Author Topic: HP-15C ran out again, after 12 hours of availability on their website hp.com.  (Read 7678 times)

0 Members and 1 Guest are viewing this topic.

Offline Lawsen

  • Frequent Contributor
  • **
  • Posts: 251
  • Country: us
The HP-15C a Kinpo made Sunpak 6502 core processor (my mistake, it is ARM 7TDMI) and powered by two Li button cells, CR2032, unlike the HP-12C Platinum series is Kinpo made and powered by one Li button cell, but similar Sunpak 6502 core processor (no) and with algebraic OS, too.  The HP-15C has no algebraic as in the HP-12C Platinum edition.  The HP-15C is exactly the same keyboard and function as the Saturn series processor, 85 thousand transistors at 0.25 mW and tested to endure no current at 4000 V static electricity according to this website:

http://www.hpmuseum.org/tech10.htm

Is the new Kinpo made HP-15C based on the Sunpak 6502 core (ARM 7TDMI) as durable as the original Saturn series processor?  Will the battery life be as long as the Saturn with the three SR44 silver oxide batteries? 

The HP-15C market is very volatile.  It is like a speculator or bubble inflation market.  It is meant to do math, why so much speculator like a house?  This is absurd.  I have a lot of favorites in HP products like the HP-34401A large multimeter and HP net book and tablets and printers.  I repair their giant plotters, but do not own one myself.  I have never seen an HP product bought to be resold for higher prices.  What happened to our economy? 

Alternative to the HP-15C limited edition, might be to purchase a clone made in Switzerland, but I have not tried this, yet:

http://www.rpn-calc.ch/#introduction

I did not received my HP-15C limited edition, yet to take it apart to see what is inside it.  Some HP calculators are not take apart able like the HP-28S and 32S Two, those are permanently snapped together.  Surf here:

http://en.wikipedia.org/wiki/HP-10C_series

http://mycalcdb.free.fr/main.php?l=0&id=815

Surf here, a better view of the HP-15C limited edition:

http://eda360insider.wordpress.com/2011/09/17/eda360-and-the-hewlett-packard-15c-limited-edition-calculator/
« Last Edit: September 29, 2011, 05:35:12 am by Lawsen »
 

Offline slateraptor

  • Frequent Contributor
  • **
  • Posts: 833
  • Country: us
A few comments on the subject:

The HP-15C is exactly the same keyboard and function as the Saturn series processor...

The new 15c is essentially a Saturn emulator...one which doesn't emulate the [ON] + [multiply] self-test at that despite the instructions given in the manual.


Is the new Kinpo made HP-15C based on the Sunpak 6502 core as durable as the origianl Saturn series processor?  Will the battery life be as long as the Saturn with the three SR44 silver oxide batteries?

I don't think the question is of durability. After all, we are talking about packaged silicon in an enclosure not intended for regular disassembly. Computation-wise, this is not an issue; if you're using a 15c to perform a calculation that requires more than a few seconds, then you're probably using the wrong tool for the job to begin with.

I think the question is really about power robustness, i.e. given that the calculator's architecture is an emulator of legacy hardware, how will it perform with respect to battery life? I would like to think the technology node that the 6502 is fabricated at suggests comparable longevity. In terms of mAh, three SR44 ~= two CR2032, and visible form factor isn't affected so this isn't really an issue (unless you're one of those "purist" nonsense types). If the 6502 implements low-power states, then I don't think power will be an issue. Nevertheless, if the calculator lasts a years before having to replace batteries, then I'd be fully satisfied; judging by the performance of my 35s, I don't think the new 15c will have any issues meeting this benchmark.


The HP-15C market is very volatile.  It is like a speculator or bubble inflation market.  It is meant to do math, why so much speculator like a house?  ...  I have never seen an HP product bought to be resold for higher prices.  What happened to our economy?

This happens all the time, e.g. think of every new Apple release. And if you think technology scalping is bad, you need to look into ticket scalping; it's so bad that some states in the US have outlawed it. The real problem are these business-oriented idiots who dive into the tools of a niche community without an understanding of how the product became so popular to begin with. For all we know, these guys probably based all their business decisions on the nominal selling price of 2nd-hand calculators on eBay. In the end, however, these scalpers are the only one's who will end up losing.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 27939
  • Country: au
    • EEVblog
The new 15C and 12C do NOT run on the Sunplus 6502, they use the Atmel ARM processor running at 30MHz.
Battery life sucks, with the same massive design mistake I explained back in blog #4 or something.

