Works on nano! But, not with USB power... Had to use power jack on Prop board:
(also I had to solder extra Prop Plug header that comes out sideways...)
I won't claim a P2 bug this time, until I investigate some more...
I did just verify that PNut appears to compile the bytes in the correct order.
So, it looks like rdbyte is swapping order within words.
But, like I said, I'll look harder at my code.
Here's the part that reads the characters:
BTW: I've got 7 more of these boards... If anyone else wants to play around with this, I can make a pretty good offer: $20+shipping, including LCD and 40-pin male header pins installed.
Was reading the new docs today and now thinking there might be a way to use the streamer for LCD (or other things) without a streamer generated pixel clock...
I noticed a wait/interrupt event that is NCO rollover.
If I understand this right, this is when new data is read in or outputted from streamer to pins.
So, maybe one could set up a interrupt or wait loop that toggled the pixel clock pin whenever that happened?
Was reading the new docs today and now thinking there might be a way to use the streamer for LCD (or other things) without a streamer generated pixel clock...
I noticed a wait/interrupt event that is NCO rollover.
If I understand this right, this is when new data is read in or outputted from streamer to pins.
So, maybe one could set up a interrupt or wait loop that toggled the pixel clock pin whenever that happened?
Chip talks about the new 'Streamer Empty" event in this posting - it's an addition to the next release.
The NCO rollover has been there.
I'm still trying to wrap my head around all this video stuff. I've never done video before. My idea of a front porch is totally different!
It seems like one of these events will be able to help out, I'm not sure which yet.
I'm looking at/playing with your OneBitFont code and playing with P2 features and I see this gigantic font data structure (4kB) and then this HUGE bit buffer for the display (~16kB) and think, "wow, that's a lot of memory!"...then I realize on the P2, that's less than 4% of my HUB!
The P2 'mul' seems to work fine as a replacement for you multiply routine. I replaced 'call #Multiply' with 'mul y,x" and it still works. I'm sure I'll break something this afternoon!
I'm looking at/playing with your OneBitFont code and playing with P2 features and I see this gigantic font data structure (4kB) and then this HUGE bit buffer for the display (~16kB) and think, "wow, that's a lot of memory!"...then I realize on the P2, that's less than 4% of my HUB!
The P2 'mul' seems to work fine as a replacement for you multiply routine. I replaced 'call #Multiply' with 'mul y,x" and it still works. I'm sure I'll break something this afternoon!
Woo hoo! Just got back from Thanksgiving with the in-laws and there were two packages waiting: an A9 board and the 4.3" display! Took no time at all to get them up and running (running Lenna_V2)! Thanks Parallax and Rayslogic!
Comments
Now, I can see about adding in a font and try it on Nano...
Here's a screen shot and code:
(also I had to solder extra Prop Plug header that comes out sideways...)
I'm defining the string to write like this:
But, I think PNut is rearranging my characters (see pic).
Is there an easy fix for this?
You should be able to do sequential RDBYTE's to read those characters in.
I did just verify that PNut appears to compile the bytes in the correct order.
So, it looks like rdbyte is swapping order within words.
But, like I said, I'll look harder at my code.
Here's the part that reads the characters:
DAT 'font drawing code DrawFontCog org 0 mov row,#7 mov col,#20 mov pMess,##Message MessLoop rdbyte c,pMess cmp c,#0 wc,wz 'zero marks end of string if_z jmp #TheEnd mov s,c 'make char source address shl s,#4 '*16 add s,##Font1 'make destination start address mov x,row shl x,#4 '16 rows per char mov y,BytesPerRow call #Multiply mov d,y add d,col add d,##HubLine1 mov y,#16 '16 rows in character CharLoop rdbyte charByte,s add s,#1 wrlong charByte,d add d,BytesPerRow djnz y,#CharLoop 'Next character add col,#1 add pMess,#1 jmp #MessLoop TheEnd jmp #TheEndMy LCD driver is reading in words, which has high and low bytes swapped...
Guess my photo data is also in words, so it worked out.
Think should make the LCD driver read in bytes...
I can link to PSB or 4.3" breakout at rayslogic.com for the datasheet...
The PCB has no specs, just something I drew up rather quickly one morning...
I could post the schematic and PCB image, if that helps...
There are some extras on the board, that I don't know if they work yet:
pads for uSD jack and pads on bottom for SQI flash chip and eeprom.
regards
Lachlan
Oh, and I need to know if this is for P123 or DE type board, so I know which 40-pin header to solder on...
I noticed a wait/interrupt event that is NCO rollover.
If I understand this right, this is when new data is read in or outputted from streamer to pins.
So, maybe one could set up a interrupt or wait loop that toggled the pixel clock pin whenever that happened?
"Hello World."
Look like more fun times ahead, now with an LCD too!!
Chip talks about the new 'Streamer Empty" event in this posting - it's an addition to the next release.
The NCO rollover has been there.
I'm still trying to wrap my head around all this video stuff. I've never done video before. My idea of a front porch is totally different!
It seems like one of these events will be able to help out, I'm not sure which yet.
I was supposed to ship another one today, but had to play soccer mom all day
I probably should post a circuit diagram and stuff soon...
I'm looking at/playing with your OneBitFont code and playing with P2 features and I see this gigantic font data structure (4kB) and then this HUGE bit buffer for the display (~16kB) and think, "wow, that's a lot of memory!"...then I realize on the P2, that's less than 4% of my HUB!
The P2 'mul' seems to work fine as a replacement for you multiply routine. I replaced 'call #Multiply' with 'mul y,x" and it still works.
Lots of RAM and processing power.
Just did 8-bpp VGA with Ariba's driver and it uses up more than half of RAM...
Was feeling a little more squeezed there...