First Impression of HP 35s



#2

Hello all,

I've been visiting HP Museum for about a year now. I enjoy this site because I have been using HP calculators for about 22 years. For the past week, I have been thinking a lot about HP calculators - mainly because of the 35s. Here is my 1st impression of the new calculator from HP.

When I found out about the new HP 35s, I knew I had to get one, even though my HP 32sII works just fine. My first HP calculator was the HP 15C, built like a tank and very practical in use. It was the best engineering calculator in the 80's. However, I lost that calculator many years ago.

Although I missed the HP party in San Diego, CA, I purchased the 35s as a sort of "thank you HP" for several reliable, high quality precision tools that they had created through the years. I just received the 35s this week. The key layout reminds me of the older C series calculators and is a positive improvement over the cluttered 32sII key layout. If there is one change I could make on the keyboard layout, it would be to have a separate STO key. To my eyes, the annunciators seem a bit too small on the display. Also, the 35s LCD screen has more reflections than the screen on my 32sII; with indoor lighting, it is more difficult to see the numbers on the 35s.

I'm not sure how to benchmark calculators; however, during some calculations, such as nCr (n=800, r=400), my 32sII found the result faster than the 35s. Other times, such as integration of the bessel function with x=2, the 35s was slightly faster. Again, these were subjective tests, nothing scientific about them.

I doubt I will use the 35s to its full capability (i.e. programmable features), unless I catch the RPN programming bug. Despite my display preferences, I hope the 35s sells well and will encourage HP to design future calculators with the same user-centered design approach and reliability that they have been famous for.

The 35s is a well thought out machine. I like the case that is supplied with it. The manual appears thorough (similar to the 32sII manual) and contains some errors in grammar and details. I wished HP had provided an electronic copy of the manual on the "bonus" CD video. Perhaps it will be available on the HP website.

The last time I was excited about a calculator was when I purchased a 15C in 1985. I'm glad I purchased this new 35s and am looking forward to future scientific/engineering RPN calculators from HP.

Sincerely,
David Bailey,
graduate student,
Industrial and Systems Engineering
Virginia Polytechnic Institute and State University,
Blacksburg, VA

Edited: 7 Oct 2007, 1:56 p.m. after one or more responses were posted


#3

Welcome David, Hopefully you will be with us for a long time.

The cover of the 35s is the best one ever. However, you are right about the STO key. Also, the Polar/Rectangular conversions we have to perform ourselves. Overall, I am impressed.

All of the attendents received a 35s (Thank you HP!), and since I have one, one of my goals is to make someone else the 2nd person at the Cal Poly Pomona campus to have one. (I'm sure I am the first).


#4

Hi, Eddie,

Quote:
The cover of the 35s is the best one ever.
What do you mean by "cover" here?

#5

The hard cover that is on the calculator case.

#6

Hi Eddie,

I'm sure whoever will get your 35s will greatly appreciate it. On the 35s, HP placed the Polar <==> Rectangular conversion functions within the DISPLAY menu. On the 32sII, the conversion functions are placed over the numerical 4 key. I find the Polar/Rect conversions on the 35s are just as easy as on the 32sII. On the 35s, if I want to convert polar coordinates (25 degree, 15) to rectangular (x,y), press the following keys:

15


right-shift (blue) key


Theta


25


left-shift (yellow) key


DISPLAY


xiy (selection 9)


ENTER may be required depending on how you select in the menu

The (x,y) coordinate is (13.5946, i6.3393) in degree MODE. The conversion on the 32sII is similar except there is no menu to use. I like the way complex numbers are shown on the 35s. I have to use the X exchange Y key on the 32sII to see the imaginary part.




Clarification: By reflections of the LCD screen on the 35s, I mean the glossy look of the screen. Placed at certain angles, the 35s screen is overwhelmed by the light source, hence, obscuring my view of the display information. The LCD screen of the 32sII is not glossy and does not reflect as much light into your eyes; yet, the information is just as viewable as on the 35s.




Cheers,

David Bailey

Edited: 7 Oct 2007, 1:58 p.m.

#7

I am curious. In addition to the 35s, what other hand-held RPN scientific calculator does not have a separate STO (or RCL) key?




David Bailey

Edited: 7 Oct 2007, 2:14 p.m.


#8

The third generation 28c, 28s, 48/49/50 series.

Also a number of algebraic: 8s, 10s,10b, 10bii,


#9


  • Q: I am curious. In addition to the 35s, what other hand-held RPN scientific calculator does not have a separate STO (or RCL) key?

  • A: The third generation 28c, 28s, 48/49/50 series.

