Nonpareil news



#2

For those of you that are interested in Nonpareil, but don't read my blog, I thought I'd mention some of the recent progress.

I've added additional models to the simulation, including the 22, 27, 29C, and 67 (sans card reader). The 29C and 67 (as well as the 19C and 97) use a highly optimized label search loop for performance, and it depends on some very subtle and undocumented behavior of the ACT chip. I still don't really know exactly how the ACT behaves under these circumstances, but based on analysis of the code, I was able to come up with a hack that makes it work.

The 67 seems fully function with the exception of the card reader
and the PSE function. PSE seems to complete immediately rather than delaying, and I'm guessing that this is due to a test for a magnetic card being inserted.

Note that Nonpareil provides continuous storage for all models, and it is possible to keep multiple state save files, so one can effectively save 67 programs and data despite the lack of card reader simulation.

I've fixed a subtle bug in my sound code, so now the 41C TONEs and BEEP work reasonably well on both Linux and Windows.

The KML syntax for specifying keyboard mapping has been improved considerably. Now the GDK Keysms used to name the host keyboard keys can be specified by name rather than obscure numeric values.

Maciej Bartosiak has made a lot of improvements to the rendered images used for the Voyager calculators, as well as developing new ones for the Spice series. I'll be incorporating those, and I'm working on better visual effects for button highlight (when the cursor is over a button but not pressed) and the actual button press.

These improvements will be in the next official release of Nonpareil, which I hope to complete in January.

Eric


#3

Great news!

Any word on the "housing" to put these into? :-)


#4

Not much to report on that. Richard Ottosen has acquired some ABS stock and is going to try assembling a case out of that. The one I showed at the conference was acrylic.

It seems to be difficult if not impossible to buy sheets of ABS that are smooth on both sides. Normal ABS sheet stock has a texture on one side. But Richard thinks that this may work OK for his assembly technique.

I hope to make a lot of progress on the firmware over the holidays.


#5

ABS sheet smooth on both sides should not be too hard to source as textured sheet is more costly to produce. What's the desired thickness and color?

Edited: 12 Dec 2005, 10:47 p.m.


#6

0.63 inch, in 12x24 inch sheets or multiples thereof. (One 12x24 inch sheet is laser cut to make one calculator.) Color isn't important for prototypes, but we expect the most desirable color will be black.

I just did a Google search for "abs sheet", and the very first listing has it in either smooth or one side textured, so I'm not sure why Richard had so much trouble finding it. Maybe it just isn't carried by the retail establishments.

#7

Any progress in getting a 10C ROM to work on nonpareil?

Arnaud


#8

I haven't yet got a 10C ROM dump. But I expect to have one soon.

Out of curiousity, why are you interested in the 10C? The only reason I plan to add it is for general completeness.


#9

It would be great for those of us that lack a 10C in our collections, just to know exactly how badly they sucked. 8)


#10

Hi, Howard:

Howard posted:


"It would be great for those of us that lack a 10C in our collections, just to know exactly how badly they sucked. 8)"

    They didn't suck at all, the HP-10C has all the physical quality, ergonomy and durability of all other Voyager-series models, plus all the elegance of a well-thought, non-cluttered keyboard.

    You'd be hard-pressed to get 10% of such quality and elegance in newer or contemporary HP models, let alone other brands.

Best regards from V.

#11

I agree with Valentin about the 10c not really stinking all that badly :-)

but it was a shame that HP "crippled" its abilities so much. I think that's probably what people mean by complaints about the 10c.

There are many programs that would not fit on the 10c that ran just find on the old HP25, for example, because of the scarcity of memory on the 10c.

I would really like to know what crazy business case someone at HP made to introduce the 10c anyway.

There just wasn't enough price differential between the 10c and 11c to convince many people to buy it over the slightly more expensive 11c. Hence the relative scarcity of the 10c.

That said, Valentin is quite correct...don't go talking bad about my 10c! :-)
Gene


#12

Gene,

I remember buying an HP10C and HP11C whn they came out. I quickly realized that the programming features of the HP10C were ver similar to those of the HP-55 -- that is ... no subroutines, no labels for program branching. That was a let down in the era of King HP41CV.

I found a buyer for the HP10C at the company I was working for and kept the HP11C.

In the last few years I bought two HP10C from eBay to complete the Voyager set.

Namir

#13

Ahem, sorry, Valentin, I should have said "..sucked as programmables compared with the 11C and 15C." After all, we won't get to experience the physical quality via a simulation, no matter how faithful.

I know you've written articles on how to overcome the limitations of the 12C, which is similarly crippled as a programmable, but the fact remains, the limitations are there to be overcome.

Now that I look, I don't see a "Long Live the HP-10C" on your site. How does the 10C stack up? (So to speak. 8)


#14

Hi, Howen:

Howen posted:

" I don't see a "Long Live the HP-10C" on your site. How does the 10C stack up? (So to speak. 8) "

    Pretty well, asuming it's still boxed.

    There's also no "Long Live the HP-16C" there, though it will be in the very next issue of Datafile, due in a few days.

Best regards from V.
#15

My only interest is for completeness. I have the real stuff if I want it but it won't last for ever...
Anyway, I wouldn't use it :-)

And if you need it, I have a slightly battered 10C that I could send you for surgury if you don't have any other

Arnaud

I still have to send you a postcard :-(

Edited: 15 Dec 2005, 3:11 a.m.

#16

Questions (maybe the answers to those should be in the README file?)

1. When emulating an HP41, if I have a rom image, how do I turn it into a .mod file?

2. Is it possible to change the default behavior of "save" so that it doesn't actually create a new image of "nonpareil.exe"?

