Hey Guys,
Wondering if we can create png files and display them on the LCD with a 5-10 sec refresh rate without using up too much resources.
The Idea of using chromium to do the layout is easiest but a bit expensive resource wise.
I think that installing imagemagick in order to create the image file (rendering the text in a predefined template image) and then display the image on the lcd using fbi, could work.
Does anyone have experience with using imagemagick on a pi? is it resource intensive?
That's possible to use imagemagick and I've run a few tests. We already use fbi to display QR-codes. Take a look at scripts like these https://github.com/rootzoll/raspiblitz/blob/master/home.admin/XXdisplayLCD.sh
I was surprised that Imagemagick with fbi was a more resource intensive than simply running chromium or epiphany. I've started writing svg-files that chromium can display in an xserve session. We can take a look at this issue
https://github.com/rootzoll/raspiblitz/issues/88#issuecomment-467578776
in which we discuss how to setup a browser.
Issue #388 has a mock up of a status page based on svg. This is just a proof of concept and I'm not sure if it's worth the effort and the computing power on this little computer.
@frennkie and @fluidvoice were involved, too.
This is my latest svg:

That's possible to use imagemagick and I've run a few tests. We already use fbi to display QR-codes. Take a look at scripts like these https://github.com/rootzoll/raspiblitz/blob/master/home.admin/XXdisplayLCD.sh
I was surprised that Imagemagick with fbi was a more resource intensive than simply running chromium or epiphany. I've started writing svg-files that chromium can display in an xserve session. We can take a look at this issue
in which we discuss how to setup a browser.
Issue #388 has a mock up of a status page based on svg. This is just a proof of concept and I'm not sure if it's worth the effort and the computing power on this little computer.
@frennkie and @fluidvoice were involved, too.
Does anyone have some benchmarks on the pi? using chromium and svg images to render the status screen? I can definitely make it look badass as long as the work is not for nothing.
will look into it
We were seing that using a broswer to display a html page on the LCD it takes 300MB+ mem constant even with a small broser like Epiphany. Also running X server with all needed parts to build a GUI for the LCD starts with 150MB+ constant mem.
So we want to research in having a script/info-service building a image to display per framebuffer (fbi command). This mechanism could also be adopted for ePaper displays later. Getting touch working while the LCD is in framebuffer-mode would then make the 4-Button UI possible.
Related issues for Infoservice and Touch to keep in view:
https://github.com/frennkie/blitzinfod/blob/master/server.go
Looks like X server is running with less mem foot print as assumed in the beginning - option is back on the table.
I still really like the HTML Info Page. But I would not run a Browser on the RaspiBlitz. Instead the HTML Info Page will be hosted by blitzd and can be access from
@arcbtc had the idea to use a really small device (ESP32?! Node-MCU? RPi Zero?!..) with a small to medium sized display and use that to show the RaspiBlitz Info Page (physically separated..).
Most helpful comment
This is my latest svg: