HP 15C Clone

[Bunuel66]

Hi everybody.

I'm new on the forum even if a old HP fan ;-) I found that recently: http://www.rpn-calc.ch/
What did you think of it?
Wouldn't it be also an option for a 34s case?

Regards from southern France

[Daniel CLEMENT]

Hello (from France too--but more to the North!)

For one thing I can't access the video of the prototype in action. Have you seen it? Is it conclusive?

I do think it is interesting, but from the rendered pictures I would say that the matrix display is dud. The indicators appear to overlap with the commas (see pic. 16).

Does the video confirm this, or is it a side effect of the rendering?

Regards, Daniel

[Bunuel66]

I have not had any trouble to see the video. Maybe your brower is missing some plugin.
That said, the display is clearly a dot display not the quality of the original but let say it is good enough. Yes the rendered pictures show some overlap but it doesn't look that bad. May be fixed later on.
From a quality standpoint I'm expecting something like the old TI Galaxy 67 which was one of the greatest TI product.
I'm a bit dubious about legal aspects. I'm not sure the HP15C are open source and free neither the saturn emulation. I'm sure they pick all that in the 'free world' of the web but it's still questionable.
The price is not that bad, but I expect to see the actual product, not just some drawings and a breadboard. BTW I had a similar idea using an arduino kit. Justthe idea, not any time to start ;-(

Regards

[Walter B]

The video works here as well.

With respect to the item itself, a dot matrix display is a welcome step towards more flexibility. The "running" display looks like with a 7-segment LCD, however. 132 x 16 pixels would allow for alphanumeric output of two lines of text or digits like on the 42S - that's what we always wanted to get from HP. I'll contact this guy and ask a few questions in German.

Walter

[Walter B]

OK, controller info is found here. 8kB RAM :-) and 32kB flash ROM :-/ On the dm XVC, a serial port is (shall be) accessible from outside through the battery door (RS232 TTL).

I'll wait a little bit until I see the plastic parts as well.

Walter

[Jean-Michel]

Salut Daniel.

Would you mind telling us where you are from?
I'm living near Orléans and would be very pleased to meet another HP fan if you don't live to far from me, and also if you agree, of course :-)
I'm already in touch with others from foreign countries, but we will probably never meet... due to the distances!

A bientôt j'espère. ("Hope to see you soon", for non French speaking persons)

Salutations
Jean-Michel.

[Bunuel66]

Southern France = Toulouse ;-)

[Jean-Michel]

Bunuel66 : je demandais à DANIEL d'où il était, pas à toi.
J'avais bien lu que tu es du sud de la France...

(For non French speaking : nothing important.)

[Daniel CLEMENT]

Hello Jean-Michel

I'm living in Caen, Normandy at present. But Orleans is my birth place and I've been living there for 20+ years.

Salutations, Daniel

Now returning to the DM XVC discussion...

[robert rozee]

the name Michael Steinmann is attached to the video file, with someone by the same name writing some comments on facebook:


http://www.facebook.com/group.php?gid=34501949976&v=wall

i'd have delved a little deeper, but don't have a facebook account.

the item itself looks quite genuine at first glance, and not a monumental task to accomplish when you think about it. is kind of funny really, that hp will be left with egg on it's face if a young fellow in europe trumps them at re-releasing one of their more famous calculators on a shoestring budget.

[Jean-Michel]

Hello all.

I watched the video as you did.
Where are the molded parts ? I just see a printed plastic film instead of an actual keyboard with molded keys.
The website says :"the 3D models necessary for the moulds were also used to render the pictures."
Good to know, but you sure have an idea of the HUGE cost of moulds for plastic injected parts, so could someone explain to me how a single person could make this incredible investment ?
Or am I missing something ?

Kind regards from France.
JM.

[Raymond Del Tondo]

Maybe he used kinda "3D printer" , and the model is made of ... soap ?

[Jeff Kearns]

Gents,

The models to which he refers are almost certainly 3-D models generated in a 3-D CAD package like SolidWorks. Not 'soap' models from the 1970's. I can't imagine how he could fund the actual mold injection process but I like the effort he has put into this so far.

This is a very interesting project as far as I am concerned. I hesitate to criticize what this fellow (or anyone) is doing until I know more - BUT - I did fill out the form for a pre-order!! I would buy one for sure if it is as advertized. Reminds me of the 41CL project. I want one of those too!

