Sampo, the problem with ASIC is you have to spend millions to get a working chip. FPGA is much, much more flexible and basically the stepping stone to designing ASIC (i.e. a lot of the tools are the same, but if you screw up you can just reflash instead of having to order a new chip). So its worth creating FPGA before going to ASIC; it is still much cheaper (and obviously more flexible) if you will be making less than 100,000 (?) chips. That's why people like Studer or Digico use FPGA in their mixers. In the case of Studer, you can reflash the whole desk and exactly choose the number of buses, channels, processing, etc...
Also for what its worth, none of these configurable chip technologies are quite as simple as you make them out to be. Yeah, you can write code, but you need to take a lot of other stuff into account before your design will run. Like timing. For example, if there is some sequence of logic operations (multipliers, memory reads, etc...), can it actually complete in the period of one clock cycle? Sometimes that period can be as small as 1ns. Compile time can be 45 minutes even for simple stuff. Sometimes modules will simulate fine but not actually run in hardware (requiring more equipment to debug). Although its way easier than EE/old-school IC design, you still can't be 'just' writing code (for that you may as well use ARM/x86). You can get an FPGA dev board with plenty of oomph and high speed USB for $60, but you'll have a hell of a time squeezing any complex C-compiled code on there. If you are interested in higher level FPGA design tools, check out Bluespec<http://csg.csail.mit.edu/pubs/memos/Bluespec/chipdesign.pdf> (also here <http://www.bluespec.com/high-level-synthesis-tools.html>). I haven't used it myself, but it is supposed to be pretty useful. Otherewise, VHDL or Verilog are the way to go. Intel has some cheap Atom dev boards with both FPGA and 1.6Ghz Atom, network, PCI, etc... I haven't had great experience putting audio on ARM platforms if anything else is happening at the same time (like control, network streaming, etc...). But most of our ARM experiments have been on phones and raspberry pi, so that's probably not the best platform. Digging up code from 4 years ago for those who are (still) interested, will try to get it up beginning of next week if I can track it all down. Ben On Fri, May 24, 2013 at 7:21 AM, Dave Malham <dave.mal...@york.ac.uk> wrote: > For info, some approx prices ( inc uk vat); > Beaglebone £36 > AM3359 CPU £30 > ADAU1966 DAC (16 channel 24 bit balanced out, 200kHz > £13 (yes that's thirteen...) > The evaluation board for the DAC is around £250 so not a viable option for > production but a good way of trying it out as it goes to a standard loom > with D type connectors at one end and xlr's at the other. > > Particularly in terms of the DAC, I don't think daisy chaining of cheaper > stereo units could possibly be cheaper, though if you wanted to go for 8 > channel chunks there's the AD1933/4 at about half the price of adau1966. > > Dave > > On 24 May 2013 09:36, Michael Chapman <s...@mchapman.com> wrote: > > > Sampo à écrit: > > > > Seriously, if even this kind of a list really wanted to make something > > happen, we could easily forget about PLAs, FPGAs, Arms, Intels and > > everything else. The price of a pure pedal-to-the-metal ASIC is so low, > > and it's allure so great to major hardware manufacturers (if done > > right), that we could actually jump to that final semiconductor > > technology right now. Seriously. :) > > ___ > > > > I'm usually negative about IP (patents, registered designs, ...), but an > > anecdote: > > Knew a zoologist who invented a wonderful gadget. Published the design. > > Then went to a manufacturer. > > They agreed it was wonderful, but said that tooling up to make it did not > > make commercial sense, as once they'd 'created the market' they would be > > undercut and they'd lose out. > > So if you want to be sure of a manufacturer you may need to be able to > > licence it. > > Sure you can insist on sensible pricing as part of the licence (and not > > take any revenue oneself ...). > > > > Michael > > > > > > PS Minor point,not worth responding at the time: But my 'daisy-chaining' > > argument did involve a 'chain', i.e. a bit of wire between boxes ... for > > the synch, the clock, whatever ... > > MC > > > > _______________________________________________ > > Sursound mailing list > > Sursound@music.vt.edu > > https://mail.music.vt.edu/mailman/listinfo/sursound > > > > > > -- > As of 1st October 2012, I have retired from the University, so this > disclaimer is redundant.... > > > These are my own views and may or may not be shared by my employer > > Dave Malham > Ex-Music Research Centre > Department of Music > The University of York > Heslington > York YO10 5DD > UK > > 'Ambisonics - Component Imaging for Audio' > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > https://mail.music.vt.edu/mailman/private/sursound/attachments/20130524/5faf7979/attachment.html > > > _______________________________________________ > Sursound mailing list > Sursound@music.vt.edu > https://mail.music.vt.edu/mailman/listinfo/sursound > -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://mail.music.vt.edu/mailman/private/sursound/attachments/20130524/73d9794e/attachment.html> _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
