Update regarding the cases (and some other things)


Staff member
Mar 4, 2003
So - where are the cases?
Not here, as you probably guessed (I would've made pictures otherwise).

I'm pretty much enraged now - and I've reserved next Thursday / Friday for a short trip to Greece if need be!

I've talked with my contact today. Unfortunately, he was on a trip through Germany from Tuesday until yesterday (I knew that upfront), and apparently, Greece companies are similar to some Chinese companies... they work sloooow unless you pay them a daily visit.

Well, he's been there today and has seen some of the finished parts and talked to them.
The machine is working again, they've made some minimal tweaks in some molds so that the case parts fit together better (minimal tweaks means: +/- 0,1mm)... yeah, don't ask me why they didn't do it before... I have no clue either :(
The reason probably is that you can only do that during production. You produce some parts, measure them, tweak the molds, produce some more, etc.
So that can't be done while a different production is currently running.

Anyways, as I mentioned: If I don't receive the cases or see pictures of their existence this week, I'll pay them a visit.
So I reserved two days next week (Thursday and Friday) and will travel there - unless I get pictures or the cases before that (my contact person will be there on Monday, so we'll know more then).

Rest assured, I'll take care of that :)

Any other things that have happened this week?

Yes. We finally fixed the DPad (it was a bit mushy before, as Fahrstuhl also stated on the keymat-test-thread).
The fix was a pretty simple one:
The current DPads have a recession for the keymat pads of 1,9mm.
However, a DPad (when being pressed in a direction) moves skewed - which doesn't work well with such a recession.
Pandoras DPad has 0,95mm recessions, and putting that one on the keymat works fine.

That was the last thing to fix with the keymat... so we're really done now :D

PCB-wise, we have the issue that the boards with 4GB RAM freeze after a while whereas the 2GB RAM boards work stable.
This could be because the only CPU-Boards we use for testing are the unoptimized ones, it could be the 4GB RAM is not properly setup in software (timing) or it could be a signal integrity issue, which can be simulated (and I hope Juliano will help us again here).

zmatt finally found some more time to debug and test on the PCBs this weekend, so we should hopefully know soon if Nikolaus' handmade fixes work fine (which would mean we can finalize and order the mainboards for the mass production).

Yes, we're moving slowly, but we are moving forward.

And I know this is quite a long wait - and honestly: If you ever consider starting such a project... don't. It's really really stressful.
And unless you are as lucky as I am having such an awesome community like yourselves being so patient and following every little step... well, it could certainly drive you crazy.

I honestly do NOT want to be in the shoes of some Kickstarter / Indiegogo guys when backers get impatient with all the delays...

Once more, I can only say THANKS!
We'll get there.

And next week we'll either see cases or you'll see me flying to Greece :)


Very Active Member
Sep 29, 2014
I can't believe how frustrated you must be with the case people, its so close I bet you can almost taste it!

the dog ate my homework err i mean cases !


Nov 19, 2005
a message to the mould-company:
ehlo! we... need... cases... ...now !!!
please form-noaction transform into a really fast form-action company and deliver nice cases asap, so our dragon doesn't have to waste time and money visiting you again. ok?
thanks in advance.

ps. here is hope that the 4gb ram issue is just a software timing thingy.

in the meantime, drink a good cup of coffee/tee and meditate to relax a bit.
we *will* manage this last chapter too! :)


Still fresh, damnit!
Oct 6, 2008
Somewhere off the coast of the EU
Thanks for keeping in touch, ED! I can only imagine how frustrating placing a big order then not hearing anything from them could be. Pandora wait frustrations++

I too hope you can find a fix for the 4GB ram parts soon and without too much to and fro. That shouldn't slow down getting units out to prototype devs though, should it?

I hope these frustrations don't stop you evolving the pyra parts over the next few years, and even trying out some of the alternative uses for these parts we've been thinking of (like the pyramote). While you're selling a lot more interesting stuff on the shop these days, it's really because of these projects that I know about it.

Keep up the good work!


