the experience of programming: old vs. new



#16

Good day, all.

I hope this doesn't come out sounding heretical. ;) I have spent some time the past few weeks slavering over 'Net sales / auctions of classic HPs like the 41C/CV/CX ... but if I am honest with myself, I must admit I really don't have the handy cash to become a collector - if I did have it, I would jump.

So with that in mind I have also given the 35s a look, and it appears (and according to some reviews, does have) a good deal of resemblance to an 'old classic.' I find myself wondering, outside of the obvious differences in build quality, number of available registers, etc., will I have a comparable experience programming something like the new 35s as I would if I broke the bank and got an old model from the 70s?

I'm not questioning whether money spent on a classic calc would be "worth it," I just don't think it would be wise for me. But I would like to have roughly the same experience I would have had back in the 1970s if I'd been a) more aware and b) more true to myself.

Thanks for any impressions/opinions.

Regards,

Glenn


#17

Glenn, if what you are interested in is pure RPN programming, get a 33s. It is significantly faster than the 35s, its functions work better, and it does not have any "bugs" that I am aware of.

If you don't need scientific functions, get a 12c, especially if you can find one of the newer "60 times faster than the original 12c" ones.


#18

Interesting. I hadn't heard the 33s was faster ... of course it is tough to get past all the complaints about the keyboard design! :D


#19

Keyboard design schmeeboard design. It works. All the keys you need are there.

#20

Glenn:

I believe I understand your desire to have a machine with the older/'classic'/simple feel.

Firstly, I must precede the following by saying that I am not familiar with, nor have I tried the 33s.

I have used the 35s quite a bit and do find it does truly have the 'feel' - both in programming and esthetics of the old/classic HP Programmables (I have used a HP65 since 1974 and have programmed the 25, 67 41 through the 48 series).

1. With the 35s, the ENTER^ key is properly sized and positioned. Personally, I find using any RPN/RPL machines (like the 49/50 series) with the 'newer' ENTER^ key size/position to be quite disturbing.

2. The programming is very 'comfortable' compared to the older line-oriented RPN machine.

3. The keyboard feel is quite close to the classic feel (I've recently been let down by the feel of the 20b - mushy and scratchy).

You may find the code listings on my webpage of interest:
http://tchrapkiewicz.googlepages.com/hp35s
I oonfigured the machine with 100 data registers, and some of the code on my page demonstrates some functionality with that register set.

Good luck - for the price, I believe it's worth trying the 35s.

Regards,
TomC


#21

Thank you, Tom - your post pushed me over the edge. ;^D Just ordered a 35s. GB

#22

Why not start with a few emulators? There are 41 and 42 emulators for just about every platform including PDAs/iPhones.


#23

I have one of the 41 emulators on a Windows install somewhere. I don't know, I am very fond of the actual physical artifact as opposed to an emulation. That sounds foofier than I mean it. :)

Of course a good emulator _would_ give me and idea of what the process of programming was like on, say, a 41C.


#24

That was the point. Before you spend a lot of $, give it a test drive first.

For programming I find that the 41 and 42 have the most to offer. There is no publicly available 35s emulator. Cost wise it is a low risk.


#25

Too late now perhaps but one machine I rate very highly that is far better than most people give it credit is the 32sII. Not only does it have the superb Pioneer series ergonomics, it has one feature that is closer to the older machines than the 35s - a lack of memory!

But it still has enough memory to write serious programs and I'd rather have a machine which feels designed for the size of memory it has instead of a new machine that has huge memory but due to its programming model, it is almost impossible to take advantage of the large memory size.

The 32sII should definitely have its own feature page in the museum and I hope one day, will be regarded as a classic. It does nearly everything the 35s does (the 35s is actually like the 32 with a few extra functions), has hugely better ergonomics and because it is such a relatively new machine, prices are not as mad as the older machines.

My 35s sits on a shelf unused. My 32sII remains on my desk. It is a fun machine.

Mark


#26

Both my 35s and 32SII sit unused. I obtained a 32SII because of the positive hype, and now I know where the 35s got its brain dead complex number support.

For pure RPN and portability its the 15C and 42S for me.

Edited: 5 June 2009, 9:48 a.m.


#27

I must admit to being very curious as to why complex number support is so important to so many people. I'll start a new thread.

I suppose with the 32sII you have to say that complex number support is a bonus rather than something that seems to have been an essential design consideration from the start (ie, 15C).

Mark

#28

Quote:
Both my 35s and 32SII sit unused. I obtained a 32SII because of the positive hype, and now I know where the 35s got its brain dead complex number support.

For pure RPN and portability its the 15C and 42S for me.


The 32sii actually has a few features that are superior to the 35s: (1) Polar to Rectangular conversions (2) faster integrator and solver (3) better interface and footprint (3) better overall build quality with no missed keystrokes. -- I am sure there are others.

As for the complex number support; it may be lacking compared to the 15C and 42S, but it serves most purposes. It is a great little machine nonetheless and arguably easier to keystroke program (or at least edit programs) that the 15C. The only downside for me is its limited memory!

