dave Frederickson Wrote:  Yes, Oliver De Smet's emulator. Have you been successful dumping other ROMs? Or, back to my previous question, have you tried to dump the ROM?

Dave

No, never did it. Despite working in the past with low level hardware reverse engineering I never did it for HP calculators. This is a kind of a new hobby for me, you know.

EDIT: But I will have a look into it, just for fun, when I have time.

[Dave Frederickson]

According to this post this capability has been around for 10 years. Please let me know if you get around to trying it.

Dave

[Mark Hardman]

(06-27-2014 10:46 PM)jebem Wrote:  but from my research in the Net, apparently no one could dump it...

Your research is faulty.

I had no problem dumping the HP-27S ROM following the instructions and utilities found here:

CPROMUPL.ZIP

[Dave Frederickson]

Apparently I was asking the wrong person.

Hey, Mark ...

[Don Shepherd]

(06-27-2014 10:58 PM)jebem Wrote:  However, I must confess that the way Thomas and others here take advantage of the Solve feature really puzzles me...

That is understandable. Thomas and I have been working with the solver for years and over that period we have found tricks to capitalize on its features to do things that would not be possible in a simpler solver application, and also to do things that the designer probably didn't count on (like exiting a loop early by doing an illegal operation like divide by 0).

A simpler solver is able to solve equations, like Pythagorean theorem:
SQ(A)+SQ(B)=SQ(C)

Enter value for A, press A. Enter value for B, press B. Then press C and it calculates the value of C. If you enter a number before pressing the menu key, it just assigns the value you entered to that variable. But if you don't enter a value before pressing the menu key, it evaluates the equation and tells you the value the variable must have in order to solve the equation. If you don't have an equals sign in your equation, it tells you the value that the variable must have so that the equation equals 0.

Read the manual to see how the S command works (Solving for). The general form of that is: IF(S(A): calculate A here - A: calculate B here - B).

Here is a simple solver equation for playing the famous "high/low" guessing game. The calculator thinks of a number between 0 and 99, and you try to guess it.

First, after CALC'ing the equation, press INIT. This sets the number of guesses to 0 and loads N with a random number, 0-99. [this takes advantage of the RAN# solver function on the 27S and 19b; on the 17b/17bii, there is no random number function so getting a random number takes more work, but it can be done using CTIME]

So, INIT generates the random number you will try to guess. So enter your first guess and press GUESS. That just stores the value you entered in the variable GUESS, it doesn't execute any code. Then press EVAL for evaluate. This executes code. It increments the guess counter G by 1. It then checks your guess against the random number N. If it is equal, it returns the number of guesses you required (and you are done). If your guess is too low, it returns 0, else your guess is too high and it returns 99.

This is what the code would look like on the 27S:

Code:


IF(S(INIT):L(G:0)XL(N:IP(100XRAN#))-INIT:
0XL(G:G(G)+1)+IF(GUESS=G(N):G(G):IF(GUESS<G(N):0:99))-EVAL)

Unless I have made an error above (which is possible since I don't have a 27s to test on), this should let you play the high/low game. You should always guess the number within 7 guesses if you use a binary search approach, by the way.

This is a simple example of using the solver to play a game, something not possible on a traditional equation solver.

[Marcus von Cube]

(06-27-2014 10:18 PM)jebem Wrote:  Because I felt the need to do it, I have put together this small listing with links to HP-27S references here in the MoHPC:
...

And a few ones on the HP-17B series (similar to HP-27S Solver):
...

This is probably worth an article.

[jebem]

(06-28-2014 12:05 AM)Mark Hardman Wrote:  

(06-27-2014 10:46 PM)jebem Wrote:  but from my research in the Net, apparently no one could dump it...

Your research is faulty.

I had no problem dumping the HP-27S ROM following the instructions and utilities found here:

CPROMUPL.ZIP

Hi Mark,
What I meant is that anyone can find copies of the ROM images for other models like the 42S or 28S, but it seems that no one can find it for the 27S model... and a search in Google reveals some people claiming that they failed dumping the 27S despite having success with other models...
But maybe you can point us to a link where we can download it?

I have read that article from Christoph Gießelink a couple of times, but if you read my posts you will see that I just acquire this 27S for the first time... so I had no time to investigate ROM dumping at this point, besides it seems it requires a HP-48 series machine as a man in the middle (did you try to use a HP-50G?).
However I understand that it is handy to be able to run the 27S in a emulator, of course

[walter b]

(06-28-2014 07:17 AM)Marcus von Cube Wrote:  

(06-27-2014 10:18 PM)jebem Wrote:  Because I felt the need to do it, I have put together this small listing with links to HP-27S references here in the MoHPC:
...

And a few ones on the HP-17B series (similar to HP-27S Solver):
...

This is probably worth an article.

+1

d:-)

[jebem]

(06-28-2014 12:35 AM)Don Shepherd Wrote:  

(06-27-2014 10:58 PM)jebem Wrote:  However, I must confess that the way Thomas and others here take advantage of the Solve feature really puzzles me...