Thanks

/ji


#17

Apologies if this is old ground. I just recently looked at this myself, so it was new to me.

The V41 source includes a utility to go the other way,.mod->.rom, but as far as I know, there's no utility to create a MOD file from one or more roms. It's not a completely automatic process in that direction. If it's your own module, you need to write an info file for it. Here's an excerpt from the info file for the CCD module in the V41 distribution:

FILE NAME: C:\Program Files\V41\MOD\Ccd.mod
FILE FORMAT: MOD1
TITLE: CCD Module
VERSION: B
PART NUMBER:
AUTHOR: W&W Software Products GmbH
COPYRIGHT (c) W&W Software Products GmbH
LICENSE: GNU General Public License
COMMENTS:
CATEGORY: Custom peripherals
HARDWARE: None
MEMORY MODULES: 0
EXENDED MEMORY MODULES: 0
ORIGINAL: Yes - unaltered
APPLICATION AUTO UPDATE: No - do not update this file
NUMBER OF PAGES: 2

ROM NAME: CCDL-1B
ROM ID: WW1B
PAGE: May be in more than one location
POSITION: In lower relative to upper page
PAGE GROUP: 1
BANK: 1
BANK GROUP: 0 - not grouped
RAM: No
Write Protected: No or Not Applicable
FAT: Yes
XROM: 9
FCNS: 64
XROM Addr Function Type
09,00 754B -W&W CCD B 4K MCODE Programmable
09,01 7764 B? 4K MCODE Programmable

And so forth. That file contains a second ROM section for the upper 4K page. Both end up with something like this (taken from the second one in the CCD file.)

11,41 169C5             8K MCODE (Not decoded)
INTERRUPT VECTORS:
Pause loop: 000
Main running loop: 000
Deep sleep wake up, no key down: 000
Off: 000
I/O service: 3C3
Deep sleep wake up: 2D3
Cold start: 000
CHECKSUM: 37F (CORRECT)

All of that corresponds to how HP-41 roms are organized, which could be generated from the rom itself. But the metadata at the front needs to be filled in by a human being, I think.

Exactly how the info and rom files are glued together in the .MOD is a matter for the source code to tell.

If it isn't your own ROM, the V41 package will likely have what you need. Otherwise the other TOS resources will probably cover it.

Edited: 13 Dec 2005, 8:25 p.m.

#18

Quote:
Is it possible to change the default behavior of "save" so that it doesn't actually create a new image of "nonpareil.exe"?

I've never seen it do any such thing, nor do I have any idea how it could. The save commands writes a state save file with a ".nst" extension.

#19

Quote:
When emulating an HP41, if I have a rom image, how do I turn it into a .mod file?

Currently Nonpareil only supports hard-addressed ROMs. If you create a .mod file that forces the address (e.g., 8000, A000, C000, or E000 for the base of an 8K ROM), it should work.

Better .mod file support is on my TODO list.

#20

I can't wait - the (potential) new artwork looks stunning, keep up the good work...

Mike T.


#21

Thanks! But all credit for the artwork goes to:

  • Maciej Bartosiak - for 41C, Voyager series, and Spice series
  • Xavier Théry - 67
  • Dave Hicks, MoHPC - for photographs used for other models

My own graphics skills are feeble at best, so it's a good thing that I know how to program. :-)

I'm trying to replace the horrible white box key highlighting that's been used up until now in Nonpareil (which is a side effect of the GTK+ button widget) with a better looking effect.

Christoph Gießelink has sent me his updated NSIS script for Nonpareil, so the next release will have a real installer and uninstaller.


#22

Speaking of installers, It may be a matter of me using the SVN sources, but 'scons install' seems to miss rom/67.obj.

Other than that, I think the 67 support is "PDC" - Pretty Darned Cool, like the rest of Nonpareil. Thanks!


#23

Thanks, I just added it to rom/SConscript.

Looks like the cyclic dependency problem with images has come back with the latest release of SCons. Sigh. I suppose I should try to reduce it to as small a test case as possible, and ask about it on the SCons mailing list.


Possibly Related Threads...
Thread Author Replies Views Last Post
  Trivial news of the 43S (no Prime) Walter B 117 2,913 11-22-2013, 03:26 AM
Last Post: Raymond Del Tondo
  Nonpareil Rom files davorin 2 177 09-20-2013, 06:20 PM
Last Post: Didier Lachieze
  News from the 43S Walter B 78 1,708 07-08-2013, 10:33 AM
Last Post: Victor Koechli
  Great news - Vicinno's HP 15C Scientific Calculator iPhone app is FREE now John 21 901 06-07-2013, 05:49 AM
Last Post: Mike (Stgt)
  Nonpareil + HP10C Mike (Stgt) 0 109 05-13-2013, 04:58 AM
Last Post: Mike (Stgt)
  Good news for HP-emulator fans! :-) fhub 44 1,622 02-19-2013, 04:11 PM
Last Post: Marcus von Cube, Germany
  Some news for old HP emulator fans! ;-) Olivier De Smet 3 232 02-10-2013, 08:47 PM
Last Post: Earl Kubaskie
  Sad news Eric Smith 7 383 02-08-2013, 09:27 AM
Last Post: Eddie W. Shore
  Latest DIY calculator news Eric Smith 17 579 09-25-2012, 10:16 AM
Last Post: uhmgawa
  Good news :-) fhub 12 422 09-24-2012, 08:10 AM
Last Post: Frank Boehm (Germany)

Forum Jump: