Threaded Mode | Linear Mode

Alberto Candel · 07-25-2014, 05:34 AM

How many bad flash blocks does your Prime have? I have one unit with 12 and another one with 6. I wonder if this is such a significant number that may affect the performance of the Prime.

To check, you hold down C F O while turning the unit on, then run the test by pushing key 5 (Evaluation), then key 4 (CHK Bad Flash).

PS. I have found a related post in the old HP forum, but not much follow up on it.

akmon · (This post was last modified: 07-25-2014 06:10 AM by akmon.)

I get eight bad blocks. How dangerous is it?

eried · 07-25-2014, 06:16 AM

1 BAD here

ColinJDenman · 07-25-2014, 09:52 AM

5 (classed as "NG" on the test result page)

toml_12953 · 07-25-2014, 11:39 AM

I have 6:
0, 1, 397, 514, 747, and 1427

Tim Wessman · 07-25-2014, 02:49 PM

(07-25-2014 05:56 AM)akmon Wrote: I get eight bad blocks. How dangerous is it?

Not at all. They have been marked as problematic and are not in use. They won't be increasing or anything and having blocks "suddenly" go bad. They basically come that way from the chip manufacturer.

ndzied1 · 07-25-2014, 04:34 PM

Ah, How do I get out of this mode and back to the normal calculator?
Thanks

Kevin Ouellet · 07-25-2014, 05:02 PM

(07-25-2014 02:49 PM)Tim Wessman Wrote:
(07-25-2014 05:56 AM)akmon Wrote: I get eight bad blocks. How dangerous is it?

Not at all. They have been marked as problematic and are not in use. They won't be increasing or anything and having blocks "suddenly" go bad. They basically come that way from the chip manufacturer.

Does it mean that our calculator will have fewer disk space for program storage if it has many bad flash blocks?

Alberto Candel · 07-25-2014, 05:17 PM

(07-25-2014 04:34 PM)ndzied1 Wrote: Ah, How do I get out of this mode and back to the normal calculator?
Thanks

ESC to the long list, then Key 9 (Reset)

eried · (This post was last modified: 07-25-2014 06:56 PM by eried.)

(07-25-2014 05:02 PM)Kevin Ouellet Wrote:
(07-25-2014 02:49 PM)Tim Wessman Wrote: Not at all. They have been marked as problematic and are not in use. They won't be increasing or anything and having blocks "suddenly" go bad. They basically come that way from the chip manufacturer.

Does it mean that our calculator will have fewer disk space for program storage if it has many bad flash blocks?

In the computer SSDs there is a bit of slack for bad blocks, probable these are the same chips, <1% is acceptable. Some say that the upcoming hardware revision will be hand-check by white clothed technicians assuring zero-bad-blocks.

Michael Lopez · 07-26-2014, 12:46 AM

Mine reports 5 bad blocks: 0,1,236,606,1033.

Cheers,

Michael

DrD · 07-26-2014, 09:34 AM

Idle time challenge #1

With a flash memory size of 256Mb, and a typical block size of 64KB, what is the smallest ppl program, using your calculator's bad block value, to rate your calculator's flash memory quality index?

Too much idle time challenge #2

If you are especially clever, extend the smallest ppl program, using each technical parameter available within the Prime, to assign an overall performance factor when run on an hp Prime.

Winner of the "Smallest hpppl Program" challenge receives less idle time.

Disclaimer: Not affiliated with anything useful, hp or otherwise. Open to everyone.

Thomas Radtke · 07-26-2014, 11:52 AM

(07-25-2014 02:49 PM)Tim Wessman Wrote: Not at all. They have been marked as problematic and are not in use. They won't be increasing or anything and having blocks "suddenly" go bad. They basically come that way from the chip manufacturer.

Out of interest: Why did you make that number easily accessible? Isn't this asking for trouble where there's no problem really? No criticism, I like this kind of openness, of course.

dan_h · 07-26-2014, 02:24 PM

7 bad blocks: 0 1 82 286 584 776 1895

Jsather · 07-26-2014, 03:26 PM

(07-26-2014 09:34 AM)DrD Wrote: a typical block size of 64KB

Is 64KB per Block authoritative? That would be 2^28 / 2^16 = 4096 blocks total. I have eight bad blocks 0 1 273 942 967 1052 1319 1711, and notice that no one has reported a bad block higher than 2048. It could be that there are 128KB per Block in this Flash or perhaps they are 64K and only 2048 are being tested. 8/2048 is .004 so I'm guessing I have .4% bad blocks.

After I escape back to the Evaluation Menu, my Flash test is marked "NG" (probably you need 1% failures to rate an NFG) whereas the other tests result in "OK". Who out there is getting NG and who is getting OK for their flash test?

Alberto Candel · 07-26-2014, 05:26 PM

I have 2 units, with 12 and 6. The numbers are:

12 Bad Blk: 0 1 196 450 477 873 1036 1063 1169 1865 1906
6 Bad Blk: 0 1 3 5 539 620 1050

0 and 1 seem rather frequent.

Jsather · 07-26-2014, 06:11 PM

Thanks to a picture from ereid at http://www.hpmuseum.org/forum/thread-1803.html , we know the flash memory chip is a K9F2G08UOC whose manufacturer's data sheet can be found at http://www.wasuntech.com/FileUp/file/201...5_1660.pdf .

On page 11 of that advanced data sheet, it states that the chip will have a minimum of 2008 valid blocks and a maximum of 2048 which means the manufacturer guarantees no more than 40 bad blocks, roughly 2%. Reading between the lines, a few bad blocks is to be expected for this technology and we should expect a few in a typical HP Prime.

The data sheet also says that the first block will be valid and instructs us not to attempt to program or erase blocks marked invalid by the factory.

That satisfies me. No point fretting over a few bad blocks. I would only advise HP to change their diagnostic test to indicate "OK" instead of "NG" if the number of bad blocks is acceptable by their standards.

toml_12953 · 07-26-2014, 07:35 PM

(07-26-2014 11:52 AM)Thomas Radtke Wrote:
(07-25-2014 02:49 PM)Tim Wessman Wrote: Not at all. They have been marked as problematic and are not in use. They won't be increasing or anything and having blocks "suddenly" go bad. They basically come that way from the chip manufacturer.
Out of interest: Why did you make that number easily accessible? Isn't this asking for trouble where there's no problem really? No criticism, I like this kind of openness, of course.

It reminds me of the old MFM hard drive days. Manufacturers always included a sticker on the drive that showed bad sectors. I never had to do anything with the list since DOS format automatically locked out bad ones. Nowadays with IDE drives, any flaws are hidden and never cause a problem either.
Tom L

eried · 07-26-2014, 09:07 PM

(07-26-2014 07:35 PM)toml_12953 Wrote:
(07-26-2014 11:52 AM)Thomas Radtke Wrote: Out of interest: Why did you make that number easily accessible? Isn't this asking for trouble where there's no problem really? No criticism, I like this kind of openness, of course.
It reminds me of the old MFM hard drive days. Manufacturers always included a sticker on the drive that showed bad sectors. I never had to do anything with the list since DOS format automatically locked out bad ones. Nowadays with IDE drives, any flaws are hidden and never cause a problem either.
Tom L

S.M.A.R.T. probably track all these stuff in case of IDE/SATA drives. I checked my SSD and it says 0 damaged blocks but lots of 'data noise' (probably due the cable looping another sata cable)

Tim Wessman · 07-27-2014, 12:04 AM

(07-26-2014 11:52 AM)Thomas Radtke Wrote: Out of interest: Why did you make that number easily accessible?

That is purely coded and made by the device manufacturer and used for factory purposes.