That is understandable. Thomas and I have been working with the solver for years and over that period we have found tricks to capitalize on its features to do things that would not be possible in a simpler solver application, and also to do things that the designer probably didn't count on (like exiting a loop early by doing an illegal operation like divide by 0).
(...)

Thanks, Don!

Quote: "(like exiting a loop early by doing an illegal operation like divide by 0)"

Impressive stuff I would say.
Yesterday I have ordered the MoHPC DVD v7 set that will include the 27S documentation.

[Dave Frederickson]

(06-28-2014 08:02 AM)jebem Wrote:  Yesterday I have ordered the MoHPC DVD v7 set that will include the 27S documentation.

A wealth of information. Don't forget to get a copy of the PPC DVD and if you're going to get involved with the 41 or HP-IL, TOS has a DVD, too.

Dave

[jebem]

(06-27-2014 11:12 PM)Dave Frederickson Wrote:  According to this post this capability has been around for 10 years. Please let me know if you get around to trying it.

Dave

Job done!

I have now my own copy of the HP-27S ROM image dumped from my physical calculator s/n: 3122S04492.

I spent this Saturday's morning collecting and organizing the required information from the Christoph Giesselink site.
Then I had to make a RS232 serial cable to connect the man-in-the-midle HP-48G+ to my PC.
Only then I did the ROM transfer.

I have to say that this is no challenge at all. Easy job in fact. Worked at first try.
Thanks to Christoph Giesselink and others work!


ROM transfer via IR from HP-27S to HP-48G+




Transfer finished. The HP-27S did a warm boot. The HP-48G+ showing the new string.




Transferring the ROM to a PC using XModem protocol. Windows 7 64-bit running Hyperterminal.




Striping the extra data from the ROM file at PC.




ROM file checksum OK.




Testing on the excellent Christoph Gießelink's Emu42 running the HP-27S ROM image.

[Dave Frederickson]

I'll bet you had fun! Was it not a worthwhile experience? Kind of a rite of passage.

Dave

[jebem]

(06-28-2014 05:33 PM)Dave Frederickson Wrote:  I'll bet you had fun! Was it not a worthwhile experience? Kind of a rite of passage.

Dave

Yap, that was a good time
I have sent you a PM.

[franz.b]

Hi everyone, I try to resurrect this old thread.
I have a 27s that does not turn on anymore and I would like to try to fix it because it was my father's, but for now my efforts have been in vain.
I was able to open it, but the problem does not seem to be in the rubber strip, which, although in good condition, I replaced .. I obviously cleaned all contacts, even those of the screen, but the calculator does not turn on.
The pcb is very clean and the copper shining, but there is a lot of dirt on the keyboard. opening it is extremely difficult and destructive, is there any way to check if the keys produce sufficient electrical contact?
Here I found a lot of useful information but but I can't measure the resistance of the keyboard, I don't think my analog instrument is suitable, the indicator barely moves ... or is it the fault of the dirt on the contacts? I also don't understand if the protective film of the screen is also polarized, or if the polarized film is applied to the glass. The external part, superimposed on other LDCs screens does not produce any effect.. could it be that I simply cannot see that the calculator is switched on?

Thanks in advance

[Massimo Gnerucci]

From a reset condition it should wake up by beeping since beeper is set on by default.
To be sure try a division by 0, this should make it beep.

If you don't hear anything (and the beeper is connected to the PCB) it should still be dormant.

[franz.b]

(11-06-2020 08:42 PM)Massimo Gnerucci Wrote:  From a reset condition it should wake up by beeping since beeper is set on by default.
To be sure try a division by 0, this should make it beep.

If you don't hear anything (and the beeper is connected to the PCB) it should still be dormant.

Infinite thanks! it makes a sound!
I connected an external piezoelectric buzzer to the springs and short-circuited between points P1 and P2 which are the contacts of the ON button. To tell all what happened, the beep came out only by touching P2 connected to nothing else
Now you have to understand why you don't see anything on the display ... Is there any possible test to do on the display alone to see if the problem is there?
Any suggestions for fixing the PCB on the keyboard / screen contacts without twisting the hooks (which otherwise I will end up breaking in attempts)?
Can the display depend on the components mounted on the PCB? (I don't know if they are capacitors or resistors ... but they are the only parts that can be replaced). I don't think the oscillator died as a sound was emitted ...

[jebem]

(11-08-2020 06:39 PM)franz.b Wrote:  Now you have to understand why you don't see anything on the display ...

Just to eliminate that possibility, did you increase the LCD contrast to the maximum position?
Sometimes, specially when the power supply is running below the nominal value, the LCD appears to be blank when in fact what is required is to adjust its contrast.

If I remember correctly, on the Pioneer machines the LCD contrast is adjusted by pressing and holding the CLR key and then press the + key several times to increase it, and - key to decrease it.