John,Due to the continued issues I'm experiencing with MDAC build quality, sound quality variability and our longer term plans to progress forward we are considering changing the direction as to how we handle these MDAC upgrades.
Currently each MDAC upgrade is taking about 4 to 5 days to complete as I'm essentially customising the upgrade for each unit this is obviously unsustainable, extremely frustrating and not to mention financially unprofitable.
The plan that is now brewing in my mind (Im putting it out here for feedback) is to offer a whole new MDAC Mainboard that we would manufacture here in Europe thus insuring quality and consistency. To allow us to meet the price targets we would purchase key expensive items in Asia cheaper passive components etc would be purchased locally here in Europe.
As the PCB is built by SMD machine, labour is such a small part of the board cost we could still meet our price targets for the upgrades with a higher quality and more professional product.
To help finance this endeavour Im thinking upon the lines that say 50 perspective customers send instalments of GBP100 per month and within 4 months (GBP400) we will be able to ship out upgrade boards (for simple DIY fit) or I can upgrade units here in CZ as Im currently doing being a simple board swap it would be a much quicker process. It would also allow owners outside of Europe to upgrade there unit.
You would be essentially getting a new product with only front panel, PSU and chassis reused from the original MDAC.
Id go the full hog with the redesigned PCB, not only including the Toy / Fusion upgrades but include Dual pair of ESS9018s, AV bypass and the potential for an extra say GBP50 and a months extra design effort to add Async USB to 768Fs!! and DSD support etc.
The first 20 or so customers who have paid the Toy / Fusion upgraded MDACs can then send them back to me here in CZ once the upgrade PCB is available and Ill transplant the Precious Bulk Foil resistors onto the new MDAC PCB (the PCB will be design for both options) for the cost price of the new PCB which I'm guessing at around GBP120
This route has many advantages not only is it a vastly better product without the restrictions of trying to add the modifications to the existing PCB, you have the DUAL ESS9018s and the potential of DSD & 786Fs support with each unit tested and QCed by myself here in Europe more importantly we would be helping the local European economy, shifting production back from China.
Just an idea and early days, wondering if there would be interest as the current method of upgrading the PCB is unsustainable in the longterm while this would be a 4 month plan so far about 130 people have expressed interest in the upgrade at 3 to 4 days a unit (hopefully I can speed this up for the remain units here) Im never going to be able to upgrade all units! While it would add a delay for a few you do get a vastly superior product and the payments can be spread over 4 months
I would only need 50 commitments at GBP100 per month to go ahead and design a new replacement board incorporating the Toy / Fusion upgrades + all the juicy extras such as dual 9018s (and DSD / 768Fs support if the consensus is to spend the extra GBP50 and month extra hardware development) it would be a REFERENCE grade DAC (with all the lessons learnt from the current upgrades etc) any interest or comments?
Hi John
Following closely the thread lately, it seems you re indeed struggling very much with the upgrades - with excellent feedback so far, Congrats! - but you certainly would feel better spending your time on other topics/designs/developments.
Now, the approach you propose seems very much reasonable, and I feel like you will receive support from this community.
Do you envisage different versions of the upgrade (i.e. "basic": upgrade of the full PCB with dual ESS, improved USB and DSD, "Toy+": basic + Toy components, "Fusion+": basic + Fusion components)?
Could you please get a bit more precise (as far as feasible atm) regarding associated final costs?
And last point, how long would you estimate design work for the new PCB + validation on your side, before going production?
Yes indeed, the upgrades have proved much more difficult then I expected - units sound different and some units prove to be a REAL battle to bring them to the Toy / Fusion standard. I'm learning as I upgrade each unit, but then a unit can will throw a spanner into the works
The good news is that with each unit I learn a little more how to improve the MDAC nothing like get ones hands dirty!
If the consensus is to go with the USB2.0 design this will add an extra £50 to the unit cost to cover the month extra hardware development time (Dominik will be working on the software for many more months).
Standard MDAC2 USB 2.0 (786Fs & DSD support) = £450 + Postage.
Toy MDAC2 USB 2.0 (786Fs & DSD support) = £496.98 + Postage.
Fusion MDAC2 USB 2.0 (786Fs & DSD support) £786.24 + Postage
The above prices will be £50 less if we decide to go with the current 24/96KHz USB1.1 implementation.
2 months design time for the (standard 24/96KHz USB1.1 version) or
3 month design time for the USB2.0 DSD / 786Fs version
Then 2 months for testing and first production units
I might be able to reduce the design times... I've indicated what I'd like to believe is worst case...
There is little extra in the price for what would be a stunning new design!
John,
for the long term - count me in as well. I'm not sure whether to bring home my original MDAC and bring it back to you once the PCBs are available or whether to pay extra GBP120 with the benefit of having the unit ASAP with the current L2+fusion upgrade, I shall decide during the next week (assuming you didn't upgrade my unit so far).
Some points I have on my mind (you don't need to answer here, just ask yourself):
While enthusiasm is good, how real is the 4 month plan? Does it include time reserves for when problems show up? Consider "lessons learned" from design time of BDAC/MIMP/.. (not the product itself, just PCB design time). Won't you run out of money before it's done?
How real is the cost estimation for the PCB? Won't it get way more expensive? Are there multiple PCB manufacturers (competition) as a backup?
What about approach to all the customizations both hardware and software? Will you perform slight alterations to the PCB design to make ie. front 5-pin miniXLR PCB-mountable? Is it worth it?
Any QA process to test the PCBs prior to soldering components on them? I know Europe *might* be different, but there can still be a number of badly made PCBs and dicovering that too late might generate more smoke than intended.
What about expected sound quality? Are you certain enough that it's going to be better than original MDAC? Won't different components and a different PCB make a negative impact? Sound signature change?
What about the new features? Any list? Won't a dynamic list (adding new things on the fly) prolong the design time / add differences between PCBs?
Speaking of features - how far are you willing to go? Perhaps a new USB chip (XMOS-based) to eliminate linux/windows usb3 and windows 8 issues and allow 24/192 over USB?
I'm really not trying to scare you, just trying to provide some things to consider, so that the entire process doesn't end up badly for you (designer), us (customers) or both. You know, you have some reputation and history to protect. ... No pressure.