Close
0%
0%

PiBoy-Zero

yet another portable console! ;-)
roughly the size of a gameboy micro, with a 2.2" display

Similar projects worth following
the idea is to build upon the great PiGrrl - a gameboy-like portable console that is raspberry-pi driven - but to use a single carrier pcb to reduce the size to a gameboy-micro like form-factor.

this project will build upon the open-source (thnx! :-D) plans for different adafruit products and their "learning"-resources to add the different features - and ultimately build a truly compact and portable console

some of the adafruit projects this is inspired by and will be based upon:
the original piggrl: https://learn.adafruit.com/pigrrl-raspberry-pi-gameboy
(its ancestor: https://learn.adafruit.com/game-grrl )
the pocket variant: https://learn.adafruit.com/pocket-pigrrl
and its second iteration: https://learn.adafruit.com/pigrrl-2


Design Goals:

- reduce ratsnest of wires, done with a custom pcb

- use good-quality buttons with better "feel" instead of cheaper/harder 6mm tactile switches

- hand-solderable, using 0805 sized smd components and slightly longer pads on the pcb


Features:

- 2.2" 320x240 display (for comparison, the gameboy-micro had a 2"240×160 display)

- 3d printable case, with overall size: 48x103x21mm (gameboy-micro: 50x101x17.2 mm)

- power button, for controlled shutdown of the pi (via a pushbutton - push for on/off, hold to force off)

- audio-out (via pwm, like the raspberry 2)

- 1200mAh LiPo battery connected to an adafruit powerboost 500c

- USB-A socket (reusing the connector that comes with the powerboost pcb, connected to raspberry-zero testpads for USB)

- OS: raspbian running emulation-station (and some support daemons to tie in the hardware)


x-zip-compressed - 1.08 kB - 03/14/2016 at 14:16

Download

Adobe Portable Document Format - 163.38 kB - 03/14/2016 at 14:16

Preview Download

Adobe Portable Document Format - 256.51 kB - 04/02/2016 at 14:20

Preview Download

View all 3 files

  • current state & ABXY buttons

    JohSchneider11/02/2016 at 18:14 0 comments

    been using the Pi-Boy on and off for the last couple of months - works beautifully :-)

    some things need to be redone though;

    - messed up front panel (heat-setting M3-thread *after* painting was a dumb idea...)

    - the power button needs a redesign (button-cap is missing)

    - have to add a bit material to the back/front to close the slight gap between the two


    on a side note: what turned out really well where the ABXY buttons - the combination of the soft-tactile omron switches plus a coat of epoxy feels just right while playing :-D

    the way i build them was:

    - print the button-piece (one part, all four are connected by a bridge - to ease assembly) with the letters indented

    - fill the letters with acrylic paint

    - put a drop of XTC-3D (some other epoxy might work too) on top to form a slight dome

  • milestone: it lives!

    JohSchneider04/09/2016 at 15:27 6 comments

    for the thirtieth log something special - it lives! :-D

    - battery: check

    - display: check

    - controls: check

    - sound: check

    all systems go:

    there are still many rough spots:

    - on the software side there are many daemons still missing (pwm back-light, power management, ...)

    - the system needs further configuration: the buttons currently only work in emulation-station menu, not the emulators (probably sth minor)

    - the case needs further refinement, lots of sanding, priming and a few layers of paint :-)

    - the setup/configuration steps need more documentation or better: scripted as far as possible (rough script already on github)

    - print and assembly instructions?

  • case prototyping

    JohSchneider04/09/2016 at 14:28 0 comments

    creating a 3d printable case from the rough sketches done earlier

    the overall shape was put toghether relatively quickly - but getting the d-pad to feel right took some iterations, the first few where still cross-shaped

    turned out, that shrinking and making it round works better for the alps switch. connecting the ABXY buttons eases installation

    also playing with the back part, for example splitting it in two with overlapping part along the seam:

    which would be printed standing up:

    this would have the upside of a sturdier frame - but on the downside prints waaay longer, with increased risk of print failure and the split/seam doesn't look as nice as a single-pice backside does, paintability would probably be another issue - so i'll stick with the single-piece for now

    btw: the front piece in its current form uses the pause-print-insert-M3-trapped-nuts-resume-print approach:



    STLs to print your own are on github :-)

  • and another minor mistake...

