DM 42

[Thomas Okken]

(05-26-2017 12:52 PM)toml_12953 Wrote:  

(05-26-2017 07:34 AM)Paul Dale Wrote:  I'm fine with the 42S's behaviour. CPXRES and REALRES are required.

The question was more: should a calculator switch between real and complex mode automatically? It shouldn't unless told to. This works both ways: (0 + i)² should give (-1, 0) not -1.

Maybe for you it shouldn't but for me, it should always give an answer in either the real or complex plane as appropriate. If I could pick only one method, I'd have the calculator always give a numeric answer whether real or complex. If I was working a real-world problem and saw a complex result, I'd know there was no real solution. If the calculator always reported an error when given real inputs, there would be no way of knowing what the actual answer was.

Tom L

That's exactly what CPXRES mode does: with real inputs, you get real results if they exist, and complex results if necessary; with complex inputs, you always get complex results. This is the default mode on the HP-42S and in Free42, and, I assume, DM42 as well.

[Guenter Schink]

(05-26-2017 01:36 PM)Ángel Martin Wrote:  

(05-24-2017 01:53 PM)Guenter Schink Wrote:  4. In Alpha-Mode haven't found a way to put in numbers

I have the same issue: can't figure out how to enter numbers in ALPHA...

On the real 42 is simply pressing the number key, so here I had assumed the SHIFT key would be needed since the direct letters are assigned to the keys - but that does *not* work.

Is this a bug?

I don't think it's a bug. It's beta status
Michael und I had a lenghty discussion last year about a concept of alpha entry. His objective was to finalize the hardware first. Software will be tuned later.
Right now he is on vacation.
Günter

But I am not at all really involved in the development.

[Massimo Gnerucci]

Wow, oh wow, oh wow wow wow!

The wait is over.

Right size and weight. Keys not on par as my beloved 41, but not really bad at all!
A nice pouch, and how speedy...

Well done Michael.

[pier4r]

Thomas do you get a fee with the Dm 42?

If yes I may plan a future purchase, because having the free 42 on the phone is ok but a real keyboard is better.

[Thomas Okken]

(05-26-2017 06:24 PM)pier4r Wrote:  Thomas do you get a fee with the Dm 42?

If yes I may plan a future purchase, because having the free 42 on the phone is ok but a real keyboard is better.

I don't get royalties for DM42, but don't let that stop you from buying one. Free42 is GPLv2, so as long as they publish the source code, including all the additions and modifications they made, everything is fine.

[toml_12953]

(05-26-2017 04:36 PM)Thomas Okken Wrote:  

(05-26-2017 12:52 PM)toml_12953 Wrote:  Maybe for you it shouldn't but for me, it should always give an answer in either the real or complex plane as appropriate. If I could pick only one method, I'd have the calculator always give a numeric answer whether real or complex. If I was working a real-world problem and saw a complex result, I'd know there was no real solution. If the calculator always reported an error when given real inputs, there would be no way of knowing what the actual answer was.

Tom L

That's exactly what CPXRES mode does: with real inputs, you get real results if they exist, and complex results if necessary; with complex inputs, you always get complex results. This is the default mode on the HP-42S and in Free42, and, I assume, DM42 as well.

OK, now I feel like a fool (not the first time!) CPXRES mode exactly what I would use. In fact, I *will* use it on the DM42 when it's shipping in final form.

Tom L

[Dave Frederickson]

(05-19-2017 05:29 PM)Dave Frederickson Wrote:  Where?

... were beta units announced?

[Geoff Quickfall]

When the original announcement last year (June?) delivery 2016 and was subsequently postponed the beta test versions were mentioned. Since then more names were added to the list. The original orders (30?) were contacted as to beta testing.

My memory is vague on this but that is how I ended up in the first thirty for delivery postphoned until this year.

Cheers

[Dave Frederickson]

(05-26-2017 08:59 PM)Geoff Quickfall Wrote:  When the original announcement last year (June?) delivery 2016 and was subsequently postponed the beta test versions were mentioned. Since then more names were added to the list.