Jeff Kearns

[Walter B]

FYI, I did file a pre-order as well for the same reason as Jeff did. If it will be as advertized, the LCD alone in such a package will be worth the $$ IMHO - maybe (!) we can even do some repurposing then ;-)

Walter

[Hubert Weikert]

Also joined the preorder queue.

Hubert

[Katie Wasserman]

I'm just curious about this .... does it make use of the 15C ROM image, if so hasn't HP asserted it's copyright of that?

[Egan Ford]

Quote:

---

I'm just curious about this .... does it make use of the 15C ROM image

---

The video posted by Cody below answers that.

[Katie Wasserman]

Egan,

I'm feeling pretty stupid, I don't see how that video answers my question. Is there some function demonstrated that results in a different result than a real 15C would? If not, how can I tell if it's an Voyager processor emulator running the 15C ROM image or a really good emulation of a 15C from scratch?

-Katie

edit....

I found this on the developer's site:

Quote:

---

runs on a battery saving LPC1114 ARM processor emulating the NUT processor

---

Now that DOES answer my question. Since it would just be plain silly to emulate that processor without running the original ROM image.

Edited: 8 July 2011, 12:56 a.m.

[Lode]

I thought I had read somewhere that the original ROM image was lost even to HP themselves. Is it not lost after all then?

[Ángel Martin]

I believe that applies to the 42S code, not the 15C.

The urban legend grows...

[Didier Lachieze]

Well I think that the 15C code is also lost, unless I'm wrong, only the 41, 71 and 75 source code were published by HP and are still available (not including RPL units here).
However 15C rom image is present in every 15C unit and can be extracted with the right tools and cleverness. Same apply to 42S.

Edited: 8 July 2011, 6:37 a.m.

[Ángel Martin]

One thing is to extract the code and another ro understand its intricacies and HW dependencies. Those would have to be touched up to make it compatible with the new CPU, or the design chganges. This is close to mission impossible without the SOURCE code, IMHO.

[sylvandb]

Quote:

---

One thing is to extract the code and another ro understand its intricacies and HW dependencies. Those would have to be touched up to make it compatible with the new CPU, or the design chganges. This is close to mission impossible without the SOURCE code, IMHO.

---

There isn't near enough code to be anywhere close to "mission impossible" if someone is familiar with the processor.

A man-day or two per k-byte of instructions will usually suffice if you have some idea of what the code does. It goes quicker if you have interactive debugging (trace, breakpoints, etc.) or slower if there was intentional obfuscation or extreme byte-saving tricks.

In some ways it's easier if it was written as assembly (or machine) code originally. At least then you have the same structure as existed in the original. On the other hand, well written high-level language usually compiles to an obvious and easily understood structure so you can get lucky that way. Unfortunately most code isn't well written, and a high-level language tends to multiply the resulting executable size by an order of magnitude or more, so usually it is easier if it was originally assembly code.

The 15c is more code than the other Voyagers, but it still isn't very big. Of course, an emulator might be easier than reworking the old code... That's what HP did.

(I used to take apart a lot of code (mostly Z80, some 8080, 6502 and 8086) and could read and write hex 8080 and Z80 code as if it were the assembler mnemonics. Spending a few weeks taking something apart, patching and putting it together will do that for you.)

[Mike Morrow]

Were one really really motivated, or if one had the appropriate secretarial resources, the HP 42S firmware could always be non-invasively **slowly*** extracted *manually* using the built-in RAM/ROM scanner, 16 hexadecimal nybbles at a time. The 64K-byte ROM would likely require one person less than ten 8-hour days to display and transcribe approximately 8000 segments of the firmware in this manner, and then double check. It would be much faster if some voice recognition process was used. I wouldn't want to do it...mind-numbing no matter what way it was done.

I've valued that memory scanner facility and appreciated HP designing easy access to it. I've only used it to check the firmware version on 17BII or 42S units I've come across or to alter the RAM location that temporarily doubles 42S speed. This scanner is a feature that is rare on HP products. It is non-existent on most calculators, including the Voyagers and the HP-41. A big disappointment found after I bought an HP 48GX was that, unlike my old HP48SX, HP had eliminated such easy owner access to machine internals.

Edited: 8 July 2011, 2:04 p.m.

[Jeff O.]