JeffK


#29

One benefit the 35 has with P->R conversion is that you can convert immediately on input.

However, I've just realised I don't know how to split the result. The 32 returns two results but the 35 is one so if I've just done a conversion and got 3i4, how do I isolate them?

Another area where the 32 totally blows away the 35 is base conversion. I need this a lot. I detest the way the 35 works with explicitly needing to specify the base each time. Absolutley daft design. I honestly cannot find a single advantage for the 35's method over the tried and trusted methods in all previous calcs.

The 35 does have logic operations though which is a big plus.

Mark


#30

Quote:
However, I've just realised I don't know how to split the result.
Have you been successful composing them in polar notation in any other way than just typing them in? Say, you have r in register x and phi in register y. How would you compose them? This is on of the weakest points of the 35s.
#31

Quote:
However, I've just realised I don't know how to split the result. The 32 returns two results but the 35 is one so if I've just done a conversion and got 3i4, how do I isolate them?

The deficient abilities of the rectangular to polar, polar to rectangular and related functions on the HP 35s were discussed at length in the summer of 2007 upon its release. A search of Archive 17 will yield lots of information, for example see this thread.

The following program will convert a complex number to its rectangular components.

Z001	LBL Z		
Z002 SF 10
Z003 eqn CPLX->RECT
Z004 PSE
Z005 CF 10
Z006 ABS
Z007 CLx
Z008 eqn ABS(LASTx)*SIN(ARG(LASTx))
Z009 eqn ABS(LASTx)*COS(ARG(LASTx))
Z010 RTN


...

#32

Quote:
The 32sii actually has a few features that are superior to the 35s: (1) Polar to Rectangular conversions (2) faster integrator and solver (3) better interface and footprint (3) better overall build quality with no missed keystrokes. -- I am sure there are others.
The display of the 35s is quite dissapointing. A less reflective cover and some space between annunciators and border would have been geat.

It's about time for a 35sII :^).

#33

Quote:
The 32sii actually has a few features that are superior to the 35s: (1) Polar to Rectangular conversions (2) faster integrator and solver (3) better interface and footprint (3) better overall build quality with no missed keystrokes. -- I am sure there are others.

All true and that it why I wanted one, but I expected it to be a 15C++, not an 11C++. My bad. So I'll just be happy with my 15C/42S and emulators. Emulators will live forever.

#34

Quote:

So I'll just be happy with my 15C/42S and emulators. Emulators will live forever.


True, they will and the development of emulators to preserve our computing heritage is a wonderful thing ... but in terms of devices like pocket calculators, don't you find them a strangely detached experience and very removed from operating the real thing?

Myself, I use emu48 for quick calculations but when I want to program something, I always grab the real thing. For some reason, I just cannot get into programming an emulation.

Mark


#35

Quote:
Quote:
So I'll just be happy with my 15C/42S and emulators. Emulators will live forever.

True, they will and the development of emulators to preserve our computing heritage is a wonderful thing ... but in terms of devices like pocket calculators, don't you find them a strangely detached experience and very removed from operating the real thing?

Myself, I use emu48 for quick calculations but when I want to program something, I always grab the real thing. For some reason, I just cannot get into programming an emulation.


It's a matter of perspective, I guess. My experience seems to be the opposite of yours: I regard my real HP-42S as a museum piece, cherished but rarely used, kept around more for old times' sake than anything else; when I need to use a calculator in everyday life, I use an emulator on whatever device is close at hand.

At work, I always have a PC (I'm a computer programmer so PC availability is an occupational hazard). At home, my iPod touch is never far away... And when I'm travelling, I always have my Palm with me. I have and use these devices anyway; installing calculator emulators on them means I have a calculator with me at all times, too, without needing to carry Yet Another Physical Device -- and without having to worry that my hard-to-replace vintage HP will break or get stolen.

I do miss the HP-42S keyboard sometimes, but my virtual calculators run on commodity hardware, and can always be ported to new hardware should the need arise... The HP-42S, on the other hand, is slowly dying out. That reality, sad as it may be, makes using the real HP-42S feel like a detached experience, and it makes using an emulator feel more like the real thing -- for better or worse. ;-)

- Thomas

#36

Quote:
True, they will and the development of emulators to preserve our computing heritage is a wonderful thing ... but in terms of devices like pocket calculators, don't you find them a strangely detached experience and very removed from operating the real thing?

Not really. But when I go back to using the real thing there is a noticeable improvement, but not enough to make me switch back.

I travel frequently, and used to carry a 48GX/50g (for my programs) and a 15C (small and quick). Now that I have an iPhone with 15C/42S/41CX and my MacBook with x49gp (50g emulator with ARM support, i.e. I can run my C programs too), I opted to reduce space and weight by leaving them behind.

Quote:
Myself, I use emu48 for quick calculations but when I want to program something, I always grab the real thing. For some reason, I just cannot get into programming an emulation.

