Ask / Submit

Make I2C accessible in some way (docking or solder pad on board) [answered]

asked 2014-12-03 13:24:04 +0300

dac gravatar image

updated 2015-01-16 14:33:09 +0300

eric gravatar image

As the Atom Z3700 does have I2C/SPI interfaces, please try to make this available/accessible in some way.

  • Either as Docking/TOH, as asked in this request

  • or, if that is not possible, at least try to route the necessary pins of the SoC to some accessible pins/pads on the board, in case someone wants to heavily modify the tablet and doesn't mind opening it.

I know that in the second case, the modification will be limited to more experienced/skilled/risktaking people, but at least it opens up "some" additional hackability... :-)

Edit: As Philippe mentioned, it might be anywhere between small effort to a lot of effort and/or cost, but it depends on the current board and the extent of the modification that is needed. So maybe it is possible, maybe too much effort. Let's see...

edit retag flag offensive reopen delete

The question has been closed for the following reason "the question is answered, an answer was accepted" by eric
close date 2015-01-16 14:33:39.632726

3 Answers

Sort by » oldest newest most voted

answered 2014-12-11 15:01:42 +0300

Philippe De Swert gravatar image

updated 2014-12-11 15:54:16 +0300

Actually there are quite a lot of pins coming from the chip. So adding, a connection/pin is not always possible. In some cases when working with an existing device that could mean rerouting/replacing/total pcb redesign. It is actually not very trivial to do.

EDIT: Sure we have those things in mind, but the cost/time factor is something we need to take into account also. Unfortunately we do not have the amounts of money to do real custom stuff like our bigger competitors. So if realistically possible we will sure do it, but atm it is totally unclear and most likely not do-able

edit flag offensive delete publish link more


Well, so we have to hope we are lucky in the way that routing them is possible without too much effort. Of course I can't say how much effort it is, only Jolla can make a statement on that, as they are the only ones that have access to the necessary infos and layout. I guess that in principle, routing the pin out is possible, else it would not make sense for Intel to offer it. Of course a complete pcb redesign would not be trivial to do, and maybe even too costly.

Just I think that if it was possible without too much effort, it would be a nice thing to do. Because pins that are not routed at least to somewhere on the board are dead and will never be usable in any way. Think "fm radio issue" on Jolla phone. While, even if there would be no way to have an official docking port with those pins, having them routed just anywhere where they can be accessed by someone skilled in soldering and electronics opens up some possibilities to tinker around with it, given one is ready to take the risk.

dac ( 2014-12-11 15:15:25 +0300 )edit

So what about a simple separate chip providing a useful interface (I2C/USB/GPIO/SDIO/RS232/whatever, which is itself connected to one of the "internal" available interfaces (could be even over internal USB).

MartinK ( 2014-12-11 15:15:53 +0300 )edit

@dac: Sometimes there is not enough routing space to exploit all of a chips functions. As they are sometimes designed with many features and it is up to the user to select the ones they really need and want. Which also means that often not all pins can actually be routed.
@MartinK: Such chip will still need routing and pins etc... it is possibly even more complex than exposing the pins.

Philippe De Swert ( 2014-12-11 15:57:05 +0300 )edit

@Philippe De Swert Understood. So let's just see how it is in this case. If it is impossible, well, there is nothing we can do about it. If it is too much effort or cost, sure, it would be a waste to spend much money or huge effort just for having the pads. But if it can be done with less effort, why not? I can't say. I'll be happy for a statement of Jolla on whether it is possible/feasible or not (in this specific case)... :-)

Edit: I appreciate the answer. So have a look on what is possible with reasonable effort using and possibly modifying the board you are going to use. Would be cool to have this, because it would open quite some possibilities in my opinion...

dac ( 2014-12-11 16:02:57 +0300 )edit

@Philippe De Swert: Sure, not all - what I mean is that actually about any of these interface that can be routed to suitable outside pins will do! :)

It might not be high speed, full featured or might even some quite obscure type of interface but as long as it is here, it can be used for doing interesting things. It would be also another first - after the Jolla being the first widely sold mobile phone with a generic third party expansion interface, the Jolla tablet can be the first such extensible tablet. :)

BTW, I think (and anyone more knowledgeable please correct me! :) ) that this is what a suitable interface should provide (if possible):

  • interrupt support - the accessory can notify the main device, so that the main device does not need to continuously poll the accessory for eventd. The current TOH interface has an interrupt pin and this alone makes the TOH keyboard project practical.
  • power out - the interface can power the accessory so that the accessory does not need its own power source
  • power in (optional) - the accessory can supply power to the main device (battery packs, wireless charging, solar panels, etc.)

