[sdiy] Synth Keybards and Number of Keys

[Edward King]

I disagree.

Having just gone down the road of custom-building my own keyboard, I took
several years to investigate all the options.
The real cost in building keyboards is in the initial design, tooling and
any custom silicon baking, NOT the unit cost. The unit cost is very much
secondary.

It costs a fortune to set up the tooling for an octave of plastic keys, but
very little for the actual manufacture, PCB's switches etc. A key costs
(according to the figures I was presented with anyway) approximately £0.68 +
V.A.T if it is part of a run.

I could have paid nearly £80,000 for a factory to set up the tooling and
moulds etc and the price difference of 10 keyboards to 1 keyboard was
negligable as - like most automated processes - once you get it set up, it
costs nothing.
I also looked into 61, 76 and 96 key variants and the result was the same.
Its just one of those situations where the technology (moulded hard
plastics) is great, but really only suitable for large runs where the
company concerned can justify the initial outlay but once you have paid for
the setup, the unit cost will remain pretty much the same (petrochemical
prices dictate of course).

Im happy to say that I didnt go the moulded plastic route (I actually prefer
wood) and it cost me a fraction of the above cost to get everything I
wanted, but I would be lying if I said that I didnt come close to biting the
bullet and paying for a nice shiny plastic thing that would have weighed
significantly less than my back-breaking machine (which Im going to have to
design wheels for, I think) and could have been customised beyond anything I
could get commercially.

Most of the cost of synths up until now (because the landscape is changing
dramatically) has been R&D and development hardware.
Up until FPGA's and other programmable devices became stable enough to use
commercially, companies had to either use COTS DSP's or bake their own
silicon (usually through a third party) which also costs a fortune. A lot of
stuff was done with microcontrollers and MPU's, but even these arent cheap
in the long run because someone somewhere still has to develop the code and
people cost money.

The limit of impact of cost to key-numbers is if you have an opportunity to
save money (however small the amount) by stripping the design down to 61
keys from 88 without it affecting sales, then of course you'll do it. But I
dont think thats the only driving force.

I think it is probably a balance of marketing (some people shy away from
bigger machines and thats a fact), cost and the trend. Lots of people
actually train on 61 note keyboards (go into any keyboard music school and
see for yourself) and so thats what they stick with.
If lots of music schools buy 61 note keyboards (however innocent that
decision may be) and all of their students learn on them, then what the
hell? The manufactureres see the sales doing okay and stick with what
appears to be working.
It doesnt matter to some extent what the original motivation was; if it
sells, it sells.

If on the other hand you go into a proper music school and see everyone
training on pianos, theres a good chance that although most of those
individuals will have pianos / 88 key machines at home, a fair number will
also have the bog-standard 61 key keyboard.

[snip]
>
> Cost of parts, plain and simple.  There are only a few OEM's of piano-keys
> (Fatar, Yamaha, etc.) and unless you buy in large bulk, you won't see the
> economies of scale required to turn a profit on the keyboard; the
> piano-keys part alone is a huge percentage of the total cost of the
> average synth.
>

[snip]

>
>
> Its a pure costs thing.  Most synths that you can buy on the market today,
> still surviving the onslaught of software-synthesis and PC-based DSP, have
> extremely tight profit margins.  The markup on your average synth is quite
> heavy as well; with something like 4 or 5 different tiers to pass through,
> synth production and price are managed with a very tight scale of economy.
>
> Organs, generally, have a much more targetted market willing to pay a
> higher cost for the item, which typically will be installed, rather
> prestige-like, and the 'idea of higher cost' for such an item is certainly
> more than for the consumerist-like approach in the synth market.
>


[snip]

>
> I'd personally like to see someone DIY a piano-keyboard with naught but a
> wireless-LAN connection, so I can use it to write software for the
> Nintendo DS, but hey .. ;)
>

Mine has a "streaming MIDI over Lan" aspect in the design, but its run from
the master cpu, not the keyboard itself. I dont think it would be too hard
to adapt this for your purposes since it would simply be a case of taking my
keyboard design + HID handler + MIDI handler and substitute the GBE module
and switch with a wireless card. Modularity...you gotta love it. 


		
___________________________________________________________ 
Now you can scan emails quickly with a reading pane. Get the new Yahoo! Mail. http://uk.docs.yahoo.com/nowyoucan.html