I know what you mean, however if you get use to the keymap its not that bad. I find with RPL programming that alpha entering commands is very fast on a QUERTY keyboard.
#37

Hi, Egan --

Quote:
I obtained a 32SII because of the positive hype, and now I know where the 35s got its brain dead complex number support.

It goes back further than that, as indicated in this archived thread from 2007:

http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv017.cgi?read=119576#119576

A self-quote from the thread:


"The complex-number functionality of the HP-32S, HP-32SII, HP-33s, and now apparently the HP-35s, was essentially a porting of the HP-41 Math Pac's RPN routines."

-- KS

Edited: 6 June 2009, 4:02 p.m.


#38

So for all the hardcore complex number users, what is the general verdict on the way the 48 series deals with them? It looks relatively elegant to me but I can see that having to enter the brackets each time can be a nuisance - although it wouldn't be difficult to setup a user-Enter key to combine lvls 1 and 2 into a CN automatically. Anyway, is the 15C still preferred to the 48 series?

Mark


#39

The 15C, 42S, and 48/50 all do complex numbers a bit different in terms of input and presentation, however I have no difficult with any of them. At least its there and complete.

#40

Mark --

Quote:
So for all the hardcore complex number users, what is the general verdict on the way the 48 series deals with them? It looks relatively elegant to me but I can see that having to enter the brackets each time can be a nuisance - although it wouldn't be difficult to setup a user-Enter key to combine lvls 1 and 2 into a CN automatically.

Complex-number support for the HP-48 series and beyond is mathematically complete. Numbers can be entered and displayed in polar mode (unlike the HP-28 series). However, I also dislike representing a complex number as an ordered parenthesized pair, which could also be a point on a graph, a pair of statistical data, or an open range of numbers.

The "R->C" function on RPL-based calc's combines levels 1 and 2 into a complex number. However, it differs from "COMPLEX" with two reals on an HP-42S, in that R->C always treats the two reals as the components of a complex number in rectangular format. COMPLEX treats the two reals as the components of a complex number in the selected format.

More on the general topic from the Archives:

http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv014.cgi?read=63415#63415

-- KS


#41

Quote:
Complex-number support for the HP-48 series and beyond is mathematically complete.

I guess I should have said, relatively complete.
#42

Quote:
The "R->C" function on RPL-based calc's combines levels 1 and 2 into a complex number. However, it differs from "COMPLEX" with two reals on an HP-42S, in that R->C always treats the two reals as the components of a complex number in rectangular format. COMPLEX treats the two reals as the components of a complex number in the selected format.

Karl - there is a solution to this problem in the 48 series which enables you to enter levels 1 and 2 and have them combined into a complex number of the currently selected format. It doesn't involve any programming, just a different command and a flag.



Please try this:



1. Set flag -19.

2. Put the machine into Polar.

3. Enter for example 5 and 53.1 in levels 2 and 1.

4. Goto MTH/VECTR and press ->V2

5. You now have a complex number in the correct format.



Obviously this also works in rectangular mode. I agree it isn't as neat as being able to do it with a single shift+key in the 42s but at least the solution does exist in the 48 machines and if you keep the Vector menu open in the 48, it is still very immediate.



Flag -19 sets between creating a vector or a complex number with the ->V2 command.

Mark

Edited: 7 June 2009, 12:49 p.m.

#43

Karl,

Thanks. Now I remember.


Possibly Related Threads...
Thread Author Replies Views Last Post
  First experience with a slide rule aurelio 27 1,905 08-17-2013, 01:16 PM
Last Post: Garth Wilson
  Experience (Learning) Curve Program for HP-41C/CV/CX Chris Catotti 1 257 09-20-2011, 01:46 PM
Last Post: Frido Bohn
  Eric's overlay - my experience and suggestions Cristian Arezzini 5 427 06-20-2011, 05:37 PM
Last Post: Gerry Schultz
  WP 34S experience (cable, flashing, overlay, etc) JayV 2 294 06-14-2011, 02:44 AM
Last Post: Marcus von Cube, Germany
  Wp34s - my flashing experience Cristian Arezzini 5 422 05-24-2011, 01:18 PM
Last Post: Tim Wessman
  Mostly OT: What's your experience with LED flashlights? Paul Brogger 39 2,073 07-29-2010, 01:38 PM
Last Post: Ren
  TI-84+, a humbling experience Egan Ford 7 533 04-16-2009, 04:15 PM
Last Post: Don Shepherd
  Any experience connecting HP50G with MacOSX using HPConnect? Miguel Saiz 4 383 09-21-2007, 09:04 PM
Last Post: Miguel Saiz
  50G--the accumulated experience to date? Les Wright 30 2,200 06-13-2007, 12:16 AM
Last Post: Ron Allen (Fairhope)
  Anyone have any experience of Lygea calculators? Bruce Horrocks 2 250 03-26-2006, 07:45 AM
Last Post: Bruce Horrocks

Forum Jump: