HP Forums

Full Version: HP 15C LE Memory Corruption After Self Test Performance
You're currently viewing a stripped down version of our content. View the full version with proper formatting.

In other threads today I reported failure of the self tests normal to the 15C and described on page 261 of the 15C LE manual.

But now, I can report that attempting to perform the "ON plus *" self test not only results in immediate "Error 9", but also corrupts any program that is in memory.

I had a Savage benchmark in memory consisting of the following steps:

001-42,21,12

002-44 0

003-43 8

004-42,7,9

005-0

006-42,21,0

007-1

008-40

009-11

010-43 11

011-43 12

012-12

013-43 25

014-25

015-42,5,0

016-22 0

017-43 7

018-43 32

(FWIW, putting a number in the X register, followed by "f" and "B" executes the Savage Benchmark for the number of iterations in X. If X holds the traditional 2500, the results will show in about 48 seconds.)

After turning the 15C LE off, then executing "ON plus *", I get "Error 9".

Then, when I go to program mode, here's what's I find there now:

001-42,21,11

002-44 1

003-43 8

004-42,7,9

005-0

006-42,21,0

007-44,20,12

008-11

009-43 11

010-43 12

011-12

012-43 25

013-26

014-42,5,24

015-22 0

016-43 7

017-43 32

Really really strange!

I did a reset with "ON plus -", got the expected and proper "Pr Error", then reprogrammed the 15C LE. I was able to get this conditionn to repeat.

I conclude that one won't want to try any self tests if there is something in program memory!

What else lurks out there? I've only spent a couple of hours with my new 15C LE.

I'm not all that happy with this, but it seems to have some, er, possibilities. :-)

Edited: 9 Sept 2011, 9:50 p.m. after one or more responses were posted

Programming synthetic there we will :-)
And for $79 what do you want?
Maybe changing the system and putting the WebOS

Edited: 9 Sept 2011, 9:06 p.m.

An updated 15C with micro SD card storage would have been a quantum leap in technology. never mind about keeping true to it's original specs.

But...any SD card capability would likely eat the 2032 batteries up.

Those are not valid self tests. They were supposed to be disabled or removed. I also told them to remove any mention of that in the manual. Apparently it was left in.


Simple solution, don't do it. :-)

TW

Edited: 9 Sept 2011, 9:31 p.m.

So when will the errata sheets be available ?

Monday, when I give it to the support guy to put in the call center. And they are never availiable.

TW

Edited: 9 Sept 2011, 9:41 p.m.

The SD card spec allows the cards to draw 150mA at 3.3V, and even more in some of the newest high-speed modes. Coin cells are rated for a few mA maximum.

Quote:
Those are not valid self tests. They were supposed to be disabled or removed. I also told them to remove any mention of that in the manual. Apparently it was left in.

Simple solution, don't do it. :-)


Well, er, of course.

But I have the honor to suggest that is not an elegant solution to something that is, after all, canonical in the 15C LE manual.

It is a condition fairly easily detected. :-)

Edited: 9 Sept 2011, 9:48 p.m.

That's something beta-testers missed :-)

Well, I hope ON/. works. It does on my 12C+.

Quote:
That's something beta-testers missed :-)

Time for the master betas! :-)

Quote:
But now, I can report that attempting to perform the "ON plus *" self test not only results in immediate "Error 9", but also corrupts any program that is in memory.

Huh, sounds like something is getting stomped on elsewhere as
a result of the test. The system self test does cycle through
memory alternately writing "55555555555555" and "aaaaaaaaaaaaaa"
patterns but takes care not to permanently alter data.

Will updated firmware be made available that either locks out those tests that aren't suppose to be there, or implements them?

-Katie

I think the beta testers were told the self-tests were not there so they assumed they were not there! Doh!

Quote:
The SD card spec allows the cards to draw 150mA at 3.3V, and even more in some of the newest high-speed modes. Coin cells are rated for a few mA maximum.

True. But demountable SPI EEPROM in the new HP Voyager Flash
Card format would have worked quite well from all concerns.

Looks like a missed opportunity to seed a proprietary media
design into a captive market.

2000 line programs written in the cryptic numeric codes? :)

Quote:
An updated 15C with micro SD card storage would have been a quantum leap in technology. never mind about keeping true to it's original specs.

These really could be limited edition collectors items :)

Daniel