Quote:

---

...the HP 42S firmware could always be non-invasively **slowly*** extracted *manually* using the built-in RAM/ROM scanner, 16 hexadecimal nybbles at a time. The 64K-byte ROM would likely require one person less than ten 8-hour days to display and transcribe approximately 8000 segments of the firmware in this manner, and then double check. It would be much faster if some voice recognition process was used. I wouldn't want to do it...mind-numbing no matter what way it was done.

---

I do not recall the procedure in detail, or how automated it was, but there was a way to extract the 42S rom by sending it to a 48G via the infra-red output of the 42S. A special program, called "BINPRINT" if I recall correctly, was used on the 48G to capture the characters. I could try to find details if you are interested.

[Marcus von Cube, Germany]

It's INPRT or INPRINT if I'm not mistaken and you need a well equipped 48G, memory wise. You can then "IR print" from the monitor in the 42S and capture the output with the 48G. This does not work with a 49G or 50G. It's described in the docs for emu42 (hardware emulator for the high end pioneers).

[Jeff O.]

I thought it was BINPRNT (or similar) because it captured binary. But I could certainly be wrong.

[Egan Ford]

No need to guess. It is documented here: http://hp.giesselink.com/Emu28/CPROMUPL.ZIP

Quote:

---

The INPRT program has one big problem, it don't accept 0 characters, so it cannot be used for binary transfers. Raymond Del Tondo made a derivation of this program without this limitation. I called it BINPRT and is used on the HP48 for capturing the IR output of the calculators.

---

[Gerson W. Barbosa]

Katie,

Judging by the misaligned "r" in "running" I'd say it's based upon Eric Smith's Nonpareil simulator, which uses the actual HP-15C ROM image.

Gerson.

[Egan Ford]

Exactly. That is Eric's watermark.

[Katie Wasserman]

I didn't realize it was "the mark of Eric".

Anyway, I'm sure the HP lawyers will squash this project before too long. They forced Eric to remove the 15C ROM image from his site, didn't they? They go after those 12C clone makers in China too.

[Reth]

Quote:

---

I'm sure the HP lawyers will squash this project before too long.

---

What a joke - HP gladly pay lawyers big bucks to "squash" projects rather than paying a few to produce it themselves... HP actually have shown how much they care about calculators given the fact they have lost the documentation as indicated by some.
I also pre-ordered one and hope it gets produced.

[Thomas Radtke]

It's their intellectual property, so they can do with it what they want, including nothing. But isn't there an HP emulator for the iphone?

What really bothers me is that the windows Nonpareil implementation is gone 'due to legal issues'. OTOH, *x versions are still floating around. Unfortunately, I'm unable to do the port myself :-(.

[Reth]

Quote:

---

It's their intellectual property, so they can do with it what they want, including nothing.

---

Really? Did I state the opposite?

[Thomas Radtke]

Quote:

---

Really? Did I state the opposite?

---

Nope, and I meant no offense. Hopefully, I may comment a post even when there's no need to correct anything, just to add things.

[Gerry Schultz]

I know the 41C is built with the Nut uP, but does the 15C use a modified Nut? I checked the 15C web page on this website but this wasn't clear to me.

Gerry

[Eric Smith]

The Voyager calculators, including the 15C, use the 1LF5-0301 or 1LM2-0001 Nut processors, or in later models, versions of the 1LQ9, 1RR2, or 2AF1.

The HP-41 uses the 1LA5-0001, 1LF5-0002, or 1LF5-0003 Nut processors.

The 1LF5-0301 used in some Voyager calculators is a variant of the 1LF5-0002/0003 used in some HP-41 calculators. The differences are primarily in the power supply and electrical specifications.

[Gerry Schultz]

Thanks, Eric. The part numbers sure do tell the story of the development of the Nut processor.

Gerry

[Paul Dale]

Quote:

---

I can't imagine how he could fund the actual mold injection process but I like the effort he has put into this so far.

---

It might have something to do with the 80,000 would buy figure listed on hp15c.org He does link there. Selling that many would amortise the cost of the molds to a reasonable level.

- Pauli

[exschr]

Looking at the pictures it seems to me that he uses a rubber-key-mat operating the microcontact on the PCB instead of injection mold keys like we are used to on our voyagers!



just my 2c

Roger

Edited: 8 July 2011, 10:50 a.m.

[Cody]

With so much demand for a new 15c, I have to think it would be quicker and easier to repurpose the new 12c in the same way the 30b has become the WP 34s.

