Finishing up the hardware

[Eight Bit]

Haraldur, on 20 Jun 2015 - 03:30 AM, said: Out of curiosity: will any keys in the middle of the mat, like F and J, have any markings to help key-navigation by touch? Can be done, but first we need to finish the keymat design for that.

Edit: nvm

 

[Michoko]

Thanks for the update ED! I'm a bit concerned with the tearing issue, since I'm really anal regarding vsync. If a device has in-built tearing issues, it would be really a showstopper for me. Good luck with this challenge!

 

[Butterkeks]

Good news, there's hope to have the best Christmas ever in 2015

+1

 

[EvilDragon]

I suppose this also means that there's slightly more room on the action buttons for a label, since there's no circle padding anymore. Will the action button keycaps be printed on the underside as well? (that could be harder to do, since they are not flat pieces of plastic like the rest of the keyboard)

 Are all 6 action buttons like this, or just the 4 big ones?

The print will be on the silicon in this case and the caps are transparent.

Only the 4 big ones are planned like that.

 

[Caine]

410 traces are left to be created - so if he continues in the same pace, the CPU board will be finished in about 15 days.

Could be a bit faster as well though (some non-critical traces could be made using the auto-router), but we'll see.

Still the eternal optimist I see. Development being quicker than anticipated is a rarity.

Nikolaus also started to work on the e-SATA adaptor which plugs into our USB port for those who want to connect a SATA device to the Pyra.

Damn, another weird custom-adapter. I hoped that could be avoided. In my opinion, that's a wart on the design (just like the not-quite USB3 port).I take it that chances of a regular USB to esata adapter functioning properly were ruled out from the beginning.

Nevertheless, there is a surprising amount of progress this time. Exciting times.

Good luck with the keyboard threads

 

[WizardStan]

Damn, another weird custom-adapter. I hoped that could be avoided.

The alternative is not having eSATA.They were originally going to have a USB/eSATA hybrid connector but turns out they're actually very hard to source in quantity. It was either the adapter or ditch eSATA because there was no way to get yet another connector on there.

I don't see how you can consider it a wart.

Nor the "not-quite" USB3 port. Would you rather it just be a USB2 port such that when the CPU board eventually gets an upgrade to something with more USB3 support we're forced to USB2 anyway because of original design? Naw man, the "not-quite" USB3 port is a major feature, it's an upgrade path waiting to happen. May as well take advantage of the extra pins while it's there.

I take it that chances of a regular USB to esata adapter functioning properly were ruled out from the beginning.

Absolutely not. You want to use a standard USB to eSATA adapter go right ahead. You'll only get USB2 speeds instead of eSATA speeds but if that's what you want then go for it. This adapter is only to get access to the eSATA port.

 

[levi]

Apart from that, the TE line is simply a signal coming from the SSD. It triggers whenever it awaits a new frame. This can be read out in software and be hacked into the driver so synchronize the video stream it sends.

Ah neat.  What's the SSD by the way?  So it's a kind of reverse v-sync, originating from the display device rather from the sender device

Spoiler

or, depending on what the SSD is, it's kind of like a standard v-sync, since traditionally in software you're waiting for the sender hardware to signal the software that v-sync flyback has started before rendering the top of the next frame.

About the NUBs, will be improved this time?

They will be different, and I fully expect, significantly improved.  AIUI the nub part you move is a magnet and the base is a pair of hall-effect sensors, sensing the movement of that magnet.  Those signals are then digitised and passed to a kernel module, so it's entirely down to changeable software how the nubs are calibrated etc. - no more unexpected recalibrations, once we get the software working properly.

 

[WizardStan]

What's the SSD by the way?

The SSD is the rotator chip. From my understanding the TE line is like a standard vsync: the "display" reports back "ready for more"

 

[EdenR]

It s early for pre orders?

 

[levi]

No preorders until the hardware and software is ready.  So you'll need to wait until at least the demo units are appearing at shows.

OTOH you can donate to the project now, which would make the final units slightly cheaper for anyone buying one (but you wouldn't get close to getting your own money back in kind, unless you really value community appreciation, so it's not really anything like a preorder).

 

[Caine]

Would you rather it just be a USB2 port such that when the CPU board eventually gets an upgrade to something with more USB3 support we're forced to USB2 anyway because of original design? Naw man, the "not-quite" USB3 port is a major feature, it's an upgrade path waiting to happen. May as well take advantage of the extra pins while it's there.

Let me see if I understand you correctly.

• Our OMAP5 SoC supports USB2 and ESata, but not USB3.
• Regular USB3 to ESata converters will work but only at USB2 speeds.
• The custom esata adapter will overcome this limitation allowing us to utilize USB2 as well as esata.
• If we ever upgrade the CPU board to a newer SoC, then the USB3 port will become a real USB3 port.Thus regular USB3 to esata converters will function as normal and we no longer require the special esata adapter.
  

If this interpretation is correct, then I'm all for the chosen direction. Sounds like a good middle-road and certainly a future-proof design.

 

[slaeshjag]

^ Almost right. There's a USB3 channel on the OMAP5, but it is used for the OTG port. There's no way to use that USB3 channel as both a client and host port at the same time, thus it was chosen to use it for the OTG port that can do both.

 

[Caine]

Makes sense. Thanks for clarifying.

 

[rygD]

Great news! About the NUBs, will be improved this time? I have heard that the second production for Pandora still had a high failure rate.

It is a totally different type of nub. It should be more robust and not have the issues the Pandora nubs did. This has been discussed a lot elsewhere, especially when news that the nub ED picked would be discontinued (however there are a lot of options to still allow essentially the same thing to be used by making one of the parts, which is why it is/was discontinued...I think they said the sensor and everything is still around, just not the plastic top part where you put your thumb). ED made a thread (or threads) about it, so do a search if you want the details, and other things that were considered.

 

[Linux-SWAT]

What about the audio chip ?

OP opamp or OMAP ?

 

[asimov-solensan]

Great news! So, we will see a working prototype by august, I was wondering when this would happen.

Right now I'm most worried about the action buttons. Never seen seen such solution in any gamepad/handheld. I just hope they feel right.

Thanks for keeping us up to date ED.

 

[levi]

What about the audio chip ?

OP opamp or OMAP ?

Good point.  We were told there'd be audio tests done on those possibilities, which I assume still need to be done.  CPU board needs to be finished first I'd have thought though.

 

[jp10558]

So will pre-orders only be via Dragon box? I ask because I got a Pyra credit with Ithic in the US from the ICP2 failure, and would like to use that, but otoh, I don't want to end up many months down the road waiting because I couldn't pre-order...

 

[levi]

Actually IIRC ED has said he won't do preorders per se.  It'll only appear on Dragonbox shop once it's been produced.  He still reserves the right to do a kickstarter or similar though, which is closest to a preorder, but I'd be surpised if you could use ithic credit on that somehow.  There really shouldn't be any signficant delay between kickstarted orders going out and units appearing on Link's site though.

 

[FaeMinx]

ED told me I could use my alucase credit towards pre ordering. I also have ithic iCP2 credit, but I didn't even ask about that.


Once I've pre ordered, I'll ask ED about it...


Perhaps if enough of us got together, it might save both dragonbox and ithic some money if they were drop shipped?