    JohSchneider04/02/2016 at 14:25 2 comments

    another wrong footprint...the push buttons used for start/select group their four pins in pairs - the other way round

    ... really need to learn to double check the datasheet for all components that have more than two leads - even trivial ones like buttons :-P


    oh well - the fix was to desolder the button, cut some traces and amputate a leg >:-)

    updated the pcb layout and pushed the changes to git and osh-park


    update 2016-04-13

    for anyone who happen to have this buggy version, the fix looks like this:

    cut the start/select traces at the red lines, and do not solder the button pads marked with the circle...

  • soldering the remaining components

    JohSchneider04/02/2016 at 14:14 0 comments

    mainly buttons and the usb socket...

    the usb-socket has mounting tabs which i decided to trim down at the side close to the alps-nav switch, just in case the room is needed later on by the 3d printed that'll come ontop of it

    the omron switches chosen for ABXY have some position keys on the back, which have to be trimmed also -- which was a design decision, including holes for those would have interfered with the smd-pads on the other side

    in retrospect it would have been a bit easier to mount the tft last, since it sits close to some solder-pads... but it was still doable

  • Pi Zero PWM Audio - device tree overlay

    JohSchneider04/02/2016 at 11:03 1 comment

    the adafruit guide for pwm-audio on the pi-zero ( https://learn.adafruit.com/adding-basic-audio-ouput-to-raspberry-pi-zero/pi-zero-pwm-audio ) configures the gpio-alt functions on the fly to remap the audio-pwm pins from hidden ones to pins on the gpio-pinheader
    after reading up on

    device tree configurations and tracking down all necessary pieces, i got this this code snippet/overlay to do the same at boot-time:

    /dts-v1/;
    /plugin/;
    
    / {
      compatible = "brcm,bcm2708";
    
      fragment@0 {
        target = <&gpio>;
        __overlay__ {
          pinctrl-names = "default";
          pinctrl-0 = <&pwm_audio_pins>;
    
        pwm_audio_pins: pwm_audio_pins {
    	brcm,pins = <13 18>;   /* gpio no ('BCM' number) */
    	brcm,function = <4 2>; /* 0:in, 1:out, 2: alt5, 3: alt4, 4: alt0, 5: alt1, 6: alt2, 7: alt3 */
    	brcm,pull = <0 0>;     /* 2:up 1:down 0:none */
          };
        };
      };
     };

    this maps the left channel to pin 33 (BCM=13) and the right channel to pin 12 (BCM=18)

    compiled with:

    dtc -@ -I dts -O dtb -o pwm-audio-pi-zero-overlay.dtb pwm-audio-pi-zero-overlay.dts
    and enabled by appending this line to the /boot/config.txt:
    dtoverlay=pwm-audio-pi-zero

    helpful links:

    http://elinux.org/RPi_BCM2835_GPIOs - pins and alt functions for the pi-zeros cpu

    https://learn.adafruit.com/introducing-the-raspberry-pi-zero/audio-outputs - the other adafruit guide, touching opi-zero pwm audio

    https://www.raspberrypi.org/documentation/configuration/device-tree.md - general documentation on devicetrees on raspberry

    https://www.raspberrypi.org/documentation/configuration/pin-configuration.md

    https://www.kernel.org/doc/Documentation/devicetree/bindings/pinctrl/brcm,bcm2835-gpio.txt - documentation on the gpio pinconfig via dtoverlay

    https://github.com/fivdi/onoff/wiki/Enabling-Pullup-and-Pulldown-Resistors-on-The-Raspberry-Pi - good example of gpio config via dt-overlay

