Could the base of the Pyra be used with a smartphone?


Mr_Loon

Can't Remember
Joined
Aug 30, 2010
Messages
2,332
Hi ED,

How easy would it be to use the base of the Pyra with a smartphone? i.e. What would have to be done to use the Pyra base as an iCP3 type device?

Just to clarify by 'base' I mean everything apart from the screen & CPU board.

If this is indeed possible I'd give it some serious thought as you could sell such a thing at a premium and hopefully thereby be able to reduce the retail price of Pyra.
 
Replace the CPU board with a (cheap) barebones microcontroller that handles the keyboard and game controller, talks to the bluetooth, and that's it?  Interesting idea.
 
Last edited by a moderator:
Call it the Pyratroller or Pytroller or PyCon instead?

One potential catch will be that in the end it was decided to split the symbol key pairs on the Fn layer meaning that non-Pyra use of the keymat would need an interpreter layer between the mat and the target bluetooth device as "US101 keyboard" won't work - nor would any other generic HID keyboard driver.  So - there would have to be some logic running in the unit itself to make that translation.  With that, though - and what Wizardstan said above, yeah, it could theoretically work.

However, I could also see very clear reasons why ED should avoid any project that looks or works even remotely like the device that was proposed for the ICP2.
 
 
I remember there was a Programm for the Pandora who turns it in a Bluetoothcontroller, so just port this to pyra, put the Smartphone whit some ductape on the Pyra, and this should work..

Every plan that involves duct-tape gets my thumbs up!
 
 
There is no reason not to have the symbol pairs intact on the lower FN layer in the default keymap. I would go so far as to say it would be stupid to not have them mapped because they are clearly labeled, too (in the row above). I think this actually is the smartest part of the layout, since it is easy for the user to completely remap one row and still have all symbols available with fitting labels.
 
Replace the CPU board with a (cheap) barebones microcontroller that handles the keyboard and game controller, talks to the bluetooth, and that's it?  Interesting idea.

And also gives access to the SD Card slots and charges the phone via the Pyra battery.
 
The idea is nothing new, I think I even stated that the iCP2 stuff could be put onto a CPU board to turn the base into an iCP2.
 
The idea is nothing new, I think I even stated that the iCP2 stuff could be put onto a CPU board to turn the base into an iCP2.

It's certainly not new  ;)  but IMHO it is a very good idea for a product, I know the whole business with the iCP2 left a very bad taste around here, that was not the fault of the product or the vision of the product.

The Pytroller (great name BTW) would be a perfect match for modern smartphones in many ways. Imagine a micro usb male connector attached to your phone connected to a USB hub on the Pyra side bringing : Charging from Pyra battery (optional) , KB, Gaming controls, SD Cards, USB Slot, with added Bluetooth for wireless operation. If the speakers on the Pyra are any good then perhaps an audio connection could be added....

I can understand the fear that some people might just think 'why buy a Pyra? when I can pair my Pytroller with my Nexus 5 etc.' However that's nothing to fear IMHO as Dragonbox are in charge of pricing, use the profit margin from sales of the Pytroller to subsidise the cost of the Pyra and therefore sell more Pyra! The development costs could be shared as well.

Bet a lot of people would buy a Pyra & a Pytroller, I know I would.

(Very) Hypothetical pricing model to maximise Pyra Sales :

Pyra : $450
Pytroller : $250
Pyra + Pytroller : $600
 
 
Last edited by a moderator:
But wouldnt it be cheaper, to just by the Pyra and use the Controller PND i mentioned some posts bevore ??
Im not that much in Smartphonegaming, and the games i play, dotnt need a Controller (Angry Birds, Solitär, World of Goo) or will also work on Pyra by using the Android Image..
So the Controller needet games i play, are every on Pandora/Pyra...
 
I think the idea is that a configuration that could avoid the expense of the screen and the complicated 10-layer CPU board (for a simpler 4-layer-or-less board) could be sold for less than the cost of a Pyra.

FWIW, I'm not sure abot the names suggested thus far; either pyra-troll or pyra-con seem to have unfortunate connotation to me.  Maybe do something around the name of ED's shop; the DragonTroll-er sounds plausible.
 
But wouldnt it be cheaper, to just by the Pyra and use the Controller PND i mentioned some posts bevore ??
Im not that much in Smartphonegaming, and the games i play, dotnt need a Controller (Angry Birds, Solitär, World of Goo) or will also work on Pyra by using the Android Image..
So the Controller needet games i play, are every on Pandora/Pyra...

Yes it would & it would be great to see a similar 'PND' for Pyra.

Many people here, on the boards, would have no need for a PYtroller, however it's my theory that to the global market the Pytroller would be a highly attractive product, thus sales of Pytroller + shared development cost could mean that the retail price of the Pyra could be lowered and thus the user base extended.

What I missed from the above example of hypothetical prices is that without the profit from Pytroller the retail cost of the Pyra might be $500.

NB : All the numbers for pricing are for illustrative purposes only.
 
The product idea for a Pytroller is good.  Pytroller works fine for a name until something better comes along - anything BUT Insane Clown Posse X, OK?


As for the concept of the Pytroller sales subsidizing the Pyra sales, that's silly.  EACH product should self support.  The Pyra can and should stand by it's own merits regardless of whether or not ED decides to create and sell the Pytroller. 


OK, so if we re-use as much as possible from the Pyra...


*Main board stays the same, BUT full sized USB ports and microHDMI ports are not populated or populated with something to fill the holes that will be left behind in the case plastics.


