Just received a HP35s

[Tugdual]

First impressions: the construction is actually pretty good. Not as dense as a 15C, the back makes a drum noise because the case seems pretty empty but the design is much better than a 30b, keyboard pleasant, easy to read (unlike the Prime).
Starting to use it and I find it very practical; it has a few metric to imperial conversions on keyboard, % and %CHG, bases calculations... very nice.
The screen is easy to read, pretty clear, programming is really a generation above 15C.
I keep discovering it, but so far I find this little thing simple and very convenient.
A bit surprised it didn't receive more attention, took me a while to decide to order one.

[Thomas Radtke]

(02-06-2015 09:26 AM)Tugdual Wrote:  A bit surprised it didn't receive more attention [...]

It received quite some attention in 2007/2008 until the buglist grew beyond any acceptable size.

I agree about the concept being very practical. The calculator basically adds what was missing from the 32SII and has very few glitches in terms of the user interface.

[walter b]

(02-06-2015 09:43 AM)Thomas Radtke Wrote:  [quote='Tugdual' pid='26718' dateline='1423214812']
...
I agree about the concept being very practical. The calculator basically adds what was missing from the 32SII and has very few glitches in terms of the user interface.

If they had put the cursor cross one row below and made the little beast repurposable, it would have been great. But alas ...

d:-/

[Derek W]

The 35s has been my everyday calculator on my desk since I bought it in around 2007. I prefer its 2-line display to the single line of my 15CLE, and it's less cumbersome for run of the mill calculations than my 50g.

Some users have reported heavy battery consumption, but mine is only on about its 3rd set from new. Most of the items in the bug list don't affect the sort of work I do, but the one that does annoy me is the disappearing exponent for numbers <1 in ALL mode (no 9 on the list). As an engineer working on practical projects, I rarely need more than about three siginificant figures in the mantissa, but an error by a factor of 10 could be serious.

I think the hardware quality is very good, it's just a pity that the firmware bugs couldn't be ironed out, for contractual reasons that Tim W has described in earlier posts.

[kusmi]

(02-06-2015 10:31 AM)Derek W Wrote:  but the one that does annoy me is the disappearing exponent for numbers <1 in ALL mode (no 9 on the list). As an engineer working on practical projects, I rarely need more than about three siginificant figures in the mantissa, but an error by a factor of 10 could be serious.

Yes - this bug is my biggest issue. This is the reason I abandoned my 35s after just 1 week of use.

But not all is lost: I repurposed the 35s case for my 42S ;-)

[ElectroDuende]

(02-06-2015 10:44 AM)kusmi Wrote:  But not all is lost: I repurposed the 35s case for my 42S ;-)

We have to see that!

[Tugdual]

Bugs
Yes there are bugs and bugs are risks.
A risk is evaluated according to impact, occurence & detectability.
I read the list of bugs, not sure I will see them very often and/or the impact would be huge.
I'm not launching rockets to Mars so I guess this calculator is very acceptable for my daily activities.

Some negative aspects:
- the size of casing. The 35s is bigger and fairly thicker than a 15C which is an issue IMO
- the speed: the 15C is about 10 times faster according to my tests

Still, the 35s is attractive, and I'm aligned with Derek:

(02-06-2015 10:31 AM)Derek W Wrote:  The 35s has been my everyday calculator on my desk since I bought it in around 2007. I prefer its 2-line display to the single line of my 15CLE, and it's less cumbersome for run of the mill calculations than my 50g.

[Dieter]

(02-06-2015 10:31 AM)Derek W Wrote:  The 35s has been my everyday calculator on my desk since I bought it in around 2007. I prefer its 2-line display to the single line of my 15CLE, and it's less cumbersome for run of the mill calculations than my 50g.

That's exactly my impression as well.

(02-06-2015 10:31 AM)Derek W Wrote:  Some users have reported heavy battery consumption, but mine is only on about its 3rd set from new.

Here it's the fifth set since late 2007. With everyday use, a battery set lasts for more than a year. I think that's far from "heavy consumption".

(02-06-2015 10:31 AM)Derek W Wrote:  Most of the items in the bug list don't affect the sort of work I do, but the one that does annoy me is the disappearing exponent for numbers <1 in ALL mode (no 9 on the list). As an engineer working on practical projects, I rarely need more than about three siginificant figures in the mantissa, but an error by a factor of 10 could be serious.

I think we should not blame the 35s here. In any case it displays an "E" at the right display border, so you know there is an exponent – yes, it's out of the display, and I also do not like the switch to SCI display for any value below 1 (that's why I use the WP34s in ALL 3 mode). But with the limited number of display digits there is no other choice except reducing the mantissa digits. Remember, the mode is called ALL, not MOST or SOME. ;-)

(02-06-2015 10:31 AM)Derek W Wrote:  As an engineer working on practical projects, I rarely need more than about three siginificant figures in the mantissa, but an error by a factor of 10 could be serious.

Then simply set the 35s to SCI 3 or ENG 3, or use its default setting FIX 4. That's what worked for generations of 67/97/41C/15C users – they simply had no ALL mode.

