An almost naked Pyra


EvilDragon

Administrator
Staff member
Joined
Mar 4, 2003
Messages
29,986
Age
46
Location
Ingolstadt
Sorry again for the long time without an update - once again, I'm stuffed with work... so running out of time.

This post will be a quick summary, but at least I'll have some nice pictures for you :)

Let's start with the status of the case...

The first parts of the case arrived last week (the full basement, so the most complex part).
And even though the last details to make it fit perfectly have not been implemented yet, it fits almost perfectly and looks and feels great!

The PCB fits perfectly, all ports and openings work without any problems as well, the keymat also fits fine.
The screws fit so much better than in the Pandora case as well (even though this is not yet the real material).

The shoulder buttons and the top of the case have been finished end of last week, so they will be shipped soon as well (I'll probably get them next week), the DPad and stylus should follow one week later.

So yes, we're getting there!

Okay, now we'll continue with the keymat...

Now that I got the real case, I could finally start testing and tweaking the keymat as well.
Basically, it is working, but it needs some improvement:

  • The ABXY-Buttons need to be a bit higher. Right now, they don't stick out too much and when you press them, they are flat with the surrounding. This makes it hard to navigate from one button to the next one. I've checked different buttons (PlayStation and SNES controllers, Pandora, etc.), and they all stick out more than what we've got right here. This is an easy fix!
  • While the flat keyboard buttons look awesome, there certainly is a reason why all mobile keyboards I've seen so far are curved a bit. While typing works okay, accidental button presses easily happen, as the space between the keys is narrow. The mobile keypads solve that with a bit of an elevation in the middle and a recession at the edges of the key: Your finger automatically finds the absolute middle of the key, and as the area surrounding it is a bit recessed, you also don't easily hit that key. This also is an easy fix, I just need to do some more testing.
So not perfect yet, but we're getting there :)


And finally, the status of the PCBs...

As Nikolaus solved the issue with the I/O-Errors of the SD Card slot, he is now able to boot into Linux to do some testing.
Finally, we could test the remaining parts: The Wifi module (incl. Bluetooth) works fine as well, the LCD including the rotator work, the 3G/4G-Module works, etc.

However, there is one thing Nikolaus is fighting with right now:
The system runs a bit unstable if it uses too much power. Running two cores at the same time will cause the system to freeze at some time, running only one core will run mostly stable (except for if you stress the CPU).

This is another one of those issues where the solution will probably be simple, but finding it is a long and hard way.

He already traced it down to most likely be the charger / power management chip setup we're using.
It could be a software issue (both ICs are being configured by software during bootup), it could be a simple hardware thing where only a capacitor needs to be added.

This seems to be the last hurdle we need to take, as all the other hardware has been tested and is working already :)

Now, let's guess:
Will Nikolaus solve the issue or will all the case parts arrive here first?
My guess is that both will probably happen around the same time, which would make me a happy camper.

As usual, if you have any more questions, let me know.

But now, enjoy the pictures I took of the case with a PCB put in!
 

Attachments

  • DSC02882.JPG
    DSC02882.JPG
    1.1 MB · Views: 2,375
  • DSC02883.JPG
    DSC02883.JPG
    5.4 MB · Views: 2,395
  • DSC02884.JPG
    DSC02884.JPG
    806.3 KB · Views: 2,259
  • DSC02885.JPG
    DSC02885.JPG
    1.1 MB · Views: 3,031
  • DSC02886.JPG
    DSC02886.JPG
    5.9 MB · Views: 2,378
  • DSC02887.JPG
    DSC02887.JPG
    1.2 MB · Views: 2,347
  • DSC02881.JPG
    DSC02881.JPG
    1.2 MB · Views: 2,875
  • DSC02888.JPG
    DSC02888.JPG
    1.1 MB · Views: 2,247
  • DSC02880.JPG
    DSC02880.JPG
    5.7 MB · Views: 4,391
Starting to shape up nicely. I figured there would be some bumps in the road.
 
Great newspost and beautiful pictures, thank you for keeping us updated!
These problems with the CPU made me think about possible hardware problems with the GPU - I know there's no real driver for it yet, but have you guys tested it at least with the hacked-up driver that was once shown to be somewhat usable with Reicast? It would be a shame if someone wrote a working driver just to find out there's a hardware problem with the first batch.

Edit: Oh, and what about light reflections on the keymat? That was once a minor selling point of the Pyra as an improvement to the Pandora because of the flat keys, and eventhough it shouldn't be as ugly with the transparent tops of the keys and is just about the looks of the device, it would be interesting to see pictures of it when you get the new keymat and/or reading a short statement by you about that.
 
Last edited:
The flat keyboard certainly looks awesome! I hope that having to curve the keys a bit doesn't change it too much, though I understand why its necessary. Thanks for all the hard work!
 
That clear plastic case looks really nice. I've been getting really excited about the Pyra. I love my Pandora, but this seems to fix every minor hardware and aesthetic issue that I had with the device. Happy to see it coming together so well!
 
The PCB fits perfectly, all ports and openings work without any problems as well, the keymat also fits fine.
I am probably remembering it wrong, but weren't there supposed to be some dip switches reachable via the battery compartment ?
The screws fit so much better than in the Pandora case as well (even though this is not yet the real material).
No personal "quest", I am just curious: One thing that came up rather often was the horrible choice to let the screws cut themselves into the plastic. Was this addressed and if so, how (guessing from the pictures, metal inlays weren't an option) ?

