WP34S (SENDL) Is it Practical/Possibe

[BarryMead]

One can build custom RAM program groups with the assembler/library tools, but RAM space is limited. After experimenting with the WP-34S for several months, it became obvious to me that the only way to create different "Flash Libraries" is to duplicate the whole realbuild toolchain and recompile the libraries as part of a whole new calc_full.bin flash image. To me this seems a bit awkward, and I was hoping that it would be possible/practical to add a one extra SEND command to the P.FCN catalog. What I envision is a new command named SENDL which sends the "Flash Library" programs from the calculator to the emulator saving them into the wp34s-lib.dat file on the computer, or into the flash of the calculator. If this could be done then users could use the standard wp34s assembler/library tools to add/subtract functions from their wp34s-lib.dat libraries. To those who have the full toolchain and have built custom library and realbuild flash images, does this seem practical or possible? Does this sound like a good idea / bad idea?
All thoughts are welcome.

[Bit]

Check out the PSTO and PRCL commands. You'll need to process programs one by one and so they may not be as convenient to use as you'd like if updating very many programs at the same time but they may be a better option than reflashing.

[BarryMead]

(02-08-2015 06:32 PM)Bit Wrote:  

(02-08-2015 06:10 PM)BarryMead Wrote:  

Check out the PSTO and PRCL commands. You'll need to process programs one by one and so they may not be as convenient to use as you'd like if updating very many programs at the same time but they may be a better option than reflashing.

I now see how one can add/subtract a program to/from the flash memory within one calculator, but how does one keep one's collection of calculators and computer based emulators synchronized? I have 3 WP-34S calculators, and two computers running several versions of the QT and non-QT emulator and repeating these same PSTO commands on all would not be practical. I was hoping there would be some way to SYNCHRONIZE the flash library of the calculator with the emulator or the emulator with the calculator. If this capability existed, I could see where members of this community could write nice GUI "Drag-n-Drop" library management tool calling the command-line perl scripts in the tools directory to allow even inexperienced novice users to exchange/manage user programs and libraries. This is one area where the WP-34S is less user friendly than other calculators due to the complexity and difficulty in compiling/packaging user supplied programs and libraries.

[Marcus von Cube]

I know this is a problem. The present implementation of the communication protocol can only send/receive a single block of data, together with its CRC to insure the integrity of the transmitted data. The block size is limited by the size of volatile RAM which is only 4 KB. I'm using about the half of it as the communication buffer in case of a full transmission of the non volatile user RAM (SENDA). The destination memory area is only updated when the block check is ok.

Until now there is no provision of sending data in chunks which need to be assembled on the receiver side. SENDL would require a major overhaul. As a workaround, You should be able to send a single program directly from flash to the emulator (or vice versa) by selecting it from the catalog and issuing SENDP. The receiver will add it to RAM from where it can be stored in the library. I know its tedious but doable.

[BarryMead]

(02-08-2015 07:17 PM)Marcus von Cube Wrote:  I know this is a problem. The present implementation of the communication protocol can only send/receive a single block of data, together with its CRC to insure the integrity of the transmitted data. The block size is limited by the size of volatile RAM which is only 4 KB. I'm using about the half of it as the communication buffer in case of a full transmission of the non volatile user RAM (SENDA). The destination memory area is only updated when the block check is ok.

Thanks Marcus for explaining why this feature was never added. I figured there was some kind of "Technical" reason why it was not practical. The way you guys have "Thought of Everything", in this calculator, I knew that there must be a good reason why a feature this obvious was not in the mix.

Take Care, Barry

[MarkHaysHarris777]

(02-08-2015 07:22 PM)BarryMead Wrote:  

(02-08-2015 07:17 PM)Marcus von Cube Wrote:  I know this is a problem. The present implementation of the communication protocol can only send/receive a single block of data, together with its CRC to insure the integrity of the transmitted data. The block size is limited by the size of volatile RAM which is only 4 KB. I'm using about the half of it as the communication buffer in case of a full transmission of the non volatile user RAM (SENDA). The destination memory area is only updated when the block check is ok.

Thanks Marcus for explaining why this feature was never added. I figured there was some kind of "Technical" reason why it was not practical. The way you guys have "Thought of Everything", in this calculator, I knew that there must be a good reason why a feature this obvious was not in the mix.

