[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Date Index][Thread Index][Author Index]

Re: FCB1010 wishlist (was: LD Community)

To those curious about future fcb1010 upgrades:

It is VERY UNLIKELY that many of your ideas will ever be incorporated into
future firmware releases to the FCB1010.  Why such a dismal view?  Well,
I've reverse engineered the electrical schematic of the FCB1010, and along
with several other interested people, reverse engineered the sysex format
for saving patches (leading to the software editor now available on my 
group FCB_sysex_tool, or on the general FCB yahoo group).  In addition I've
begun a reverse engineer of the firmware in the FCB1010 with a couple of
other interested people (with the idea in mind to do an update ourselves).
So I know a lot about how the FCB1010 was designed and implimented.

Behringer cut a lot of corners, most likely to maintain some overall target
cost of the unit (which is probably the number one reason why the unit 
so well, sadly).  They chose a serial EEPROM chip (the main non-volitile
storage chip - permanent memory) which is only 2Kbytes big.  This is a 
chip. For about $3.00 they could have used a pin-for-pin chip from the same
manufacturer which is 64Kbytes big (32 times the storage for only twice the
price).  Note for all you tweakers that this is a tiny surface mount chip
soldered directly on the board, so its not easy for a trained tech to
change, let alone a random user.

Because the chose a tiny little wimpy memory chip, they had to jump through
a lot of hoops in implimentation to get 10 banks of midi data storage out 
it.  Basically they do not have room to store the data direcly as midi 
they have a simple compression scheme which only allows for their idea of
usefull midi data assignments.  Their data storage routine is SEVERELY
CRIPPLED.  There is no possible way to support multiple midi channels for 
expression pedal, no way to support sending note messages to more than one
midi device, and most importantly, no way to support sending arbitrary
strings from a button press.  Unfortunately, the way they've designed it,
without a complete code redesign, this cannot change.

If they redesigned the firmware implimentation (and this would be a huge
redesign effort) they could support these things a la the PMC controller.
However they'd still have the upper limitation of 2K of non-volitile data,
which would mean something more like only being able to use 2 or 3 banks
rather than 10.

I think this is one more case of a marketing department compromising the
overall quality of a final product.  By saving that extra $1.50 on the cost
of goods, they forced their designers to severely compromise the
functionality of the product.  Who really knows what would have happened if
they put more memory in there, and released a product that could be 
over time, but was more expensive.  Maybe they wouldn't have sold as many 
it was more expensive.  Or maybe there would be lots of people wanting to 
esoteric midi stuff drooling over it and buying two.  Basically you can't
have your tea and eat it too - either cost or functionality, its always a
big question.

Bottom line:  Keep sending your ideas to Behringer.  They need to know what
their customer wants so they can design good products.  But don't expect
results any time soon.  They've designed themselves into a cost cutting
corner, which they could only get out of by issuing new and improved