[sdiy] $53 Intel Edison dual 500MHz Atom 1GB RAM SOC board has I2S
nvawter
nvawter at media.mit.edu
Wed Dec 3 17:27:25 CET 2014
hmm yeah, can you tell what the power consumption is? I was vexed by
these specs:
Power
Input:
3.3V - 4.5V
Output:
100mA @ 3.3V and 100mA @ 1.8V
It definitely does look capable and compact. And being x86 will
probably run
many nice things like a realtime linux kernel. (Anyone familiar with
those,
I'd like to get up to speed).
Also, do you know conceptually, from an API standpoint, how one puts it
into
low-power mode? Is it simply a function that the operating system can
take
care of automagically?
btw I have a feeling in the near future, users are going to want synths
that
act more like "internet-of-things" devices. That is, realtime upgrades
and
what not. I'm still trying to conceive of the perfect architecture for
doing that.
If anyone has any ideas, I'd like to hear them. It would of course be
excellent
to be able to design patches in a browser on a phone then transfer them
to the synth
without having to stop playing music, etc. Also, to have
"subscriptions" to patches,
extract patches from songs, and share patches over social networking.
I'm sure it's
possible, but the question is how to maintain this capability as the
availability
of hardware and software platform shifts along underneath us!
On 2014-12-03 09:01, Robin Whittle wrote:
> Hi Terry,
>
> I was using the term "DSP" rather loosely. A quick look at this board
> made me think it would be possible to have two or more channels of
> audio
> going in and out of the board, with all the audio processing software
> written in C++ (GCC - g++), using ordinary 64 bit x86 floating point
> instructions. It has 1GB of RAM and 4GB of Flash.
>
> There would probably be lower power and more compact ways to do intense
> DSP work with suitable chips, which would be BGA etc.
>
> I was just thinking of writing from the ivory tower comforts of C++ in
> a
> familiar Linux environment and having the whole thing run on its own
> from Flash memory, using a watt or so (I guess, maybe more for the two
> cores working hard), all in a space about the size of a matchbox or so,
> without requiring heroics with BGA chips or even much surface mount.
>
> This link should provide the latest forum messages of interest:
>
> https://communities.intel.com/search.jspa?q=I2S+Edison
>
> I haven't looked at them in detail.
>
> In principle it would also be possible to connect a USB audio
> interface,
> but I think there is only one USB connection and the directness and
> compactness of an actual CODEC chip, or separate ADC and DAC chips,
> going straight into a dual-core Linux SOC costing $53 . . . The 1.8
> volt
> signal levels will require some level translation, since I think CODECs
> normally use 3.3V.
>
> - Robin
>
>
> On 2014/12/04 12:24 AM, Terry Shultz wrote:
>> Hi Robin,
>>
>> Are there any DSP Libs available for the INTEL ATOM?
>>
>> We rejected it a few years back at GM due to the lack of any DSP Libs
>> and the ARM platform Could use the iEEE 758 ? libs as a starting
>> point.
>>
>> My Memory on this is somewhat dated but it was a point of contention
>> and Intel had already sold off their ARM business to Marvell I believe
>> by that point.
>>
>> Perhaps you have more current info and could illuminate this a bit
>> further.
>>
>> I presently use Sharc and Ti DSP and several TI device I am using have
>> ARM 9 + DSP.
>>
>> DSP functions are not clear on this device but I2S is always welcome
>> if they are not MUXed with other functions necessary.
>>
>> This is yet another thing to look out for on some of the ARM devices
>> as it seems Audio I/O is not as forefront as Video needs and other I/O
>> ports used for Memory, Network etc.
>>
>> Just my 2 cents.
>>
>> best regards,
>>
>> Terry
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
More information about the Synth-diy
mailing list