  • solder&test MK2

    JohSchneider04/01/2016 at 20:15 0 comments

    ok, again repopulating the board with the components grouped by function (the raspberry-zero and the powerboost 500 are already mounted)

    first: soft-power

    connecting the lipo battery and hitting the power button the pi boots, and pulling the gpio down powers it of :-D


    second: audio

    NOTE: the components in the top right corner are left/right channel = two times the same set, but their placement got a bit mixed up = is not symmetrical... oh well, just have to follow the silkscreen :-P

    Testing the audio with a pair of headphones: works! :-)

    (when nothing is playing - eg the channels are silent - there seems to be a tiny bit of high frequency noise coming from somewhere - and sd-io is also slightly audible... i wonder if that is just due to the anaolog/pwm nature of the system or bad board design/soldering :-S )


    third: backlight and tft display:

    the 2.2" display is held by a metal bracket, which in turn is glued with strips of double sided tape to the donor-pcb.

    a sharp knife makes short work of these:

    reapplied double-sided tape to the bracket and glued it to the pi-boy pcb - aligned to the right edge of the silkscreen square

    Note: are some contacts on the front-side of the pi-boy-pcb that need to be masked of with insulating tape befor glueing down the metal braket!

    transplanting the tft screen:

    * the ribbon cable of the screen was desoldered from the donor-pcb with the previously mentioned tea-candle method

    * pre-tinning the contacts on the pi-boy pcb

    * and finally re-soldering the 14 contact

    * double/tripple check the solder-joints/solder bridges since the contact pitch is ultra small :-P

    powering the hole thing up and we have picture! :-D

    (upside down, but thats easily fixed in the config)

  • hand soldering the raspberry to the baseboard

    JohSchneider04/01/2016 at 13:54 2 comments

    similar projects that mounted the raspberry onto another pcb seem to reflow it in some way or another, with a hot-air rework station or similar tools. the same is doable with a good fine tipped soldering iron.

    these steps seemed to produce good results:

    1) widen the mounting hole near the hdmi to 3mm, which will be used as position key (and later on by a screw holding the case together) - done by hand, very carefully :-)

    2) pre-tin all the surface mount pads on the base-board with a slight (!) amount of solder

    3) use a M3 screw to get an accurate position, clamp the boards together and solder by first heating up the rim of the pin-hole, then feeding solder into it until the hole is roughly half filled. at this point it should connect with the pad below and visibly suck the solder a bit down

    the tip i used fit halfway through the pin-holes, which i could put into the soldered holes that seemed to have a hard time connecting to the pad below (was the case with some of the holes connected to the ground plane, since they sucked the heat away pretty fast)

  • candle-light desoldering

    JohSchneider03/31/2016 at 19:13 0 comments

    not wanting to discard the perfectly good smd-components that are already on the Mk1 pcb a candle-light desoldering session was in order: the Mk1 pcb below - roughly 1cm above the flame of a small tea light candle - and the Mk2 above ready to receive the transplant

    moving the candle under the specific components to desolder; small tweezers in one, the solder iron in the other hand - worked like a charm :-)

  • PCBs Mk2 arrived :-)

    JohSchneider03/31/2016 at 13:57 0 comments

    things of beauty :-D

    time to desolder the Mk1 and populate the Mk2...

View all 31 project logs

View all instructions

Enjoy this project?

Share      

Discussions

John Wells wrote 01/29/2017 at 18:36 point

I've assembled the power, pi, and buttons, but I'm having trouble desoldering the LCD ribbon. I tried the candle technique but it didn't seem to get hot enough and I ripped the traces off the ribbon. I will try again tomorrow getting the candle closer to the bottom of the board. If anyone has any other tips for this that would be helpful.