No, not "The List"! I've observed that there's no official standard for "The List". Take "The Book" List for example. Who knows how long various threads have contained partial lists. Are they compiled somewhere? To this day I've no idea how I got on the short list for a Pre-Production FRAM71 module.

Thanks, Geoff.

[Geoff Quickfall]

No idea where the "list " is. I remember emailing Michael last year and asking if he needed beta testers as I helped beta test the HP 41 clone. That was at the announcement that the DM 42 was a project.

The book list, well I have the original from HHC2014 and emails that add to it. That list resides with me, as to the book it is a WIP that is close to completion.

Problem with that is family and work, family including two elderly parents (over 90) at a very disruptive time of their lives. Everything this last year has been delayed.

Geoff

[Ángel Martin]

(05-26-2017 04:53 PM)Guenter Schink Wrote:  

(05-26-2017 01:36 PM)Ángel Martin Wrote:  Is this a bug?

I don't think it's a bug. It's beta status
Michael und I had a lenghty discussion last year about a concept of alpha entry. His objective was to finalize the hardware first. Software will be tuned later.

ok, it'd be great to have an SSB/RSB-like document with all the known functionality not implemented yet, will save quite some time poking around ;-)

[Thomas Okken]

I just released Free42 2.0.3, which fixes acos(cos(1+i)) and asin(acos(atan(tan(cos(sin(1+0i)))))), and a few other edge cases.

I found that some of the incorrect branches were because of negative zeros. I added code to test for those where I found them, but the more rigorous approach would be to change the < and > operators to not consider -0 < 0. That would make the internal semantics of Free42 match those of the HP-42S. I'll do that in the next release.

(05-26-2017 01:17 PM)jch Wrote:  I had to use a paper clip to reset my DM42 after I tried to create a 100 x 100 matrix.
That yields to a 'Malloc Error' and I had to reset the calculator. I first tried a gentle pressure, but it requires a firm one.

That is something you should definitely report to SwissMicros. When memory allocation fails while trying to create a matrix that is too large, Free42 handles the NULL returned by malloc(), and cleans up gracefully and returns the "Insufficient Memory" error message, just like the HP-42S. It should not freeze!

Perhaps they decided to catch malloc() failures at the system level because there are a few places in Free42 where it doesn't handle memory allocation failures properly, but the thing to do is to fix those places *in Free42*, not make malloc() do weird things when it can't satisfy an allocation request.

[rprosperi]

(05-28-2017 08:57 PM)Thomas Okken Wrote:  I just released Free42 2.0.3, which fixes acos(cos(1+i)) and asin(acos(atan(tan(cos(sin(1+0i)))))), and a few other edge cases.

Of course you did! I just downloaded and installed 2.0.2 last night onto 2 PCs to use for testing with the DM42.

Are state files, etc. compatible so I can just replace the .exe files (Windows).

Also, FYI, the README.txt file in the .zip still says you can keep the state and data files in the install folder by creating a file named 'portable' in the install folder, but it appears this option is deprecated. Not a big deal, just letting you know.

Thanks Thomas for continually improving Free42!

[Thomas Okken]

(05-28-2017 09:09 PM)rprosperi Wrote:  Are state files, etc. compatible so I can just replace the .exe files (Windows).

State files are backward-compatible, so you can always replace an executable with a newer version. The Binary and Decimal versions can also read each other's state files.

When you downgrade to an older version, though, reading the state file may fail. Also, state files are not interchangeable between platforms, so you probably won't be able to use one from Windows on Android, or DM42, etc.

(05-28-2017 09:09 PM)rprosperi Wrote:  Also, FYI, the README.txt file in the .zip still says you can keep the state and data files in the install folder by creating a file named 'portable' in the install folder, but it appears this option is deprecated. Not a big deal, just letting you know.

That should still work, I haven't removed that feature.

[rprosperi]

(05-28-2017 09:47 PM)Thomas Okken Wrote:  That should still work, I haven't removed that feature.