IF these "conditions" can be met with one of the available interfaces, then fully featured accessories/TOHS/docks, etc., can be made for the tablet.

But even if some or even all of these "conditions" can't be fulfilled, it would still be worthwhile to include the interface as ti would still be useful and because it is easier to supplement the missing functionality given the tablet form factor (include a battery, an on/off switch, etc.).

Anyway, thanks a lot for joining the debate from Jollas side, much appreciated! :)

MartinK ( 2014-12-12 01:52:12 +0300 )edit

answered 2014-12-11 15:24:15 +0300

Macilaci457 gravatar image

I was investigating available devices with I2C and as much as I was excited about it, I have to tell most devices that have real world use is already onboard, making the port unnecessary. I can not name more than 1 really wanted featues added with an I2C TOH for the original Jolla, and that is the keyboard.

edit flag offensive delete publish link more



Well, I was not saying they should spend hundreds of kEuros for it. But as the interface is there, it would be nice to have it accessible, if it is not too much effort.

I know that the keyboard is the only project that got significant attention, and which you can buy now or soon, but there are other people who tinker around and try out things...

see e.g. Kimmo Lindholm with eink display, oled display, FM radio

If it is realized as solder pads, it would mostly be for hackers or people who have some skills and interest in modifying things and "making". It is not meant to be for people who just want a tablet for doing normal stuff. A bit more "unlike"...

dac ( 2014-12-11 15:50:41 +0300 )edit

True. I could only think of a second one: But it'd be vastly less popular than an external keyboard...

IZ1IVA ( 2014-12-11 15:52:26 +0300 )edit

And talking "unnecessary" features, I think everyone can name a lot of features in phones, tablets or any other products that are, for part or sometimes even most of the audience, unnecessary. Talk about (all for phones now): 3d cameras, cameras with 42mpix, 3d display, fullhd, widi, miracast, usb2go, nfc, fm radio, fm transmitter, voice control, microhdmi, millions of apps in app store, root access, smartcard reader, exchangeable battery, expandable memory... etc etc... There are always some people claiming that this or that feature is unnecessary for them. And they are right, subjectively, because they don't use it. I guess many of these features are not needed or extensively used by a lot of people owning the devices. But still, this is about potential. And I2C does deliver some possibilities. Not for everyone, agreed. The same that 3D printers are not useful for all, or arduino boards, or raspberry pi... or gpio on those...

The Jolla phone didn't have the I2C for keyboard only. The same that the raspberry pi had the gpio not for a specific application. But it delivered a "general" potential and possibility to add it. And something else can be added too, if enough people join up to do it. What that can be is up to the users. And that is one aspect that I like. Look at all those capes, shields and hats available for those boards that you can buy. There it just took the critical mass. Jolla is just not there yet, but who knows about the future...

dac ( 2014-12-11 16:42:04 +0300 )edit

answered 2015-01-16 14:33:19 +0300

eric gravatar image

Jolla Roadmapping Meeting - Friday 16.01.2015

Thank you for your contribution to the Jolla Tablet project!

We receive a lot of ideas and suggestions, which we are grateful for, but all of them can't be implemented. We've reviewed your idea and unfortunately it is not something we can implement, at least right now.

However, we will keep all suggestions, and might revisit this decision in the future if we see an opportunity to implement this feature-request in an upcoming product/SW release.

Best regards,
The Jolla Tablet Team

edit flag offensive delete publish link more

Question tools



Asked: 2014-12-03 13:24:04 +0300

Seen: 649 times

Last updated: Jan 16 '15