While typing works okay, accidental button presses easily happen, as the space between the keys is narrow.
Are the spaces between the keys smaller than on the Pandora ? It sad that the keys need a convex top now > especially as I remember you playing around with the idea of a concave top.
 
Looks good with the transparent case and the new keyboard really is a vast improvement also when it comes to the looks. Good work.
 
I want one so badly.

Hey not that I'm complaining but when can we see a pyra with the screen?
Acording to the newspost, mybe next week as the top casepart and the shoulderbuttons will shipped then to evildragon..
Wouldnt it be possible to just use a printed dpad and a Pandora Stylus for testing purposses?? As soon you have the upper part of the Case and the Shoulderbuttons??
They should also work, and you can build an working prototyph until the final case parts arive..
 
I am probably remembering it wrong, but weren't there supposed to be some dip switches reachable via the battery compartment ?

No personal "quest", I am just curious: One thing that came up rather often was the horrible choice to let the screws cut themselves into the plastic. Was this addressed and if so, how (guessing from the pictures, metal inlays weren't an option) ?


Are the spaces between the keys smaller than on the Pandora ? It sad that the keys need a convex top now > especially as I remember you playing around with the idea of a concave top.

Hey pal, here the current keyboard if It didn't changes since the video.


Let's start with the status of the case...

Okay, now we'll continue with the keymat...

Now that I got the real case, I could finally start testing and tweaking the keymat as well.
Basically, it is working, but it needs some improvement:

  • The ABXY-Buttons need to be a bit higher. Right now, they don't stick out too much and when you press them, they are flat with the surrounding. This makes it hard to navigate from one button to the next one. I've checked different buttons (PlayStation and SNES controllers, Pandora, etc.), and they all stick out more than what we've got right here. This is an easy fix!
  • While the flat keyboard buttons look awesome, there certainly is a reason why all mobile keyboards I've seen so far are curved a bit. While typing works okay, accidental button presses easily happen, as the space between the keys is narrow. The mobile keypads solve that with a bit of an elevation in the middle and a recession at the edges of the key: Your finger automatically finds the absolute middle of the key, and as the area surrounding it is a bit recessed, you also don't easily hit that key. This also is an easy fix, I just need to do some more testing.

And finally, the status of the PCBs...
Finally, we could test the remaining parts: The Wifi module (incl. Bluetooth) works fine as well, the LCD including the rotator work, the 3G/4G-Module works, etc.

This seems to be the last hurdle we need to take, as all the other hardware has been tested and is working already :)

As usual, if you have any more questions, let me know.

But now, enjoy the pictures I took of the case with a PCB put in!

About the keyboard and the ABXY button, you just need to fix the plastic cap, right? What about the backlight?

About the last hurdle, it still remains the GPU driver, how confident are you about it?


Great work ED!
 
Last edited:
On the first picture the headphone port looks buried deep inside the case. Have you tested it? This might cause the connector to not fit properly.
Also I'm curious how the keyboard backlight looks with the transparent case. I know, we'll most likely not get a transparent case :( but I'm curious anyway.
 
Awesome pictures ED! Getting really hyped for this thing.
When you get the upper part of the case for the screen can you sure these part
ICnqGGr.png

aren't to sharp. On the Pandora they where really sharp to the point that they where almost a hazort lol. Maybe this have already been addressed but I thought I would bring it up just in case.
 
I am probably remembering it wrong, but weren't there supposed to be some dip switches reachable via the battery compartment ?

You remember correctly, however, the dip switches have been replaced with electronical switches, so you can even change between the eMMC and the MicroSD on-the-fly.

No personal "quest", I am just curious: One thing that came up rather often was the horrible choice to let the screws cut themselves into the plastic. Was this addressed and if so, how (guessing from the pictures, metal inlays weren't an option) ?

That was an issue with the soft plastic the Pandora had.
Even the current case is hard enough that it works a lot better, but with Polycarbonate, the plastic is that much resistant, that once a screw has carved it's way, it will reuse the path each time you rescrew it.

Are the spaces between the keys smaller than on the Pandora ? It sad that the keys need a convex top now > especially as I remember you playing around with the idea of a concave top.

No, the spaces are the same, but if you look at the Pandora keyboard, you can see that the edges of the keys are almost flat with the case, therefore the space feels wider.

Acording to the newspost, mybe next week as the top casepart and the shoulderbuttons will shipped then to evildragon..
Wouldnt it be possible to just use a printed dpad and a Pandora Stylus for testing purposses?? As soon you have the upper part of the Case and the Shoulderbuttons??
They should also work, and you can build an working prototyph until the final case parts arive..

A printed DPad can't be used for testings, as it's not accurate enough to 100% test the feel of it.
We want a good DPad, after all :)

Perhaps the CPU is overheating when it becomes busy. Perhaps you need something like pyrolytic graphite sheets to distribute the heat more evenly.
http://www.digikey.com/en/product-highlight/p/panasonic/pyrolytic-graphite-sheets

While we've already got such sheets for testing, it's not overheating.
You can run it stable with 60°C with only one core enabled, but you can easily crash it when it only has 40°C with stressing the CPU (it takes a while until it heats up when you stress).

Also, I've run the OMAP5 devboard with a stresstest for 20 minutes without any heatsink and 120°C :D
 
Back
Top