I started by deleting C:\Users\Prosperi\AppData\Roaming\Free42 (from much older versions of Free42) then I created an empty file 'portable.txt' in the install folder, renamed it to 'portable.' and ran C:\Program Files (x86)\Free42\Free42Decimal.exe (using a shortcut from the desktop) and it created a new instance of C:\Users\Prosperi\AppData\Roaming\Free42 with the various state/config files.

I also saw a comment somewhere (in project history?) saying that this feature was no longer supported after v1.5.x (don't recall exact version). So maybe it was there, went away for a while, then came back?

Would prior installations leave traces in the registry that might override the 'portable' setup?

Anyhow, not a big deal, it's just easier to clean up when removing if it's all in one place.

[Thomas Okken]

(05-28-2017 10:04 PM)rprosperi Wrote:  

(05-28-2017 09:47 PM)Thomas Okken Wrote:  That should still work, I haven't removed that feature.

I started by deleting C:\Users\Prosperi\AppData\Roaming\Free42 (from much older versions of Free42) then I created an empty file 'portable.txt' in the install folder, renamed it to 'portable.' and ran C:\Program Files (x86)\Free42\Free42Decimal.exe (using a shortcut from the desktop) and it created a new instance of C:\Users\Prosperi\AppData\Roaming\Free42 with the various state/config files.

The file must be named "portable", not "portable.". No dots allowed.
I just tried it in Windows 10; still works fine.

(05-28-2017 10:04 PM)rprosperi Wrote:  I also saw a comment somewhere (in project history?) saying that this feature was no longer supported after v1.5.x (don't recall exact version). So maybe it was there, went away for a while, then came back?

It was introduced in 1.5.1, and never removed.

(05-28-2017 10:04 PM)rprosperi Wrote:  Would prior installations leave traces in the registry that might override the 'portable' setup?

No; recent versions of Free42 don't use the registry at all.

[rprosperi]

(05-28-2017 10:24 PM)Thomas Okken Wrote:  The file must be named "portable", not "portable.". No dots allowed.
I just tried it in Windows 10; still works fine.

Ok, subtle but dumb error on my part.

(05-28-2017 10:24 PM)Thomas Okken Wrote:  No; recent versions of Free42 don't use the registry at all.

I found some Free42 file associations (with file type .dat) in the registry, but those could be remnants of much earlier versions of Free42. I presume these can be safely removed?

Thanks Thomas

[Thomas Okken]

(05-29-2017 12:10 AM)rprosperi Wrote:  

(05-28-2017 10:24 PM)Thomas Okken Wrote:  No; recent versions of Free42 don't use the registry at all.

I found some Free42 file associations (with file type .dat) in the registry, but those could be remnants of much earlier versions of Free42. I presume these can be safely removed?

Thanks Thomas

I don't know how those would have been created. The only thing Free42 ever stored in the registry was the location of the "Free42 Directory," i.e. the directory to use for skins and the state file.

So yes, those entries can be safely removed.:-)

[rprosperi]

(05-29-2017 01:30 AM)Thomas Okken Wrote:  I don't know how those would have been created. The only thing Free42 ever stored in the registry was the location of the "Free42 Directory," i.e. the directory to use for skins and the state file.

So yes, those entries can be safely removed.:-)

Removed everything, did a new install of 2.0.3 with the proper 'portable' file and all went as expected.

Thanks again.

[matthiaspaul]

(05-26-2017 12:18 AM)matthiaspaul Wrote:  Since the calculator's keyboard area is slightly recessed, it appears to be quite well suited for keyboard overlays in general. Does the case also feature small notches in the rails to the left and right of the keyboard to better afix overlays (like on the HP 48)? If not, that might be a useful tweak to further improve usability.

Noone?

What about the power supply? What kind of batteries are needed to power the calculator? Does the calculator also support being powered via USB (I assume yes, but I haven't seen this being mentioned anywhere so far)?

The calculator appears to have a size similar to the HP-42S. Does someone know the exact case dimensions already?

Greetings,

Matthias