*SoC daughter board comes out - a new as cheap as possible one that can do the functionality goes in.  THIS is the part that may crush this from an economics view.  The development time and costs of creating a new SoC board for another SoC to run essentially the same equipment that is already being run by an already relatively inexpensive SoC is a bit of tail chasing.


*Screen comes out.  Lid stays to act as a dust cover for the keyboard OR an additional plastics mold is created without the screen support horns - not cheap to remove those horns.


*I don't see the Pytroller as being able to be sold for much less than 2/3 to 3/4 the cost of the Pyra.  I.e. $300+/-.  After all, we've only shaved a screen ($50) and an SoC ($40) totaling less than $100 off the goods bill and the labor to build it will be about the same.


So - the economics of a Pyra based handheld controller-only get kind of destroyed.   But, lets leave the whole economics off the table and talk about what the Pytroller could/should be able to do.


Once it gets around those economics...  Pytroller should be able to:


1.  Connect via bluetooth to any device as a HID keyboard.  Issue:  Since it was decided that we should break the standard Key+Shift symbol pairings and Fn is not a hardware switch within the keyboard controller, no device that it is connected to will be able to read keys beyond the alphabet without a platform specific driver.  To cure this, the Pyratroller will NEED to have a processor and logic to convert presses on the keyboard layout on the Pyra keymat to mimic the behavior of a 101 US keyboard before it sends the bluetooth communications to the connected host computer.  Essentially it will need to run it's own version of FunkeyMonkey behind the scenes.


2.  Connect via USB to any device as a HID keyboard.  Same keypress to output translations as described above will need to occur.  Ideally it would then include a USB hub to make the two USB host ports into USB hub ports for additional devices.  I.e. you could daisy chain Pytrollers one to the console/device then two to one out from there or connect a USB keyboard and mouse to the Pytroller/hub.


3.  Act as a portable battery to charge other devices.  With 6000mAh of battery, the Pytroller should be able to charge someone's phone on the go.  They should be able to game while charging.


In theory, though, the Pyra itself should be able to be set up to do all of the above - and more of course.


So - the Pytroller isn't going to be particularly smaller, lighter or more functional than a Pyra.  It will need cost 3/4 as much as a Pyra.  The Pyra can do everything that the Pytroller could do.


Hmmm...  in that scenario I don't see the Pytroller subsidizing the Pyra sales, but the other way around.  Maybe ED can make the economics work better than above though.
 
As for the concept of the Pytroller sales subsidizing the Pyra sales, that's silly.  EACH product should self support.  The Pyra can and should stand by it's own merits regardless of whether or not ED decides to create and sell the Pytroller.

Without actually knowing the costs involved with producing a Pyra or Pytroller it's very hard to say in reality if that concept is silly or not.

However, at the very least, some of the development costs could be shared between the two which could lead to the Pyra selling for less and therefore having a bigger user base.

On way I could really see a Pytroller being a successful product is if Microsoft ever made an x86 based 'surface' Windows phone, there have been rumours that this might be the case and looking at it logically it would be a very sensible move in terms of bringing something unique (ability to natively run x86 'apps') to the mobile phone market. Pairing a Pytroller to such a device would surely appeal to a significant subset of those who would want an x86 based Windows phone. Look what people are willing to pay for a type cover for existing surface products look at the extras compared to a type cover that a Pytroller would bring....



 
 
But a pyratroller would need everything needed to make it act as a bluetooth/usb controller working ahead of time.  That may work out of the box with the Pyra, but it may not, and I don't see the point in delaying the Pyra for something of which the market is unknown.


Similarly, nobody's going to accept a joystick that takes ~30s to boot up, so somebody's going to need to come up with a spin of some linux that cuts out all the cruft.


A TV-out only version might make a little more sense, as you wouldn't even have to remove the HDMI port for that, and a 30s boot up time is more acceptable for that class of device IMO.   But ED would know how many people bought TV-out only Pandoras, so have some idea on how big of a market that is, and whether it's worth supporting with a different production line or not, even though I suspect a device with native HDMI out is a lot more interesting than one which required a slightly flakey cable to use.
 
But a pyratroller would need everything needed to make it act as a bluetooth/usb controller working ahead of time.

A dedicated microcontroller would suffice, doubt it would take long to boot up.
 

I don't see the point in delaying the Pyra for something of which the market is unknown.

Agreed, the primary priority should be Pyra, however if the Pytroller could be produced without causing delay to the launch of the Pyra or perhaps be produced a year or two after the launch of the Pyra then it's worth exploring IMHO.
 
If it used a 'dedicated microcontroller' then presumably that wouldn't know how to interface with the peripheral board, so a new CPU board and a new peripheral board would need to be designed.  I can't see such a project being run at the same time as the Pyra development without slowing that up.


The cheapest way to convert a Pyra into a bluetooth joystick it seems to me is to build some peripheral boards without the SD card slots and most of the ports installed, design derivative cases that don't have those holes in them (and while you're doing that you could probably get rid of the hinge lumps), and design a cut down Linux that doesn't start the wifi card or X11 amongst other things.  Depending on how much that ends up costing, it may be worth designing a new CPU board with an A9 chip on or something, but you'll probably need to sell quite a lot of units to offset the development costs of a new CPU board.


I do worry that it might not be possible to make this joystick device cheap enough for it to be a success though, using the boards designed for the Pyra.  Maybe having eliminated the SD card slots, that extra space could be used to build a low power CPU into that board, meaning you only end up building one relatively cheap 4-layer board (or however many layers the peripheral board currently has, I forget).
 
Back
Top