Dave.
 

Online Psi

  • Super Contributor
  • ***
  • Posts: 6669
  • Country: nz
Battery life sucks

Do you know roughly how bad?
Are we talking something crazy like <1 hour
Or just that it's not as efficient as they could have made it if they had put a bit more effort in?
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9353
  • Country: us
Battery life sucks

Do you know roughly how bad?
Are we talking something crazy like <1 hour
Or just that it's not as efficient as they could have made it if they had put a bit more effort in?
I will find out soon. I have one on its way and I will measure the battery current if I can find a convenient way to hook it up.
I'm not an EE--what am I doing here?
 

Offline slateraptor

  • Frequent Contributor
  • **
  • Posts: 833
  • Country: us
The new 15C and 12C do NOT run on the Sunplus 6502, they use the Atmel ARM processor running at 30MHz.
Battery life sucks, with the same massive design mistake I explained back in blog #4 or something.

Dave.

Just verified (see attached image).

Atmel AT91SAM7L128 in 128 LQFP to be exact.

http://www.atmel.com/dyn/resources/prod_documents/doc6257.pdf

Pages 12-13 and chapter 35.3 are relevant to the discussion. At first glance, it doesn't look too bad...then again, I haven't crunched the numbers quite yet.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 27939
  • Country: au
    • EEVblog
Battery life sucks

Do you know roughly how bad?
Are we talking something crazy like <1 hour
Or just that it's not as efficient as they could have made it if they had put a bit more effort in?

It draws 30mA from the two parallel CR2032 batteries every time you press a key! or do a calculation. So you can do the math on how long it will last if you run it continuously calculating.

Dave.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9353
  • Country: us
It draws 30mA from the two parallel CR2032 batteries every time you press a key! or do a calculation. So you can do the math on how long it will last if you run it continuously calculating.
But if you don't press any keys? If it is just sitting there, waiting for a key press?

Because then I don't suppose it would make any difference to run the CPU at a lower clock speed. Any given computation will take a defined number of CPU cycles to complete, and I'm guessing that 500 CPU cycles takes about the same amount of charge per cycle whether run at a fast rate or a slow rate?

If the processor is in idle mode and drawing microamps unless it is calculating, then I am not sure what could be done to make an improvement?
I'm not an EE--what am I doing here?
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 27939
  • Country: au
    • EEVblog
It draws 30mA from the two parallel CR2032 batteries every time you press a key! or do a calculation. So you can do the math on how long it will last if you run it continuously calculating.
But if you don't press any keys? If it is just sitting there, waiting for a key press?

Because then I don't suppose it would make any difference to run the CPU at a lower clock speed. Any given computation will take a defined number of CPU cycles to complete, and I'm guessing that 500 CPU cycles takes about the same amount of charge per cycle whether run at a fast rate or a slow rate?

If the processor is in idle mode and drawing microamps unless it is calculating, then I am not sure what could be done to make an improvement?

That's what a lot of people think, but it's not the case when powered from CR2032 cells like this, there is a trap for young players. I have done a video on this exact problem in the new design HP calculators two years ago. HP responded and said they acknowledged the problem and would not make the same mistake again. You guessed it, they have.



Dave.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9353
  • Country: us
That's what a lot of people think, but it's not the case when powered from CR2032 cells like this, there is a trap for young players. I have done a video on this exact problem in the new design HP calculators two years ago. HP responded and said they acknowledged the problem and would not make the same mistake again. You guessed it, they have.
I see. And I suppose the higher current draw is going to trip the low battery warning sooner as well due to the increased voltage sag...
I'm not an EE--what am I doing here?
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 27939
  • Country: au
    • EEVblog
Re: HP-15C ran out again, after 12 hours of availability on their website hp.com.
« Reply #10 on: September 28, 2011, 06:33:58 pm »
I see. And I suppose the higher current draw is going to trip the low battery warning sooner as well due to the increased voltage sag...

Yup, and that just adds insult to injury.

Dave.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf