A LCD TFT display and a Propeller. Help and comments needed...
grasshopper
Posts: 438
I was instructed to implement a LCD TFT display on my next design and want to use the propeller. I am questioning the memory and the speed of the prop if it would be able to drive a 800x480 resolution monitor?
I was thinking, and; perhaps 3 serial to parallel ICs for the RGB data lines would help. It could free up some i/o on the prop since the display requires 24, 8 per RGB.
Also i searched the forums and was unable to find a driver or any other information that was promising. Any ideas or comments?
data sheet is here
www.apollodisplays.com/pdf/T-55290D050J-LW-A-AAN_071120_Eng.pdf Its a PDF
I was thinking, and; perhaps 3 serial to parallel ICs for the RGB data lines would help. It could free up some i/o on the prop since the display requires 24, 8 per RGB.
Also i searched the forums and was unable to find a driver or any other information that was promising. Any ideas or comments?
data sheet is here
www.apollodisplays.com/pdf/T-55290D050J-LW-A-AAN_071120_Eng.pdf Its a PDF
Comments
does it really have to be THIS Display ?
Does it really have to be a Display withh RGB input?
Who wants to use RGB nowadays?
In the library there are drivers for VGA-output.
There are drivers up to 1024x768 pixels in textmode
If i saw it right in the datsheet the DataClock has to be VERY fast 30 MHz !
One assembler-command needs 4 clockcycles this is only 20MHz
Maybe it is possible to to it slower but i don't know what lower speed means to the screen refreshing rate (complete pictures per second)
best regards
Stefan
media.digikey.com/pdf/Data%20Sheets/Optrex%20PDFs/T-51380L050J-FW-P-AB%20Spec.pdf
I think 6 mhz is all it is wanting...
You will also need to reduce resolution there isn't the memory for 800x480 even at the reduced color depth.
You probably also should choose the prop crystal so the counter can generate a sensible dclk without jitter e.g. a 4Mhz crystal will run the prop at 64Mhz, divided by 2 gives you a clean 32Mhz dclk. Though the question is then whether the prop could run vga at that speed. You could try 5Mhz / 2 for 40Mhz or 10Mhz / 4 gives also gives 40
Thanks for link.
It is fuly VGA compatible
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Nothing is impossible, there are only different degrees of difficulty.
For every stuipid question there is atleast one inteligent awnser
If you dont ask you wont know
Sapieha
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
E3 = Thought
http://folding.stanford.edu/·- Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.
Post Edited (RinksCustoms) : 9/13/2008 9:33:13 PM GMT
So i translate this to "It will work". ill order up a few and make some boards. If it works out perhaps i can give away the remaining leftover prototypes.
Thanks so far but stay tuned, , ,
If I can see corectly it has one mode RGB+H+V sync in.
and one mode RGB+Composite sync in
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Nothing is impossible, there are only different degrees of difficulty.
For every stuipid question there is atleast one inteligent awnser
If you dont ask you wont know
Sapieha
Have a read of my comments here. Then talk to Baggers about timings that might be achievable.
Question: Have you seen the LCD in action? 200cd is not bright but if it's got good contrast it'll prolly be fine.
Evan
It says otherwise but you could also experiment with removing blanking altogether.
Evan
This way you not only make sense of the datasheet but also have the confidence in doing the probing when you are trying to get the Prop to go.
Evan
With that in mind the minimums for DE will need to be observed.
But there is still no bottom limit the clock so there is a good chance you can get away with running at 20 MHz and have the vertical refresh below spec.
Evan
So i translate this to "It will work". ill order up a few and make some boards. If it works out perhaps i can give away the remaining leftover prototypes.
Thanks so far but stay tuned, , ,Back to Top
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Nothing is impossible, there are only different degrees of difficulty.
For every stuipid question there is atleast one inteligent awnser
If you dont ask you wont know
Sapieha
I cant seem to find a solution, any one else trying to run a 5-7inch TFT and a propeller without having to purchase the small ones over at 4D
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Timothy D. Swieter, E.I.
www.brilldea.com·- Prop Blade, LED Painter, RGB LEDs, uOLED-IOC, eProto fo SunSPOT, BitScope
www.sxmicro.com - a blog·exploring the SX micro
www.tdswieter.com
Can you explain to me this notion of not enough memory? I am not sure why limits on memory would be a factor when I could hook up more to a pin on the propeller, right? Correct me if i am wrong but EEproms are cheap and well I seen people running a propeller using 2 gig memory drives. Would these devices be to slow?
With 32 kbytes of memory, you can get 512x384 monochrome bit mapped with tiled color, as the example. For 640x240 monochrome, similar memory requirements. For 800x600... no dice, as 800x600 requires 60kbytes for the bitmap. Of course you can use only text mode (or mixed text and graphics) using less memory. Bigger resolutions not only need big bitmaps, but also big fonts (well, the parallax font is a savior here) and so on. I'm currently using a 640x480 panel as 320x240 (LCD), it is a dual scan so I have to produce both lines simultaneously... with one cog, it is very packed in timing, I should change to a 320x240 panel, but this setup... works!, and you know the rule, if it works, do not touch it, as I had enough trouble to get it working at all in the first place.
I was thinking in using a 800x600 panel as 400x300, but only when I understand how the pixel info is encoded (it has 2 8 bit ports, one for the upper part, one for the lower). It is sold for 15 euro at a discount shop here in Germany.
Edit: Link www.pollin.de/shop/produkt.php?ts=0&p=OTg4OTk4#MTI1OTc4OTk=
Question: The link points to a DSTN panel, it is possible that this panel only supports set or cleared pixels and thus shades of a color are only achieved with setting/clearing the corresponding pixel in a period of time (so the effective intensity is going to be the integral of the set time ?), and not with 6 or so bits to specify intensity ?... well I thought it worked differently....
Post Edited (Ale) : 9/16/2008 9:09:25 AM GMT
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Timothy D. Swieter, E.I.
www.brilldea.com·- Prop Blade, LED Painter, RGB LEDs, uOLED-IOC, eProto fo SunSPOT, BitScope
www.sxmicro.com - a blog·exploring the SX micro
www.tdswieter.com
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer
Parallax, Inc.