Take Care, Barry

I was wondering a similar thing ... when you mentioned your phone jack idea... similar to the 84 plus I/O... I was wondering if there could be a SEND and a RECEIVE that would allow two WP34s units to be tied together (through the flash port) similar to the 84 plus, so that one machine could be placed in receive mode, and the other machine placed in send mode-- allowing for peer to peer cloning or backup? Is this idea NOT possible?

marcus

[BarryMead]

(02-08-2015 08:00 PM)MarkHaysHarris777 Wrote:  I was wondering a similar thing ... when you mentioned your phone jack idea... similar to the 84 plus I/O... I was wondering if there could be a SEND and a RECEIVE that would allow two WP34s units to be tied together (through the flash port) similar to the 84 plus, so that one machine could be placed in receive mode, and the other machine placed in send mode-- allowing for peer to peer cloning or backup? Is this idea NOT possible?

It is possible to link two WP-34S calculators equipped with the 2.5mm headphone jack using a male to male 2.5mm crossover cable (like the ones made for CASIO calculators) Using the following wiring:

Tip (Calculator 1 RX) -> Ring (Calculator 2 TX)
Ring (Calculator 1 TX) -> Tip (Calculator 2 RX)
Sleeve (Calculator 1 Gnd) -> Sleeve (Calculator 2 Gnd)

I ordered a pre-made "Casio Data Transfer Cable" from here but it isn't due to arrive until next month. (On a slow boat from China I suspect) I will let you know if it is wired properly when it arrives.

Using this cable you can execute SENDA, SENDP, SENDR, or SENDΣ, on the sending calculator and RECV on the receiving calculator. But there is no complete CLONE feature. If you use the "Samba Boot" keyboard command to flash both calculators with the same calc_xtal_full.bin (or whatever) flash file, they will have the same flash libraries, but their RAM's may still be different after such an operation, because the "Backup Partition" may be different between one calculator and the other. If you completely erase the flash with the "ERASE" button (on your flash box) then both calculators will be IDENTICAL, and their RAM's will be ERASED (empty).
Does this answer your question?

[Marcus von Cube]

(02-08-2015 08:00 PM)MarkHaysHarris777 Wrote:  I was wondering a similar thing ... when you mentioned your phone jack idea... similar to the 84 plus I/O... I was wondering if there could be a SEND and a RECEIVE that would allow two WP34s units to be tied together (through the flash port) similar to the 84 plus, so that one machine could be placed in receive mode, and the other machine placed in send mode-- allowing for peer to peer cloning or backup? Is this idea NOT possible?

If you look at post #3 of the following old thread, you see that it IS in fact possible:

http://www.hpmuseum.org/cgi-sys/cgiwrap/...ead=186826

You just need a cross over cable (Casio comes to the rescue here, TI cables are straight through.)

[MarkHaysHarris777]

Barry, Marcus, thank you. This is good news; I appreciate it. I suppose the crossover cable of course does not require the 1k resistors; no need, right? Yes, Barry when you get your casio cable let me know, please, how things check out; easy enough though to make ones own patch cable. Thanks again.

cheers
marcus

[matthiaspaul]

Perhaps somewhat related, I can think of a use case of the WP 34s being attached to a desktop or notebook computer via its USB / TTL serial link and used as a smart numerical keypad (similar to the HP CalcPads, but better) or kind of "remote control".

The idea would be to run the emulator on the PC to see and possibly log the data entry along with the results, and to make the data available to other applications via copy & paste, but to use the real calculator for the input (more convenient than using the PC keyboard, in particular without a numpad).
The calculator could be brought into this mode via a special startup hotkey like ON+R ("remote"). In this mode, it would echo any detected keystrokes in a rudimentary small packet format, so that f.e. the emulator could react on them in real time. Likewise, if the user would enter data into the virtual calculator (perhaps also using the mouse), the emulator would echo this to the calculator as emulated keystrokes as well. Typically, both calculator instances would display their local display contents, but there should be some means to send / receive the display contents of the opposite party as well, overriding the local display. At minimum, the calculator could blindly emit detected keystrokes and otherwise work as normal, unless it receives messages with emulated keystrokes or display contents from the other side, which than would temporarily override the local data.

I haven't looked at the sources yet, but wonder if implementing something like this would be possible without consuming too much flash memory given that an ON+D debug mode already exists. What do you think?

Greetings,

Matthias

[BarryMead]

(02-08-2015 09:54 PM)MarkHaysHarris777 Wrote:  Barry, Marcus, thank you. This is good news; I appreciate it. I suppose the crossover cable of course does not require the 1k resistors; no need, right? Yes, Barry when you get your casio cable let me know, please, how things check out; easy enough though to make ones own patch cable. Thanks again.

cheers
marcus

Marcus: My pre-made Casio crossover cable arrived today, and it is wired properly (like shown above).
I connected my two calculators to each other and transfered my first program DIRECTLY from one
calculator to the other today. (No computer required).