    Ah, but RCL is shifted in the RPL-based models, while STO is shifted on the HP-35s. Why is that?

    Answer: RCL is often unnecessary in RPL, as ENTER or other operation will recall a named variable.

    In RPN, STO need be used only once for storing a given value to a register, while RCL might be used many times.

    -- KS

  • #10

    Hi Allen and KS!

    Thanks for your response. I am curious, though, about non-RPL, non-graphing HP calculators. Is the 35s perhaps the first RPN without a dedicated STO (or RCL) key?




    Cheers!

    David Bailey

    Quote:
    The third generation 28c, 28s, 48/49/50 series.

    Also a number of algebraic: 8s, 10s,10b, 10bii,



    Edited: 7 Oct 2007, 4:14 p.m.


    #11

    Hi David,

    Quote:
    Is the 35s perhaps the first RPN without a dedicated STO (or RCL) key?

    As far as I see it is the first RPN scientific with less than 2 keys for STO/RCL. It is also the first RPN scientific with a shifted STO. For financial calcs, other members are more competent.

    Enjoy, Walter


    #12

    Quote:
    Hi David,

    As far as I see it is the first RPN scientific with less than 2 keys for STO/RCL. It is also the first RPN scientific with a shifted STO. For financial calcs, other members are more competent.

    Enjoy, Walter


    Mine has a "shifted" STO key.

    Given that there is no shift key, you simply press the button a 2nd time to access STO. Works very well indeed, and I think it's actually more intuitive than a shift key anyway.

    Dave.


    #13

    Nice watch calculator prototype! Will this make it to market?



    David Bailey


    #14

    Quote:
    Nice watch calculator prototype! Will this make it to market?

    In some way, shape or form, yes.
    Can you solder?

    Dave.


    #15

    LOL. No. I gave up soldering in 1990.


    David Bailey


    Quote:


    In some way, shape or form, yes.
    Can you solder?

    Dave.


    #16

    I can solder, and I think your watch looks awesome. Any details?


    #17

    Quote:
    I can solder, and I think your watch looks awesome. Any details?

    I don't want to give away major details just yet, but I'm fairly certain I'll make a kit available, perhaps partially assembled, and maybe even fully assembled (but that would cost a lot more). Also, full design details will eventually be released.
    So the watch you see there will be available for those who are keen enough to own one.
    More news in the coming months.

    Dave.


    #18

    For all the geeks who can't solder, but desperately need a watch with a calculator -um- slide rule ...


    #19

    And the batteries - for the whiz-wheel - never go flat.

    I wear one of these, but the trouble is that after a certain age, the slide rule is too small to read. . . . :-(

    Best,

    --- Les

    [http://www.lesbell.com.au]

    #20

    Welcome, David --

    I'll echo Eddie's greetings.

    You've made a few accurate and widely-accepted observations about the HP-35s. I share most of your sentiments.

    The shifted STO key is indeed a small annoyance -- a compromise driven by the limited supply of 43 keys, as four were dedicated to moving the cursor, one to "i", one to "EQN", and one to "( )".

    The speed issue has been discussed -- the HP-35s is indeed a bit slower than its predecessors HP-33s and HP-32SII. The HP-35s uses the same CR2032 3V "wafer" cells as the HP-33s, which experienced abysmal cell life. It's possible that some changes were made to remedy the problem.

    The faster numerical integration you observed is probably due to the problem chosen and the accuracy you selected. The HP-35s methods are the same as those of the HP-33s, which differ slightly from those of the HP-32SII, even though the syntax is exactly the same. Here's more than you may have wanted to know about numerical integration in HP's RPN calculators:

    http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/articles.cgi?read=556

    Best regards,

    -- KS


    #21

    KS,


    Thanks for the information. I wonder what the clock speed is of the 8502 processor in the 35s. The 33s uses a Sunplus SPLB31A running at 5Mhz, according to http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv014.cgi?read=63140




    As others have discussed on this forum, the HP 35s overall keyboard layout design approach seems to focus more on programmability functions than on direct input/output calculation. Maybe (just maybe) the 35s will encourage me to start programming it. I only recall programming the 15C many years ago, but not on any other HP calculator.




    The 35s is my first calculator that uses the 3-volt lithium coin batteries. Perhaps an OLED display will someday enhance battery life (and display clarity) for HP calculators?




    David Bailey

    Edited: 7 Oct 2007, 3:36 p.m.


