You've Got Mail.........DM42L

[d b]

Folks who have purchased something from Swiss Micros are receiving notice about the coming DM42L.
There's a link to a Youtube vid: https://www.youtube.com/watch?v=314JMYJE...e=youtu.be
Great news.

[Massimo Gnerucci]

(02-19-2016 09:38 PM)Den Belillo (Martinez Ca.) Wrote:  Folks who have purchased something from Swiss Micros are receiving notice about the coming DM42L.
There's a link to a Youtube vid: https://www.youtube.com/watch?v=314JMYJE...e=youtu.be
Great news.

See also http://www.hpmuseum.org/forum/thread-5716.html

[JimP]

(02-19-2016 09:38 PM)Den Belillo (Martinez Ca.) Wrote:  Folks who have purchased something from Swiss Micros are receiving notice about the coming DM42L.
There's a link to a Youtube vid: https://www.youtube.com/watch?v=314JMYJE...e=youtu.be
Great news.

OK I was toying with getting a DM-41L but I think this is going to be the one I will get. Is there any indication on how much memory this new beauty will have, or will it be precisely the same as the HP42S?

[Massimo Gnerucci]

(02-19-2016 09:50 PM)JimP Wrote:  

(02-19-2016 09:38 PM)Den Belillo (Martinez Ca.) Wrote:  Folks who have purchased something from Swiss Micros are receiving notice about the coming DM42L.
There's a link to a Youtube vid: https://www.youtube.com/watch?v=314JMYJE...e=youtu.be
Great news.

OK I was toying with getting a DM-41L but I think this is going to be the one I will get. Is there any indication on how much memory this new beauty will have, or will it be precisely the same as the HP42S?

This calculator will be available at https://www.SwissMicros.com in summer 2016. Hold your breath and stay tuned.

[d b]

(02-19-2016 09:50 PM)JimP Wrote:  Is there any indication on how much memory this new beauty will have, or will it be precisely the same as the HP42S?

The email is copped on the Youtube page, or vice versa. I know nothing more. It looks like Michael is still fine tuning this so my guess is that there will be more info forthcoming. As he says: "Hold your breath and stay tuned."

[Paul Dale]

The calculator forensics example in the video differs from the 34S's result in the last digit. The question is which one is rounding incorrectly and where?

Code:

sin  9                                          0.1564344650402308690101053194671668 9231389989208566
cos  0.1564344650402308690101053194671669       0.9999962727428850241175162050113502 4868632901799155
tan  0.9999962727428850241175162050113502       0.01745499985548866079139414092834846 834926109460325
atan 0.01745499985548866079139414092834847      0.9999962727428850241175162050113502 9455156466832379
acos 0.9999962727428850241175162050113503       0.1564344650402308690101053194660900 7173080649072903
asin 0.1564344650402308690101053194660901       8.999999999999999999999999999937535 3009012793999439
9 -  8.999999999999999999999999999937535        -6.2465E-29

This matches the 34S's result

The video of the DL42L shows -6.2466E-29 so it is off by at best 1 ULP somewhere.

And yes, this should have been done in radians not degrees...


- Pauli

[Steve Simpkin]

I wish he had shown one of the HP-42's menu in operation.

[rprosperi]

(02-19-2016 10:40 PM)Steve Simpkin Wrote:  I wish he had shown one of the HP-42's menu in operation.

Agree, this is a big issue for the 42 display; hopefully Michael will share some of these screens soon. The F6 key seen on the lower left of the DM42L keyboard presumably represents the top-right 42S key ("XEQ" on the original) for using soft menus. This is awkward to say the least, but all layouts for this landscape format will have some level of compromise like this.

[Dave Britten]

Yeah, layout looks a little odd yet, though I could probably adapt if it ended up being released that way. Personally, I'd probably just keep the 4x4 numeric section on the right identical to the 42S, take the top 6x2 portion of the 42S and put it in the upper left corner, work the rest of the keys in to the left of Enter, and put some printed/painted lines connecting the upper left six keys to the menu spaces along the LCD.

Or just release it in a portrait shell, but any port in a storm at this point.

[Didier Lachieze]

This is a good layout candidate for a landscape 42



Just need to work the alignment of the top keys with the screen on the DM-42L.

[Ángel Martin]

(02-19-2016 10:36 PM)Paul Dale Wrote:  The calculator forensics example in the video differs from the 34S's result in the last digit. The question is which one is rounding incorrectly and where?

I agree it's probably never enough number of decimals, but can't help but thinking how many decimals were used to put a man on the moon? Sobering thought...

[Paul Dale]

(02-20-2016 08:48 AM)Ángel Martin Wrote:  I agree it's probably never enough number of decimals, but can't help but thinking how many decimals were used to put a man on the moon? Sobering thought...

It isn't about the number of decimal digits, it is more about how it rounds those digits.
The most precise ever claim is obviously incorrect when the 34S rounds correctly and Intel's library doesn't -- mostly because the 34S library carries more digits for the basic functions.

To get to the moon, you don't need a large number of digits -- so long as you account for the inaccuracy in the calculations as you go.

Pauli

[Werner]

(02-19-2016 10:36 PM)Paul Dale Wrote:  The calculator forensics example in the video differs from the 34S's result in the last digit. The question is which one is rounding incorrectly and where?
- Pauli

Free42.
SIN(9°): one ULP off on Free42
TAN(0.9999962727428850241175162050113502): idem
I didn't look further.

Is is because the WP34S performs the degrees-to-radians conversion with 38 intermediate digits?

Werner

[rprosperi]

(02-20-2016 06:51 AM)Didier Lachieze Wrote:  This is a good layout candidate for a landscape 42



Just need to work the alignment of the top keys with the screen on the DM-42L.

I immediately liked this layout best out of those posted since yesterday's news of the DM42L. And now I know why... this is the landscape layout I have installed on my Android phone for Free42. On the actual DM42L, the 6 top-left keys obviously cannot align under the F1-F6 soft-keys in the display due to the location of the window frame, but I think using these 6 keys is more intuitive and usable than aligning some and putting others in different places.

Does anyone know who crafted this layout?

[Paul Dale]

(02-20-2016 12:53 PM)Werner Wrote:  Is is because the WP34S performs the degrees-to-radians conversion with 38 intermediate digits?

I don't know if it is the degrees to radians conversion or not. This uses 39 digits. The series expansion for the trig functions are calculated with additional digits too.


- Pauli

[Dave Britten]

Another random idea: Is there a version of the LCD they're planning to use that also has a resistive touch digitizer on it? Wouldn't need to be super high-res or anything, only enough to differentiate a touch in one of the 6 vertical divisions of the screen, and it could have other potential interesting uses.

EDIT: It would be something like this old Franklin Bookman (man, those were cool) that had a simple touch screen for the soft keys along the bottom of the screen.

http://www.amazon.com/Franklin-MWD-1450-...B00006HMDA

[rpn fan]

As awesome as Swiss Micros is, I would rather find a good original if we are talking about direct clones.

Now, if we were talking about a premium body for a WP-xxS, I would pay premiums for that. Because they would be moving the ball forward.

[Werner]

(02-20-2016 11:18 PM)Paul Dale Wrote:  

(02-20-2016 12:53 PM)Werner Wrote:  Is is because the WP34S performs the degrees-to-radians conversion with 38 intermediate digits?

I don't know if it is the degrees to radians conversion or not. This uses 39 digits. The series expansion for the trig functions are calculated with additional digits too.
- Pauli

The Intel library also uses extended precision for the trig functions, I don't think you'll see differences when using radians.

Quote:The most precise ever claim is obviously incorrect when the 34S rounds correctly and Intel's library doesn't

? Free42, at the very least, rounds to even (upon my request, even). Can you give an example of Free42 rounding incorrectly?

Cheers, Werner

[Paul Dale]

(02-23-2016 07:02 AM)Werner Wrote:  ? Free42, at the very least, rounds to even (upon my request, even). Can you give an example of Free42 rounding incorrectly?

We've got an example of Free42 using the Intel decimal library rounding incorrectly in this thread

I don't know of any examples where Free42, using its usual decimal library, making a rounding error -- I'd be surprised if it didn't for some function or matrix operation.


- Pauli

[Werner]

Ah, of course.. ;-)
But then, so does the 34S?
Take this example (in DEG mode)
SIN(180-a) = COS(270+a)
Take a=1e-4
WolframAlpha produces (for both, of course)
1.74532 92519 93443 48076 79896 05432 78633..e-6

Free42 produces (for both)
1.74532 92519 93443 48076 79896 05432 787e-6

the 34S produces
1.74532 92519 93443 48076 79896 05432 791e-6 for SIN
1.74532 92519 93443 48076 79896 05432 778e-6 for COS

BTW Free42 will honour the equality, whatever the value of a you might throw at it.

Cheers, Werner