Advanced Member
Oct 25, 2011
I´m confident you know the priorities and how to press them.
Like if time is available because of a secondary priority, then it is fine to give the first one more time.
Problem is when each one delays the next one, the usual kind of stuff that only money resolves (and sometimes not even that).

It is a tough place to be but I´m sure it will turn out fine. Keep cool.



Hi, how was your day?
Oct 9, 2016
Venice, Florida
My god this all sounds so frustrating. :mad:
If I was you, I whold of gone mad by now!

Hopefully everything gets better.:)


Great news, as I see it, two open items:
  • Cases
  • 4GB RAM issue
Aside from actual production and assembly..am I missing something?


Forum Addict!
Sep 28, 2011
I still don't trust the case company to do proper cases unless I see them and hold a perfectly working one in my handy.


Advanced Member
Jan 15, 2004
The Netherlands
Visit site
At least these cases are tweaked a bit again. We all know it is hard to make stuff sometimes, maybe they are having their own unlucky period? (they just lost a shipment they say)
I once worked with a guy that did great work for a really low price but he wasn't predictable planning-wise, so when you planned him into a chain of events he would cause delays and frustration. But still would work with him as he does a really good job for a low price.


I promise to cut my personal CO2 emissions by 2060
Dec 14, 2012
so... you ARE collecting Airmiles? (they are handy for discounts and what not...)... or are you flying yourself?

2 more months... I'll drink to that... where is my bottle of ouzo.... Cheers!

Don't worry ED!


Hardcore Member
Jun 5, 2008
Hey, the keymats are finished! Great news. Good to hear we are making real progress.

The RAM issue is a concern. You say you only use unoptimized boards for testing. I'm not sure what you mean with optimized in this context.

I understand the frustration Michael, but keep in mind that deadlines are based on estimates and most will slip. Yes it is a long wait, yes it gets frustrating from time to time.
But this is a shared dream of the community that we all want to happen, which happened in the past and which will happen again. If the Pandora taught us one thing, then it is patience :)

I have full confidence that you will bring this to a good end with a quality product, I'm just worried that you topple over half the way. Be sure to take a nice holiday before it becomes too much, not after.


Fairly Idle Member
Feb 4, 2017
Great news, as I see it, two open items:
  • Cases
  • 4GB RAM issue
Aside from actual production and assembly..am I missing something?

Hmm... seems to me you're missing a substantial number of steps!

I'm very happy with progress so far but there's a ways to go still, I feel. Here's a partial list from my tally:
  1. Current batch of test cases yet to be completed.
  2. Current batch of test cases need to be tested.
  3. [Possible final iteration of test cases.]
  4. Final cases must be mass produced.
  5. [Form Action has proven themselves unreliable and possibly dishonest; in the worst case there will need to be a change of supplier. If that happens there will be a delay, especially is Form Action are difficult about giving up the moulds.]
  6. Multiplexer bypass solution still to be tested; if not effective, an alternative solution might need to be found.
  7. 4 MB issue needs to be identified and resolved.
  8. Testing of current prototype hardware which is ongoing must be completed.
  9. New prototype hardware including fixes (some already identified) needs to be made. And tested.
  10. [Possible final iteration of prototype hardware.]
  11. Final hardware must be mass produced.
  12. Prototype of thermal solution needs to be made and tested.
  13. [Possible iterations of thermal solution.]
  14. Thermal solution components must be mass produced.
  15. Video cables must be mass produced.
  16. Final tweaks to keymat need to be made and tested.
  17. Keymats need to be mass produced.
  18. Mass produced components must be assembled into Pyras.
  19. The installed GNU/Linux needs to be polished for a satisfactory user experience, possibly including bug fixes in drivers.
  20. Final OS image needs to be produced and tested.
  21. OS image needs to be flashed into assembled Pyras.
  22. The product boxes must be designed, ordered, manufactured, and delivered.
  23. Printed material to be included in the box (getting started booklet or at least a safety information pamphlet) must be written, typeset, printed.
  24. Those working on the above need to do their day jobs, and have summer holidays etc.
This feels to me like quite a long list still. Certainly more than two things, more than two months. Excellent progress has been made and continues to be made and we'll have a great machine in due course.
Last edited: