So, I got an old OCZ Synapse 64Gb cachedrive that I've been using as a system disk on a HTPC, and so far I managed with the 30Gb's usable space in this firmware locked 50% over provisioning setup that is built into this disk. Today I migrated that HTPC box to a 120Gb SSD and with that I once again are looking into the possibility to somehow access the full disk space of the Synapse disk and I just have a hard time accepting there's no way to either hack the controller (Sandforce 2281) or force flashing it as a Vertex 3 - which as I've understood it is the same disk, though using another firmware...

So, is this possible or am I just gotta settle for only being able to access 1/2 the space of this thing?

regardless of what some sites say they can do with Sandforce, I've never seen any evidence of doing a single firmware hack or any actual fix of a bricked one. These things are often not simply firmware, sometimes actual chips are missing, discreets missing or in different places etc. I think it would be a fruitless endeavour.

Much appreciated. And yeah, this is what I expected - if you can't find sh!t about something, googling it, it's usually not out there.

Follow up question though, the firmware overprovisioning on a disk like this, is that an automated thing from the controller side or is that something that needs a companion software like Dataplex for it to be used?

it is written into the firmware. Controllers can have many different configurations of NAND chips, and you will find firmware does most of the heavy lifting.
Edit - this explains it ok: http://www.kingston.com/en/ssd/overprovisioning

Thanks for the info. I'm gonna slop this into my workstation and do some health- and performance tests and if it holds up, I'll use it for either application caching or the Windows page file... I'm a compositor and FX artist, so I already have a mix of SSD's and mechanical disks set up to handle slow- and fast caching of simulation data, renders, footage, view port caching, etc. and I'm sure I can fit this into that, somewhere.

Well, once again thanks for the info, much appreciated, and I'll set the OP title to solved.

Edit: Oh, replied-to posts aren't editable, so this is not the way they do things here? Or is there another way to mark a thread as solved?

_________________
*** Compositor and FX artist - Random tests and RnD at Vimeo.com/Farmfield ***

we don't worry about a solved moniker, often a thread will get extra stuff added or taken up again years later.. pretty laid back. I think posts are editable for a short time after you submit, then go to read only. cheers

ISTM that all those numbers are wrong. AFAICS, the author has confused GB with GiB.

For example, Intel's 240GB SSD 530 has 16 x 16GiB NAND flash chips, giving it a total physical capacity of 256GiB.

http://www.storagereview.com/intel_ssd_530_review
http://www.storagereview.com/images/Sto ... Bottom.jpg
http://www.storagereview.com/images/Sto ... CB-Top.jpg

Intel's specification states that the user capacity is 468,862,128 sectors. That's a capacity of 240GB.

http://www.intel.com.au/content/dam/www ... cation.pdf

Therefore, the over-provisioning figure is 14.5%, not 7%.

(256GiB - 240GB) / 240GB = 0.145

http://www.google.com/search?q=(256+GiB+-+240+GB)+/+240+GB

_________________
A backup a day keeps DR away.