(02-06-2015 10:31 AM)Derek W Wrote:  I think the hardware quality is very good, it's just a pity that the firmware bugs couldn't be ironed out, for contractual reasons that Tim W has described in earlier posts.

It's a pity that the (real) bugs were not detected during tests before production started. I wonder what kind of tests HP did (or usually does) before a calculator is ready to be marketed. Maybe someone with insight may comment on this.

Dieter

[Dieter]

(02-06-2015 01:09 PM)Tugdual Wrote:  - the size of casing. The 35s is bigger and fairly thicker than a 15C which is an issue IMO

I think it's just the right size and feel. :-)

(02-06-2015 01:09 PM)Tugdual Wrote:  - the speed: the 15C is about 10 times faster according to my tests

Wait... the original 15C from 1982 is faster than the 35s? I'd say the 35s has about 3x–4x the speed of a 41C and 10x–20x that of a 15C.

(02-06-2015 01:09 PM)Tugdual Wrote:  Still, the 35s is attractive, and I'm aligned with Derek:

Me too. :-)

Dieter

[Michael de Estrada]

I prefer my 33s and haven't used my 35s in years. It doesn't even have batteries in it anymore, which lasted me less than a year.

[Tugdual]

(02-06-2015 01:27 PM)Dieter Wrote:  

(02-06-2015 01:09 PM)Tugdual Wrote:  - the speed: the 15C is about 10 times faster according to my tests

Wait... the original 15C from 1982 is faster than the 35s? I'd say the 35s has about 3x–4x the speed of a 41C and 10x–20x that of a 15C.

Sorry I totally meant the 15C LE
Of course the original 15C is a totally different league!

[Maro]

I agree that the 35s is a really good every day calculator with sound manufacturing quality. However, since I need to do a lot of calculations using complex numbers, there is one point that I really don't like with the 35s. This is the polar representation for which the "theta" symbol is horrible since it is hardly distinguishable from number 8.

[Michael de Estrada]

(02-06-2015 03:33 PM)Maro Wrote:  I agree that the 35s is a really good every day calculator with sound manufacturing quality. However, since I need to do a lot of calculations using complex numbers, there is one point that I really don't like with the 35s. This is the polar representation for which the "theta" symbol is horrible since it is hardly distinguishable from number 8.

If you work a lot with complex numbers, you are better off using a 50g.

[Tugdual]

(02-06-2015 03:44 PM)Michael de Estrada Wrote:  

(02-06-2015 03:33 PM)Maro Wrote:  I agree that the 35s is a really good every day calculator with sound manufacturing quality. However, since I need to do a lot of calculations using complex numbers, there is one point that I really don't like with the 35s. This is the polar representation for which the "theta" symbol is horrible since it is hardly distinguishable from number 8.

If you work a lot with complex numbers, you are better off using a 50g.

Or a Prime.

[Maro]

(02-06-2015 04:18 PM)Tugdual Wrote:  

(02-06-2015 03:44 PM)Michael de Estrada Wrote:  If you work a lot with complex numbers, you are better off using a 50g.

Or a Prime.

Oh yes, that's what I'm actually doing, using my 50g (or the 42S) for most calculations and from time to time playing around with the Prime (waiting for a largely bug-free software and a significantly improved RPN implementation).

Some other points that I don't understand with the 35s is that not all functions are implemented in complex mode like for many other models. Ok, those functions that are missing as direct complex implementation, like hyperbolic functions or complex conjugate, can easily be programmed, but it would be nice to have it directly.

[Thomas Radtke]

(02-06-2015 10:31 AM)Derek W Wrote:  I think the hardware quality is very good, it's just a pity that the firmware bugs couldn't be ironed out, for contractual reasons that Tim W has described in earlier posts.

What were they? Do you have a link by chance?

[Tugdual]

Anybody knows if it is possible to have a comma for decimal delimiter instead of a dot?
Pressing On + dot on the 15C did the trick before.
I suspect this feature was dropped on the 35s because the comma is used for vectors.

[Michael de Estrada]

(02-06-2015 05:28 PM)Tugdual Wrote:  Anybody knows if it is possible to have a comma for decimal delimiter instead of a dot?
Pressing On + dot on the 15C did the trick before.
I suspect this feature was dropped on the 35s because the comma is used for vectors.

Yes. Type <left shift> DISPLAY 6

[Tugdual]

(02-06-2015 05:39 PM)Michael de Estrada Wrote:  

(02-06-2015 05:28 PM)Tugdual Wrote:  Anybody knows if it is possible to have a comma for decimal delimiter instead of a dot?
Pressing On + dot on the 15C did the trick before.
I suspect this feature was dropped on the 35s because the comma is used for vectors.

Yes. Type <left shift> DISPLAY 6

Thank you!

[Tugdual]

I haven't been able to figure out how to enter text in a program.
I could understand that the trick is to use EQN and then I entered text letter by letter using RCL.
Next line is a PSE to show text.
Problem is that I get a different text, it says 'syntax error' which somehow is correct since the EQN means nothing. Anybody can help me to understand what I missed?