Hi JohnW,
I am trying to figure out what it is you are trying to do here with these 'development kits'. What exactly are you expecting the people who buy these kits to do with them?
Do they develop new software component or just test software that is on the kit or pushed onto the kit? Do they do listening test or functional testing and give you, and us, feed back? Or is this just a way to raise fund by giving people early previews of what you already have designed? Or all of above?
WRT software I hope both will take place - testing and feedback on the our "OEM" software and also 3rd party software installations on the streamers RPI "heart" - I encourage both, and we will offer all the support we can.
So, Yes very much ALL of the above - I need to start funding the purchase of "production" component stock - as full SMD reeled stock is typically 2K to 5K Pcs. we need to purchase 5K pcs of each component - even though we might only produce 100pcs MDAC2 etc. so the solution is to offer as many common designs as possible - the 100 "at cost payments" does not pay for the total stock that needs to be purchased - so funds from projects such as DEVDAC and the Mini Streamer are intended to help bridge the funding gap.
1. We start to ship "units" so we are preparing ourselves for higher quantity items such as the MDAC2 etc. We are putting the logistics in place to handle outgoing shipments, packaging & paperworks etc.
2. A more staggered approach to increasing the production rates - more hands on experience on the SMD line etc. - without the pressure of a 100 odd boards that might need rework due to some unforeseen production related issue on our side.
3. Exposing the bottle necks we will face (such as production space) - it might not sound much, but even a single extra bench added upstairs in our listening area will make a dramatic difference - at the moment we are working on the production boards on our lab benches which are more for R&D then production - I'm coming to appreciate the importance of the different requirements. Being able to manufacture PCB's / units without interrupting R&D is important as we move forward.
I've now purchased the equipment for the production bench ( I cannot say its a production line as we only have a single bench). Once the MDAC2 orders have been met, we only plan small "bespoke" designs so 10 - 20 Pcs a month to keep to boredom of retirement at bay and producing "personalised" designs I believe in.
Also, how and when (a final date for end of testing phase) is all this going to be concluded, resulting in a version of MDAC2 any time soon?
The MDAC2 DD2A will be the first "MDAC2" upgrade design to be released - I'm sure you appreciate that much depends on the progress of the first items to be produced. The MDAC2 design does not live in isolation, as we move forward experience is feedback into the design so its constantly being improved - until it finally meets the production line. I see the path as clear - DEVDAC and the Streamers, then the MDAC2 DD2A & Detox. So once you see the DEVDAC and the Streamers shipped, then you know MDAC2 DD2A is next.
One thing I am noticing, and please correct me if I am wrong, is that you seem to be developing and shaping everything around the streamer, as oppose to the streamer be just a 'optional' component of the DAC.
This has very much been a constant bug bare of yours - and I think everyone else can also see this. Considering the "Streamer" as simply a streamer is a very narrow view of the hardware used that also happens to handle the streaming function.
The CM3 module is used to implement the "brains" of the advanced DACs - driving the UI / displays etc. The CM3 interfaced to our Audio co-processing system is an integral part of the more advance designs where we use the CM3 CPU's processing power (discreet DAC and DD2A) - with these designs we also so happen to leverage the Linux environment to implement streaming services.
But stressing once again its not simply about streaming services as for me they have no relevance, but it should be considered more as a media player so one can replay directly from internal memory requiring no network or wireless connections, no external connected device - this is how I'll use my DAC most of the time.
The reason we use the CM3 and not simply the cheaper RPI solution is so that we can optimise the "Linux hardware" for audio applications with careful design to insure all clocks and switching PSU's are synchronised to the Audio master clock, without these design details it would be no different from just a separate Linux PC connected to a DAC with all the EMC and noise issues this entails.
Many of the lessons learnt from the Detox R&D are incorporated into the "streamer" solution. A great part of the "streamer" design's complexity is managing these clock domains, PSU's and interface noise. I'm personally unaware of any other Audio hardware (asides from our own pending designs) that goes to these lengths.
In a world where we cannot escape less physical media we need to engineer "Audiophile" solutions otherwise its over for our small world... I'm not interested in streaming services - this is a secondary function, but a self contained media player is interesting so long as it does not loose sight that we are a HiFi product not a computer.