    #22

    Quote:
    KS,


    Thanks for the information. I wonder what the clock speed is of the 8502 processor in the 35s. The 33s uses a Sunplus SPLB31A running at 5Mhz, according to http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv014.cgi?read=63140




    As others have discussed on this forum, the HP 35s overall keyboard layout design approach seems to focus more on programmability functions than on direct input/output calculation. Maybe (just maybe) the 35s will encourage me to start programming it. I only recall programming the 15C many years ago, but not on any other HP calculator.




    The 35s is my first calculator that uses the 3-volt lithium coin batteries. Perhaps an OLED display will someday enhance battery life (and display clarity) for HP calculators?

    I can't see how OLED will ever match LCD for low power consumption.

    Dave.

    #23

    I bought HP35S a month ago, more or less for its nice retro look and improved memory usage (compared to HP33S which I used until now). From my point of view (civil engineering, bridge design practice), the pros and cons of this calc are as follows:.

    Keyboard
    Nice retro shape, good key response.
    Bad positioning of some frequently used commands to shifted keys and vice versa:
    - STO, x^2 used very often but placed on shifted keys
    - GTO, "i" seldom used, these could be on a shifted position (well, "i" is possibly more used in electrical calculations? - then the position is fine)
    - R/S quite frequently used, much more than EQN; R/S should be closer to the numeric keys block, possibly at EQN position?

    Display
    Quality is good, appears the same as on my HP33S (#CNA 51500563) which was described as very bad by some people here (probably older models than mine 33s).
    Totally wrong behaviour when showing "ALL" with Exx part of the number out of display - HP33S works slightly better here.

    Memory usage is much better, only now the 30 kB RAM can be really used (much better than HP33s). However, while the access to the variables 1-~800 by indirect addressing is fine when programming, it is very slow when used in hand calculations. I would like additional set of RCL() and STO() keys which would ask for the adress rather than for the label. This way, the sequence "123 STO I RCL (I)" could be replaced by much shorter "RCL() 123" (or, maybe better "123 RCL()").

    Also, the nice possibility of addressing label AND line number slows down normal hand calculations - one HAS to run program A001 with XEQ A and ENTER which is annoying. On other calculators (e.g. the old TI59, the superb old SHARP PC-1211 very well described in http://membres.lycos.fr/albillo/calc/pdf/DatafileVA027.pdf, CASIO FX-602P etc.) one could define a function and run it just by pressing a function key (one keypress); on HP33s, it's XEQ A (two keys), which is still fine; on HP35S, three keys are too many. This may seem to be a negligible issue, but when running a (trivial) program many times, it's not pleasant at all.

    Minor flaw: Pressing just E3* results in "Syntax error" while on 33S it was OK. Now one has to (remember to) enter 1E3*.

    Overall, HP35S is really a nice piece of hardware - a "must have" for many users. However, as most complex and programmed calculations are done on a PC now, the calc should be designed with more priority for quick hand calculations.

    VQ, Prague, Czech Republic


    #24

    VQ,


    Thanks for your very thorough observation of the 35s. I, too, like the retro look of the keyboard - reminds me of my beloved 15C. I don't program calculators, but I see what you mean concerning the extra keystroke, ENTER, required on the 35s compared to your 33s.




    Concerning the powers of ten E key, I guess HP had decided to standardize its use even if the mantissa is 1. That's ok as long as the E key works the same on near future RPN calculators and is well documented.




    I understand what you mean concerning the SHOW key. Only HP can answer why the display does not also show the exponent when you perform the following:


    (1) enter an integer larger than 999,999,999,999 or smaller than -999,999,999,999;


    (2) press left-shift key;


    3) press SHOW.




    Cheers,


    David Bailey

    Quote:
    I bought HP35S a month ago, more or less for its nice retro look and improved memory usage (compared to HP33S which I used until now). From my point of view (civil engineering, bridge design practice), the pros and cons of this calc are as follows:.

    Keyboard
    Nice retro shape, good key response.
    Bad positioning of some frequently used commands to shifted keys and vice versa:
    - STO, x^2 used very often but placed on shifted keys
    - GTO, "i" seldom used, these could be on a shifted position (well, "i" is possibly more used in electrical calculations? - then the position is fine)
    - R/S quite frequently used, much more than EQN; R/S should be closer to the numeric keys block, possibly at EQN position?

    Display
    Quality is good, appears the same as on my HP33S (#CNA 51500563) which was described as very bad by some people here (probably older models than mine 33s).
    Totally wrong behaviour when showing "ALL" with Exx part of the number out of display - HP33S works slightly better here.

    Memory usage is much better, only now the 30 kB RAM can be really used (much better than HP33s). However, while the access to the variables 1-~800 by indirect addressing is fine when programming, it is very slow when used in hand calculations. I would like additional set of RCL() and STO() keys which would ask for the adress rather than for the label. This way, the sequence "123 STO I RCL (I)" could be replaced by much shorter "RCL() 123" (or, maybe better "123 RCL()").