Thought you would want to know, Barry

[MarkHaysHarris777]

(02-10-2015 01:26 AM)BarryMead Wrote:  I connected my two calculators to each other and transfered my first program DIRECTLY from one
calculator to the other today. (No computer required).

Thanks Barry... looking forward to trying this myself. Casio cable is on the way; was noticing that the photo of the cable had suppression inductors built in just behind each plug... is that how your's arrived?

marcus

[BarryMead]

(02-10-2015 05:41 AM)MarkHaysHarris777 Wrote:  

(02-10-2015 01:26 AM)BarryMead Wrote:  I connected my two calculators to each other and transfered my first program DIRECTLY from one
calculator to the other today. (No computer required).

Thanks Barry... looking forward to trying this myself. Casio cable is on the way; was noticing that the photo of the cable had suppression inductors built in just behind each plug... is that how your's arrived?

marcus

Yes mine looks exactly like the photo. It is extremely well made and works flawlessly. The 90 Degree
plug puts less strain on the connector than a straight in cable does. That is especially true compared to
my programming connector with the long 3.5mm to 2.5mm adapter dangling out of the calculator.

I liked the 90 degree plug so much that I ordered a 90 degree 2.5mm male to 3.5mm female pigtail
type adapter to reduce the strain during flashing as well. here

[MarkHaysHarris777]

(02-08-2015 08:21 PM)Marcus von Cube Wrote:  You just need a cross over cable (Casio comes to the rescue here, TI cables are straight through.)

Marcus,

Thanks much for the tip! I do apologize, I received a message from Barry this morning indicating that the tip-ring-gnd (and Casio cable) idea was originally yours, and that you deserve the credit. I agree! So, again, thank you very much. I really think at this point it is the simplest and most solid approach to not only the update problem, but also the peer-to-peer problem; great solution/

:-)

Cheers
marcus

[BarryMead]

(02-10-2015 02:48 PM)MarkHaysHarris777 Wrote:  

(02-08-2015 08:21 PM)Marcus von Cube Wrote:  You just need a cross over cable (Casio comes to the rescue here, TI cables are straight through.)

Marcus,
Thanks much for the tip! I do apologize, I received a message from Barry this morning indicating that the tip-ring-gnd (and Casio cable) idea was originally yours, and that you deserve the credit. I agree! So, again, thank you very much. I really think at this point it is the simplest and most solid approach to not only the update problem, but also the peer-to-peer problem; great solution/

The WHOLE 2.5mm HEADPHONE JACK idea (not just the crossover cable) was Marcus von Cube's idea. I just found a roomier place in the case to glue the jack down so it fits with less cutting of the case.

[walter b]

Cf. p. 286 of the spiral bound manual. Advantage to those who can read!

d:-)

[d b]

(02-11-2015 04:47 PM)walter b Wrote:  Cf. p. 286 of the spiral bound manual. Advantage to those who can read!

d:-)

Walter; There is one problem with the threaded mode. I can't really tell which member of this forum you are accusing of being illiterate, and that's exactly what your last sentence reads like in English. Care to enlighten me?

[walter b]

(02-12-2015 04:02 AM)Den Belillo (Martinez Ca.) Wrote:  There is one problem with the threaded mode. I can't really tell which member of this forum you are accusing of being illiterate, and that's exactly what your last sentence reads like in English. Care to enlighten me?

No, Sir. I can see it, you should be able to see it as well. BTW, said link on p. 286 is in for many months - it's even found on p. 166 of v3.1 (PDF).

d:-)

[d b]

(02-12-2015 05:46 AM)walter b Wrote:  

(02-12-2015 04:02 AM)Den Belillo (Martinez Ca.) Wrote:  There is one problem with the threaded mode. I can't really tell which member of this forum you are accusing of being illiterate, and that's exactly what your last sentence reads like in English. Care to enlighten me?

No, Sir. I can see it, you should be able to see it as well. BTW, said link on p. 286 is in for many months - it's even found on p. 166 of v3.1 (PDF).

d:-)

Sorry Walter. I don't see it. I believe you that "said link on p. 286 is in for many months - it's even found on p. 166 of v3.1 (PDF)." What I want to know is which active member of this forum does not have the "Advantage to those who can read". I'd like you to spell it out plain for me.

[Mark Hardman]

(02-12-2015 06:08 AM)Den Belillo (Martinez Ca.) Wrote:  What I want to know is which active member of this forum does not have the "Advantage to those who can read". I'd like you to spell it out plain for me.

Based on the threaded view, the comment was directed at MarkHaysHarris777. This poster's frequent "stream of consciousness" (bewusstseinsstrom?) posting style has been the source of Walter's criticism in the past.