I understand why the 20b/30b project was chosen first to provide the most returns and functionality, but perhaps the new 12c can get the same treatment now to satisfy everyone.

It seems like with the new 12c platform, the biggest concerns of key feel and quality casing have already been solved. A new ROM and quality overlay seem simple compared so building a new calculator from scratch.

[Walter B]

Cody,

All you wrote is true. Remember, however, the very limited display of the 12C. IMHO, the display in Michael's clone is the decisive point as mentioned above. Personally, I won't repurpose a 12C with only a 7-segment display.

Walter

[Paul Dale]

I'm with Walter on this one. The seven segment display really inhibits what can be done. You'd be flat out doing better than the 15c and if you want that port an older version of non-pareil and you're done.

- Pauli

[Paul Dale]

I wonder if it would be possible to take the 12c firmware image and splice in the 15c rom.... It would depend on how good the NUT emulation HP wrote is.

- Pauli

[Marcus von Cube, Germany]

You will have to emulate the additional RAM in the 15C, too. If HP did not foresee this in the 12C+, you're lost. And how do you get at the 12C+ image in the first place?

[Paul Dale]

I vaguely remember someone mentioned reflashing their 12c with later firmware and if true, that means an image is available.

- Pauli

[Katie Wasserman]

I believe that TW makes the latest 12C+ available to an individual upon request. I don't think it's publicly available for download.

[Cody]

Here's a direct link to the youtube video of the prototype.

He has a pre-order page up with a ship date around November 2011, but without any pictures of the final product, I don't feel comfortable committing to an order.

[Csaba Tizedes (Hungary)]

Some idea and comments:

1.) If it have 132x16 pixel LCD why don't use normal (e.g. 5x7 pixel) alphabet? This right-angled dummy seven segment things seems so ugly, I think...

2.) 15C shape is very good, but, why we need to copy the original? Faceplate arrangement, labels, colors, etc is OK, but the left and right handed gray ends of case made to thinner will be looks so better than the original!

3.) Why don't made it thinner than original? 

4.) I don't want to take off battery cover to use RS232. After the 5th use of this function the users will be lost the cover. RS232 port without cover / or plugged with rubber cap the good solution, I think.

5.) Changeable faceplate OR "snap-on" faceplate for USER functions or for program LBLs? 

6.) LCD and keyboard backlight?

7.) And my old desire: 1:50 and/or 1:20, 1:100 scales on backplate or on the top side of calculator (for measuring drawings, maps, to made sketches on site...!)

Thanx for reading!

Csaba

[Reth]

Quote:

---

7.) And my old desire: 1:50 and/or 1:20, 1:100 scales on backplate or on the top side of calculator (for measuring drawings, maps, to made sketches on site...!)

---

and what about a bottle opener and a cork screw? ;)

[Dave Shaffer (Arizona)]

Quote:

---

and what about a bottle opener and a cork screw? ;)

---

The Swiss Army calculator!?

[Walter B]

Csaba,

Quote:

---

1.) If it have 132x16 pixel LCD why don't use normal (e.g. 5x7 pixel) alphabet? This right-angled dummy seven segment things seems so ugly, I think...

---

I guess Michael wanted to mimic the 15C.

Quote:

---

2.) 15C shape is very good, but, why we need to copy the original? Faceplate arrangement, labels, colors, etc is OK, but the left and right handed gray ends of case made to thinner will be looks so better than the original!

---

Please see above. And I assume he didn't want to use it for peeling potatoes ;-)

Quote:

---

3.) Why don't made it thinner than original?

---

Space?

Quote:

---

4.) I don't want to take off battery cover to use RS232. After the 5th use of this function the users will be lost the cover. RS232 port without cover / or plugged with rubber cap the good solution, I think.

---

I'm with you here.

Quote:

---

5.) Changeable faceplate OR "snap-on" faceplate for USER functions or for program LBLs?

---

Again. Though it won't help with the blue labels.

Quote:

---

6.) LCD and keyboard backlight?

---

Here I disagree. If you want a flashlight, buy a flashlight.

Quote:

---

7.) And my old desire: 1:50 and/or 1:20, 1:100 scales on backplate or on the top side of calculator (for measuring drawings, maps, to made sketches on site...!)

---

Reminds me on some old sliderules, e.g. ARISTO's 867U with one scale for cm and the other one for Royal British thumbs ;-) Most recent calculators, however, don't get close enough to the desk for proper measuring.

Just my 20 m€.

Walter