HYNX HY27UV0BG5M

[DRUG]

1. fun isn't it?
2. do you have a logic analyser to watch what is happening?
4. try to find a flash drive with any of the following controllers, as these shouldn't have XOR...
UT165-L48, UP10, SM321QF, SM3251 Q BA, SM3252 Q BB, SM3253 L AE
5. maybe and maybe not. the fingerprint, if a digital representation is stored on the flash, would be VERY interesting to analyse. but it may be just an unlocker to the NAND, and controller-to-NAND is same as any other. I may have to look into these. Maybe take it apart and see how many chips. If there is an extra eeprom for example(could they be this stupid?) then probably nand is normal. One thing to check is the marketing material for the flash drive. If it mentions encryption, it could be a waste of time..

1- It's fun and frustrating, the only thing that keeps me going on is the hope that one day I manage to compile all this on my brain.
2- I was considering the LA that I pmed you in the other day, but since that isn't good enough I'll keep looking.
3- All I want now is to find a way to understand how to do the pinouts correctly when I don't have acess to the datasheets.

It would also be helpfull to understand why I'm only being able to read for example 4 MiB out of 2GB flash drive ( i assume 99% it's bad wiring but I just can't guess )

[jermy]

3- All I want now is to find a way to understand how to do the pinouts correctly when I don't have acess to the datasheets.

Oh wait when you get to Monolithic

[DRUG]

3- All I want now is to find a way to understand how to do the pinouts correctly when I don't have acess to the datasheets.

Oh wait when you get to Monolithic

I've heard about how hard Monolithic chips may be, but the thing is, how do people crack the pinouts ? Do they have some kind of aid, like a LA ?

[fzabkar]

ISTM that the most logical way to analyse the wear levelling and flash translation methods would be to write a visually recognisable data pattern to each LBA. For example ...

LBA (h) xxxxxxxx
(dec) yyyyyyyyyy
pass #nnn

You could write your own tool to do this, or you could create an appropriate 4GB file and then clone it to your flash drive using Linux dd or a disc editor. You would increment the pass count for successive clones. Spildit and I have used this method to analyse the cache RAM structure of a Hitachi HDD. You could also use such an approach to analyse a RAID.

[fzabkar]

3- All I want now is to find a way to understand how to do the pinouts correctly when I don't have acess to the datasheets.

It would also be helpfull to understand why I'm only being able to read for example 4 MiB out of 2GB flash drive ( i assume 99% it's bad wiring but I just can't guess :x)

Start by showing us a detailed photo of the PCB. That should tell us which traces are in use.

If you have a CCD scanner, that takes much better images than a camera.

[HaQue]

ISTM that the most logical way to analyse the wear levelling and flash translation methods would be to write a visually recognisable data pattern to each LBA. For example ...

LBA (h) xxxxxxxx
(dec) yyyyyyyyyy
pass #nnn

You could write your own tool to do this, or you could create an appropriate 4GB file and then clone it to your flash drive using Linux dd or a disc editor. You would increment the pass count for successive clones. Spildit and I have used this method to analyse the cache RAM structure of a Hitachi HDD. You could also use such an approach to analyse a RAID.

The tool to do this is written, http://www.flash-extractor.com/downloads/ "pattern_writer". It writes numbers to each LBA. Main problem is you need a working device EXACTLY the same to test. I have bought 20pcs of 8GB Sandisk Cruzer from Officeworks from the exact same rack, same device, same day etc.. and result of device dupe count was 9 4 3 2 1 1. Similar controllers like 20-99-00515-4, 20-99-00515-5 but totally different Mix's. This is why I have bought over 500 Flash devices.. to study to help enhance Rusolut VNR.

When you get in a device to analyse, it is difficult finding another the same.. especially when companies like Verbatim kep the same looking device for years(Store-n-Go), Sandisk Cruzer, ony Microvault .. all have many 10's of inside configurations of controller+memory

[HaQue]

yes, for pinouts need logic analyser, or a good data recovery tool that supplies pinouts with support. There are MANY pinouts in use. not many standards here..