  Are you sure? yes | no

JohSchneider wrote 02/09/2017 at 18:59 point

the candle method takes some time and patience...

best use one of those third-hand-clamps to hold the pcb, point the candle at the PCB-backside roughly where the LCD-traces are (with the LCD folded away, so it doesn't get hot) and wait for the candle to heat the PCB and melt the solder

  Are you sure? yes | no

John Wells wrote 01/15/2017 at 21:01 point

I plan to build three of these in the next week or so. I will link to my build log afterwards. I have read the build log and looked at the git repo. I would also like to contribute to the project and have submitted an application. (Tbh idk if that is necessary since my contributions will likely be all software, so maybe GitHub would be a better place for me to be a contributor.) 

Is there anything not in the build log that I need to know before I order the parts and pcbs?

Also do you have any objection to my selling one of them to recoup my costs?

Also what do you think it would take to add a piezo speaker which shuts off when headphones are connected? Could it be done with a spare gpio and some software trickery to detect headphones? Maybe something like is commonly done for gpio capacitive sensing where you put a pin high and see how long it takes to drain capacitance when pulled low?

  Are you sure? yes | no

JohSchneider wrote 02/09/2017 at 20:30 point

looking forward to a build-log! :-D

and no - no objections if yo want to resell one

a speaker might be doable, but would probably need some amplifier - and the pcb is kinda full :-s

also i did exclude any form of speaker (like many other retropie/raspbery-gamboys have) on purpose... since i didn't want to annoy my peers with the retro-noise, and it just drains the battery and eats up space in the case-design :P

btw: just updated the git with software related stuff - if you have any changes/updates/suggestions or find bugs: give me a pull request :-)

  Are you sure? yes | no

Edu Arana wrote 01/15/2017 at 07:41 point

Beautiful and very well documented project! Congrats!

  Are you sure? yes | no

salmanisheikh wrote 01/10/2017 at 17:21 point

I built one back in April or May but have been so busy.  The plan is to build one 3 total and I have parts and most of 3 built up (except for the pizero and displays and powerboost on 2 of them). My colleagues at work are going to split the cost of the parts. But I don't know how to start trouble shooting the thing.  

  Are you sure? yes | no

Jeremy g. wrote 07/29/2016 at 23:57 point

where did you source your joypad? I know its alps but from what vendor did you get it?

  Are you sure? yes | no

slowill wrote 08/01/2016 at 15:18 point

I got mine from mouser.com.  Search for SKRHAAE010

  Are you sure? yes | no

slowill wrote 07/29/2016 at 04:12 point

Any tips for the D pad?  Printed out a few and having trouble getting a good fit.  Thinking about some kind of glue.   Thanks!

  Are you sure? yes | no

Jeremy g. wrote 07/14/2016 at 23:27 point

Is this stable enough to build as of yet? I finally got my hands on a bit of unobtainium.

  Are you sure? yes | no

JohSchneider wrote 07/25/2016 at 17:17 point

hehe - sure...

haven't found any bugs so far and it basically works (but haven't had much opportunity to use it either :-s)

  Are you sure? yes | no

Jeremy g. wrote 07/25/2016 at 17:24 point

No worries, I just like handheld consoles. And really have nothing to do with any of my Rasperry Pi's plus I'm fairly good with Linux and python so I'll let you know when and if I find any bugs.

  Are you sure? yes | no

slowill wrote 06/27/2016 at 01:33 point

Can you tell me what software you used for the 3d blowup models?  Thanks

  Are you sure? yes | no

JohSchneider wrote 07/01/2016 at 18:02 point

that was done with blender (cycles used for rendering) :-D

  Are you sure? yes | no

salmanisheikh wrote 05/24/2016 at 09:27 point

How do I bypass to power it up without a start switch and powerboost (not ordered yet)? 

