An image, an exhibition and 40 pages of awesomeness


EvilDragon

Administrator
Staff member
Joined
Mar 4, 2003
Messages
29,986
Age
47
Location
Ingolstadt
What have you been doing lately?

1. Schematics: 40 Pages

Well - Nikolaus surely won't forget... he created page after page of Pyra schematics.

A long and exhausting work, as it needs very high concentration.

But now it's almost finished! He already sent me a preview version which consists of 40 (!) pages just for the schematics!

Next step will be to move those schematics into the PCB layout. This is partly assisted, but still some work to do.

Well, most of that work will be the CPU PCB, as the main PCB was pretty much finished for quite a while already.

Some OMAP5 CPUs (and other parts) have already been ordered, so we should be able to create (hopefully :) ) working prototypes soon (whatever soon means :) )

What's neat: The Pyra will be compatible with the devboard. That means: We can use the SD Card from our devboard and put it into the Pyra SD-Card slot - it should boot up and work. Really helpful with development :)

However, it also means that in theory you should be able to use TIs Android Demo images and run them on the Pyra. Or run the polished PyraOS on a devboard at home.

Speaking of the OS...:

2. First Debian rootfs created on our build server!

With a lot of help from aTc, I finished setting up the build server for the OS.

We can now automatically build images with preconfigured configs and packages - a quick test today was successful.

The image already includes our own repository and our own packages.

So far, fbturbo (for a blazing fast X.Org) has been included as well as meta-packages (which we can use to decide what stuff we want to have preinstalled on the Pyra).

As Debian has almost anything we need in their repository, we can now easily create a fully fledged desktop OS.

We still need to create our own packages and include them, for example: our kernel, glshim, First Boot Wizard, SGX driver, Config files, etc.

We already created a package for fbturbo, so we know how to do it, it's just a bit of work :)

3. Electronica2014

Next week we're visiting the Electronica2014 in Munich, which is a huge electronics exhibition.

We'll visit our partners there and will discuss problems we still need to solve (i.e. getting the LCD with the rotator chip to work), visit the nubs manufacturer and talk about solutions, etc.

While they all help us via eMails as well, it's always different meeting them, being able to show what we already have and speaking with some of their experts.

We're getting there :)
 
Please wait a few days, I want to take a closer look at it as well and maybe include the kernel, so we all are working with the same kernel :)
Great!, I feel like I could be helping more. If it's just means testing out the packages and such.
 
Last edited by a moderator:
What's neat: The Pyra will be compatible with the devboard.

Or run the polished PyraOS on a devboard at home.
This is pretty interesting.  I am not sure how great it would be without everything else that will make the Pyra what it will be, but imo it allows those who may be interested in the Pyra, as well as those who want to make software, another option to try things out (and they can start early).

As Debian has almost anything we need in their repository, we can now easily create a fully fledged desktop OS.
Very cool.
 
Any thoughts what the first boot wizard will be made with? Not zenity, I assume? :p

Optimally I think it should input-wise be integrated well with the pyra controls as it's the first thing a new user sees.
 
Any thoughts what the first boot wizard will be made with? Not zenity, I assume? :p

Optimally I think it should input-wise be integrated well with the pyra controls as it's the first thing a new user sees.
Well, in theory, we could use the one we already have ;)

But if someone likes to code one without using Zenity, I'd be happy as well ;)
 
Well, in theory, we could use the one we already have ;)


But if someone likes to code one without using Zenity, I'd be happy as well ;)
Purely hypothetically, what tasks would the replacement need to be able to accomplish?


EDIT: aTc kindly pasted me a link to the current version. I think this could be doable with the doodad I did trying to make a GUI for that CLI game music player you once pointed me to.
 
Last edited by a moderator:
Is there really a reason to replace the zenity scripts? In a "if it ain't broke, don't fix it" sense? Is there something fundamentally broken about it that I'm not aware of because I've only ever used it, like, 5 times in as many years? Or is it just a "zenity looks bad and we can do better" mentality?
 
Well honestly:

  • zenity looks... if not bad at least boring
  • has confusing usability (like when doing nested menus with radio button dialogs)
  • has button labels that don't always seem to match the action being executed
  • is not tuned for the input hardware (d-pad to move between some elements and enter to confirm?)
  • has tiny UI elements compared to the screen estate. 
All in all it's good for one-off dialogs but making an entire wizard and configuration utility in it seems to be outside its comfort zone.
 
Last edited by a moderator:
Yeah, some of those are pretty good points worth fixing I guess. I've been in the business long enough to instinctively push back against "change for the sake of change" because it invariably replaces something that worked but may not have been quite as pretty with something that either doesn't work or works well enough but cost way more than originally anticipated. I have yet to encounter a project to replace a perfectly functional application which didn't result in mayhem.

edit: That being said, legitimate problems that need fixing are legitimate problems that should be fixed, and if that means replacing the whole thing then that is the cost to be paid.
 
Last edited by a moderator:
I've made it my personal mission to rid the base Pyra OS from any Zenity (or simlar crap) dialogs. It leaves a *really* bad impression on anyone who's an experienced linux user, as it really gives the impression of everything being just sloppily taped together.
 
Yeah, some of those are pretty good points worth fixing I guess. I've been in the business long enough to instinctively push back against "change for the sake of change" because it invariably replaces something that worked but may not have been quite as pretty with something that either doesn't work or works well enough but cost way more than originally anticipated. I have yet to encounter a project to replace a perfectly functional application which didn't result in mayhem.


edit: That being said, legitimate problems that need fixing are legitimate problems that should be fixed, and if that means replacing the whole thing then that is the cost to be paid.
I'm still not promising to do anything, but if I come up with something we can determine whether it's mayhem or an improvement :)
 
I've made it my personal mission to rid the base Pyra OS from any Zenity (or simlar crap) dialogs. It leaves a *really* bad impression on anyone who's an experienced linux user, as it really gives the impression of everything being just sloppily taped together.
well..imho scrapy stuff that works is infinitely better than shiny stuff that doesn't work...
 
Back
Top