How many BAD BLK?

[Jsather]

(07-26-2014 05:26 PM)Alberto Candel Wrote:  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.

Yes 0 and 1 do seem to be marked bad for everybody, and block 0 is the only block guaranteed to be good by the manufacturer. That's enigmagic.

Don't know why they're listed as invalid, but no way is block 0 invalid nor probably block 1 in everybody's Prime. Here is the direct quote from the manufacturer's data sheet: "The 1st block, which is placed on 00h block address, is guaranteed to be a valid block up to 1K program/erase cycles with 1bit/528Byte ECC."

[Alberto Candel]

(07-26-2014 06:11 PM)Jsather Wrote:  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.

My memory chips have the label K9K2G08UOM.

[Jsather]

(07-27-2014 12:47 AM)Alberto Candel Wrote:  My memory chips have the label K9K2G08UOM.

As far as this thread is concerned, the U0C and U0M have the same minimum of 2008 valid blocks from the factory. There are differences between the U0C and U0M in read cycle time and cache program support, but we can't infer much from that about performance in the Prime.