  Are you sure? yes | no

JohSchneider wrote 05/25/2016 at 11:48 point

hm, if those parts are not soldered yet, powering the raspberry normally via its micro-usb should work (not recommended if you have the powerboost soldered, since power would backflow into it)

  Are you sure? yes | no

salmanisheikh wrote 06/02/2016 at 17:15 point

I got the Powerboost500C. It seems the black connector on the side that attaches to my LIPO is blocked by the USB connector on the Pi Grrl pcb. Is that connector supposed to be used or the microusb to power the board? My LIPO 2-bin battery connector will not fit in with the USB type A connector I have installed in there blocking the way. I will try to upload a photo to show what I mean before soldering in the PowerBoost500C. 

  Are you sure? yes | no

JohSchneider wrote 06/02/2016 at 18:00 point

regarding powerboost and blocked connector: yes,that tradeoff is intentional (see project logs) - simply solder the lipo directly to the battery through-hole pins on the powerboost

  Are you sure? yes | no

commandercopper wrote 05/17/2016 at 10:55 point

Really exciting project! Adafruit released the official PiGRRL Zero a frew days ago. But I'm looking forward to your project because the size is much smaller and compact, so hopefully I'm going to build yours soon!

Link to Adafruit:

https://learn.adafruit.com/pigrrl-zero/

  Are you sure? yes | no

JC wrote 05/20/2016 at 18:11 point

I'm in the same boat. I like this one way better but this one does seem to require more skill. This project even inspired me to finally go out and get a 3D Printer!

I'm wondering BTW if @JohSchneider you've done any further development on it and taken care of some of the "rough spots" you mentioned in your last update.

  Are you sure? yes | no

JohSchneider wrote 05/22/2016 at 19:36 point

yeah,saw the official piggrl zero too - was a little disappointed that they build custom pcbs but stuck with those horrible tactile switches, and designed the case bulkier than needed :P

mine sits on the edge of the desk and waits for a me to come back and continue working on it - the dreaded 95% done ;-)

recently i was playing with yocto to build a custom/minimal image/kernel, but that didn't went far... the only two major things not finished are on software side (soft-power needs some kernel changes, and pwm-backlight needs attention too - from a configuration side) - atm i consider the hardware side as done

  Are you sure? yes | no

salmanisheikh wrote 05/17/2016 at 00:53 point

Where is the kicad board file? I am trying to figure out the value of a three components above U3.  After that and the powerboost (yet to order) and I will have one complete.  I can power directly from a +5V supply in the meantime right? 

  Are you sure? yes | no

JohSchneider wrote 05/22/2016 at 19:22 point

you can find the kicad boardfiles in the linked git repo :) (under misc/ are even generated schematics/layout-files/... for each revision i pushed to oshpark)

  Are you sure? yes | no

JohSchneider wrote 04/25/2016 at 14:11 point

@salmanisheikh: yes - the registration keys on the omron switches need to be trimmed; see /project/9467-pigrrl-zero/log/35096

  Are you sure? yes | no

salmanisheikh wrote 04/25/2016 at 14:32 point

Cool, I was going to call them nipples but that sounds perverse. Registration keys sounds better. Yeah, I filed them. I realized I forgot to buy the directional joystick thingy with my $53 (3 sets) of parts order from mouser that I got in Friday. Arrgh. My colleague is printing the cases..one in glow in the dark PLA..3 of us are building them. BTW, We found a cheap $10 alternative to the display (its exact same as Adafruit but cheaper). 

  Are you sure? yes | no

JohSchneider wrote 04/28/2016 at 15:38 point

wow - take some pictures when you're done building :-D (mine still waits for another layer of paint...)

yeah, there are other compatible screens marketed at the arduino crowd with that exact ili9341 screen out on amazon/ebay - got one of those too :P

  Are you sure? yes | no

salmanisheikh wrote 04/11/2016 at 17:13 point

I ordered and got the pcbs of your previous revision from osh-park. Can you tell me what changes I need to make to the pcb and traces explicitly? I hope I didn't just waste $35 for 3 pcb boards :(

  Are you sure? yes | no

JohSchneider wrote 04/13/2016 at 13:54 point

damn... previous as in the 2016-03-13 version? then the fix is easy, i'll update the project log ( /project/9467-pigrrl-zero/log/35097) and add a picture, showing the fix

  Are you sure? yes | no

salmanisheikh wrote 04/13/2016 at 16:44 point

Saw the update..awesome. I have cut traces so looks like an easy fix. Thanks.

  Are you sure? yes | no

salmanisheikh wrote 04/24/2016 at 03:31 point

the Omron buttons have two little plastic protrusions on bottom that prevents them from sitting flat on pcb. Did  you sand them down or let button not flush/skewed?

  Are you sure? yes | no

David Perrenoud wrote 04/09/2016 at 15:00 point

We have a similar project on Hackaday. Would you be interested in some collaboration? I think both our projects could benefit from it. [@Jean-André Santoni] and I hang out at https://discord.gg/0y3hRkVsItoaCAR1.

  Are you sure? yes | no

arnaud.durand wrote 04/11/2016 at 20:51 point

@David Perrenoud: I just discovered your project; I'm very impressed that you use RGB666 to drive the screen. Bravo! I was planning to contribute to the PiGrrl-Zero with an alternative version featuring a Sharp LS035Y8DX04A display (3.5", IPS) as I was excited with the 60 FPS promise of http://blog.reasonablycorrect.com/raw-dpi-raspberry-pi/. I didn't know this interface was available on smaller displays.

  Are you sure? yes | no

JohSchneider wrote 04/13/2016 at 13:56 point

from my experience the 60fps are nice to have, but having less doesn't kill the fun in any way - having buttons with a good response/"feel" is IMO more important :P

@arnaud.durand: i havn't seen any screens that size that have the parallel display and are readily available so far either (another plus for the adafruit screen :-)

  Are you sure? yes | no

David Perrenoud wrote 05/07/2016 at 16:28 point

@JohSchneider Are you still interested? There are around 100 people on the Gamegirl Discord chat and I would be very glad if you could advise us on the PWM/PCM audio. We can help you with the buttons in exchange.

  Are you sure? yes | no

JohSchneider wrote 04/07/2016 at 19:04 point

@david: if i read the datasheat of the display correctly (https://cdn-shop.adafruit.com/datasheets/TM022HDH26_V1.0.pdf) only the serial/spi bus is broken out - not the parallel on :-/

in a previous project i used a similar pi-tft hat with the notro-fbtft+fbcp setup at roughly 20fps - which worked well for SNES/GBA/...-games :-)

  Are you sure? yes | no

david wrote 04/07/2016 at 19:10 point

Does it really work well? I recall GBA being 60 or 30fps

  Are you sure? yes | no

JohSchneider wrote 04/08/2016 at 11:03 point

good enough for me at least :-)

dunno about the original GBA...

  Are you sure? yes | no

Jasper parker wrote 04/02/2016 at 14:51 point

ok no worries ill contact my supplier and request a few of these PCB boards slip into my usual production :)

... Wondering why Pi zero is so expensive in Australia !

  Are you sure? yes | no

Jasper parker wrote 04/02/2016 at 11:15 point

fantastic 

Will you be doing a production run of the boards and offering them for sale ? 

  Are you sure? yes | no

JohSchneider wrote 04/02/2016 at 14:02 point

so far i don't plan to...

  Are you sure? yes | no

Jasper parker wrote 04/02/2016 at 06:10 point

Will the enclosure files be available for 3D printing ? 

  Are you sure? yes | no

JohSchneider wrote 04/02/2016 at 10:48 point

yes, as soon as i'm done printing and refining them i'll add them to the git repository. so far their still in the draft stage since i'm still busy building the hardware :-P

  Are you sure? yes | no

JohSchneider wrote 04/01/2016 at 19:27 point

@arnaud.durand : just had a look at those two dpads, they are tiny! :-) but probably to tiny, the pads are 2mm diameter and diagonaly only a few milimeter apart - IMO not ideal add a 3d-printed dpad-cross above it...

@K.C. Lee: how big is your d-pad solution, and what buttons does it use (btw: had something similar in mind to protect the screen from scratches - not a thin piece of packaging plastic though, but with a piece of acrylic)

  Are you sure? yes | no

K.C. Lee wrote 04/01/2016 at 22:10 point

I made my d-pad out of 5 individual tactile momentary switches.  Sorry it was a one off DIY hack not designed for a production.

I had the side activated one, but trimmed/hack it to be a vertical mounted one.  Not sure about the exact model as I got a bag of it.  It is probably Omron brand one based on the shape, colour and the style.  

http://www.digikey.com/product-detail/en/omron-electronics-inc-emc-div/B3F-3152/SW410-ND/31741

The most important thing is the capture cap and the really sloppy mechanical - that allow it to tilt slightly to the sides. Here are the others that are probably easier to use:

http://www.digikey.com/product-search/en/switches/tactile-switches/1114206?k=&v=39&FV=fff40011,fff8005e,8840017&mnonly=0&newproducts=0&ColumnSort=0&page=1&quantity=0&ptm=0&fid=0&pageSize=500

The spring is from my junk box.  It is very weak, but just the right strength to take off the slack so the diamond doesn't wobble.  I measured about 60g of compression force with my digital scale at its final resting position.

I laid the switch on a 0.4" grid.  The diamond is 0.9"x0.9" square, so 1.27" corner to corner.  It is slightly bigger than the one on the original Xbox controller, but you can trim it back to where it hits the buttons.  

Because it is built with tactile switches, there is a very notable click and feel to it not the mushy silicon dome.

  Are you sure? yes | no

JohSchneider wrote 04/02/2016 at 03:35 point

thnx for the input!

0.9"square is more space than i have on the pcb - thats probably the only downside to the all-tactile switch design: the size can't be cut down much because the individual switches need a minimum of clearance and such... still good design idea (especially the spring)

  Are you sure? yes | no

K.C. Lee wrote 04/01/2016 at 22:21 point

BTW take a look at the cursor keys of the PC8201.  I got one and that diamond key is very good for playing games.  You just have to make those key caps for the individual momentary switches.

https://deskthority.net/wiki/NEC_PC-8201

  Are you sure? yes | no

JohSchneider wrote 04/02/2016 at 03:32 point

that's an idea for the keycap - would be easy enough to reproduce with a 3d printer. will keep it in mind for the next project that needs a d-pad :-)

  Are you sure? yes | no

david wrote 03/31/2016 at 20:30 point

Hey! Nicee work! I was planning on doing something similar, I miss creating videogames :) How do you plan driving the screen? The bandwidth to drive the screen is 320*240*60*16 = 74Mbps, this is a 75Mhz SPI clock. Is that within the contraints? I dont think so... I looked at the ILI9340 datasheet and its possible to use a parallel 8 or 16 bit wide, which could reduce theclock down to 5MHz, and reduce the CPU usage. Can you comment on that? BTW only bigger screens like 2.8'' have adafruit boards available for soldering. Maybe using an adafruit screen without the PCB board (just add your connector to the PCB)  could help you a lot.

  Are you sure? yes | no

JohSchneider wrote 04/01/2016 at 18:21 point

thnx :-)

the screen i picked will run via spi but not at 60fps, more like 20 because of the bandwidth constraints, theres a guide linked in the pitft site - basically its notro's fbtft kernel module plus a daemon that copies from fb0 to fb1 (yes... not overly performant, but works well enough for these small resolutions and emulation-station)

see components listed above for a link to the adafruit board with the target tft

  Are you sure? yes | no

david wrote 04/02/2016 at 12:26 point

And don't you think it's gonna be too low? I mean, 20fps is gonna produce many glitches in most games. That screen shot support parallen 8/16 bit mode, does the connect break out those pins for you? We can always play a bit of kernel driver hacking :)

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates