The four bold sentences sound really horrible.
Since I started interesting in this, I was at the university. Now, I am more than 10 years in the business for embedded devices and know some manufactures, consulting companies and suppliers.
For me, (just my feeling), it sounds like, that the project/product doesn't envolve during a decade. You have still the same issues as you have with the pandora.
I also never heard, that ED is not overworked and I don't want to know, how gray are his hairs now (I haven't seen a pictures of him for years).
Additionally, the interest in such a device decreased in the last years. Posts in this forum are decreasing. I am still wondering, that this project still exists and only got notified via mail that there are no real new posts.
1. Is affordable
The only ones fulfilling all the other requirements is probably Global Components and that would be stupidly expensive
2. Overworked.
Probably not. He just has enough to do with his main business that is profitable so that he can not spend a day per week to build Pyras.
The Pyra business is a loss and a labor of love.
So it makes sense for him to focus on what actually brings in the money.
3. As long as there are functioning PCB.
That is not an issue.
We simply only have a limited stock here and all the parts required are at Global Components.
There are some PCBs with issues but not the majority.
All I wanted to say is that I can not simply take hundreds of PCBs with me and finish all the Pyras at once.
There is plenty of stock left and we can order new boards any time.
4. As bad as during the last 10 years
Yes, the Pyra is a mess and you are far better getting a GPD Micro PC.
ARM is stupid and limits the software we can use.
The MicroPC also gets more battery life than the Pyra out of an Intel APU and can run any software just better.
The Pyra's design is old and the connectivity is bad. Micro USB and the likes ...
The software and OS are poorly evolving because caution and "making it all good" is preferred over including quick fixes that work for the time being.
The whole thing aims to be the perfect thing that does it all (like what the Pandora achieved) but fails at it.
I would have skipped many of the steps we do and just deliver a non-painted device which would not look as pretty but would save a ton of steps needed.
An other case of "making it all too perfect and fit it all" in my opinion.
The volume wheel is a mess and the whole audio system is convoluted.
It was great back then but is super outdated now because real analog volume controls is not possible.
An other case of "making it all too perfect and fit it all" in my opinion.
We have some stupid display rotation going on that complicates everything to the demise. XFCE or MATE can rotate the desktop just fine.
Some games might not support the rotation but sho cares.
An other case of "making it all too perfect and fit it all" in my opinion.
My fix for the "_" and ":" key are not pushed in the OS still as far as I know.
Seems like the approach I build on (we are using for more than 5 years now) is not good enough because it might not be the absolutely best and perfect way of doing it.
So better no fix than one that works and does not do any harm.
An other case of "making it all too perfect and fit it all" in my opinion.
The dual PCB design was great thinking ahead back then but is not practical for today.
Not only did we get USB-C now which forces to upgrade the mainboard and the CPU-board anyways.
Also the space between CPU-board and Main-board is so tight that not even all the availables SoCs fit.
Better ditch that, put everything on one PCB and invest in some better cololing.
Today one would use a capacitive screen.
Resistive was nice back then, however the stylus does not even hit the "x" at the top right corner (close application) reliably.
I would remove the stylus and add in a bigger battery instead.
LTE is a mess. I tried to debug this with HNS but got ignored at some point.
All in all, so many bad circumstances happened that delayed everything and delayed it all by so much.
I would say that things could have moved a lot more efficient without the "make it all perfect and satisfy everyone" mentality (it's a mess anyways).
However ED is financing the project at a loss and doing his best to deliver all the Pyras so that everyone who paid for it will get one.
Let's just wait for the next couple of months and see how efficient we can get with building the Pyras.
I already figured out a few tricks to speed up the mass-production quite significantly.
So once the black parts arrive, the production should run quite smoothly.