    Also, the nice possibility of addressing label AND line number slows down normal hand calculations - one HAS to run program A001 with XEQ A and ENTER which is annoying. On other calculators (e.g. the old TI59, the superb old SHARP PC-1211 very well described in http://membres.lycos.fr/albillo/calc/pdf/DatafileVA027.pdf, CASIO FX-602P etc.) one could define a function and run it just by pressing a function key (one keypress); on HP33s, it's XEQ A (two keys), which is still fine; on HP35S, three keys are too many. This may seem to be a negligible issue, but when running a (trivial) program many times, it's not pleasant at all.

    Minor flaw: Pressing just E3* results in "Syntax error" while on 33S it was OK. Now one has to (remember to) enter 1E3*.

    Overall, HP35S is really a nice piece of hardware - a "must have" for many users. However, as most complex and programmed calculations are done on a PC now, the calc should be designed with more priority for quick hand calculations.

    VQ, Prague, Czech Republic



    Edited: 7 Oct 2007, 4:25 p.m.


    #25

    Quote:
    I understand what you mean concerning the SHOW key. Only HP can answer why the display does not also show the exponent when you perform the following:


    (1) enter an integer larger than 999,999,999,999 or smaller than -999,999,999,999;


    (2) press left-shift key;


    3) press SHOW.



    David,
    what's even worse, in mode Display/All, pressing 4 ENTER 9 / shows only 4.44444444444E- (and a little anunciator - right hand arrow), hiding the (most?) important part of the result, which makes the "Display/ALL" mode practically unusable. It was discussed here in an earlier thread and is a bitter innovation over HP33S. Fixed notation is therefore much better for normal work. I wonder if the designers ever tried to do some calculations with ALL mode set?
    Regards, VQ


    #26

    You are correct, VQ. It is a bit annoying. With both the 32sII and 35s in display ALL mode, the irrational number pi is displayed with 12 digits:


    3.14159265359




    When I perform a division such as 1 divided by 3, the 32sII shows the repeating fraction with a 9 digit mantissa, followed by the letter E, a negative sign, and exponent value 1:


    3.33333333E -1




    However, with the 35s, this same fraction is shown with a 12 digit mantissa, the letter E, and negative sign:


    3.33333333333E - (rest is cut off)




    Blunder or intentional by HP?



    David Bailey

    Quote:

    David,
    what's even worse, in mode Display/All, pressing 4 ENTER 9 / shows only 4.44444444444E- (and a little anunciator - right hand arrow), hiding the (most?) important part of the result, which makes the "Display/ALL" mode practically unusable. It was discussed here in an earlier thread and is a bitter innovation over HP33S. Fixed notation is therefore much better for normal work. I wonder if the designers ever tried to do some calculations with ALL mode set?
    Regards, VQ



    Edited: 7 Oct 2007, 7:44 p.m.

    #27

    David,

    Quote:
    Concerning the powers of ten E key, I guess HP had decided to standardize its use even if the mantissa is 1. That's ok as long as the E key works the same on near future RPN calculators and is well documented.

    AFAIK each and every RPN scientific interpreted "E 17" or "EEX 17" as 1E17 so far, except the HP35s. So if you vote for standardization (and ease of use), the HP35s shall be changed. And I hope sincerely this feature will be brought back to normal with further calcs.
    #28

    Dobrou noc, VQ,

    Quote:
    I would like additional set of RCL() and STO() keys which would ask for the adress rather than for the label. This way, the sequence "123 STO I RCL (I)" could be replaced by much shorter "RCL() 123" (or, maybe better "123 RCL()").

    Thanks for posting this very nice idea! Maybe also Cyrille of HP is reading your post.

    With respect to alternative keyboard layouts for the 35s, some proposals were published here already (e.g. this one) and at HHC2007.

    Naskledanou!

    Walter


    Possibly Related Threads...
    Thread Author Replies Views Last Post
      My HP 12C+ impression Walter Lam 6 273 06-02-2011, 11:52 AM
    Last Post: Walter Lam
      First Impression of the New Caso Fx-CG10 Namir 27 774 12-15-2010, 12:14 PM
    Last Post: Don Shepherd
      HP 33 First impression Kamyar 2 144 04-27-2004, 08:51 AM
    Last Post: Fubar Vikinghelmet
      PB D'impression ouakti 1 107 04-05-2004, 01:53 PM
    Last Post: Arnaud Amiel

    Forum Jump: