www.picgraphic.com

[trastikata]

Hello all,

check this new site www.picgraphic.com  - some useful tools and libraries.

- Graphic screen libraries with many options and tools, with lots of examples.

- Complete overhaul of the FAT32, now full support for 8-bit and 16b PICs, with some new commands

- Some other useful things

Hope it would be of help

=================================== UPDATE - Sep 13, 2026 - ===================================

TFT & EPD Graphic Library 2 has been updated to v2.04

=================================== UPDATE - Sep 10, 2026 - ===================================

I'm pleased to announce that TFT Graphic Library 2 has been updated to v2.03 — and with this release, it's now called TFT & EPD Graphic Library 2!

Here are the highlights of what's changed:

• Hardware support — Added hardware SPI support for newer PIC18-Q/K MCUs, which use a different hardware register set than earlier parts.

• Performance — Speed optimizations in several places; touch now use fixed-point Q16.16 math instead of floating point.

• EPD support — E-paper displays work quite differently from TFTs, so if you're new to EPDs, I'd strongly recommend reading through all the posted documentation before diving in.

• SD Card — significant rework:
• • Fixed a major bug in FAT_TraverseClusters() that could corrupt a file when it was closed and reopened
  • Fat.inc now supports disk partitioning and partition addressing
  • File_Init() now correctly detects a missing card and returns an error within 2–3 seconds, instead of hanging for minutes
  • Added a number of new commands

• 5. Shapes — Added a few new shape-drawing functions.

As always, I test as much as time allows, but like any hobby project, bugs are possible. If you find or suspect an anomaly, let me know here. I have a display on hand for every model I've posted support for, so I can test against real hardware — I just can't promise any particular timeframe.

Best regards,
D

[david]

What an incredible piece of work, both in technical content and overall presentation.  A great salute to Les's compiler also.
Take a bow.

David

[Maxi]

You are wonderful trastikata

[top204]

That is beautiful!

I cannot think of any other words to state how wonderful that it. :-)

Best regards
Les

[Dompie]

Wow, that is a great piece of work and with very extensive documentation (and that is a lot of work).
AWESOME!!!

Johan

[JonW]

Nice work buddy, that must have taken you a looong time!
Ill have to give it a go now

[midali]

Wow...Amazing !

[John Lawton]

Wow, a fantastic resource, thank you for putting it all together. It looks like I'll have to explore graphic displays now you've done the hard work!

John

[diebobo]

Looks awesome !

[pjdenyer]

Had a llok at this last night, what a job you have done, great

[charliecoutas]

Amazing, very good work.

[magic_gun]

only one word " AMAZING " .. specless

[Oskar-svr]

Very interesting friend and you can see the effort and time you have invested to achieve this, congratulations, you are incredible, greetings

[atomix]

You are very talented! Excellent job!

[SCV]

Great work and a big thank you for sharing with the community.

Tim.

[Wimax]

It's a great work, simply amazing !

[normnet]

Awesome!  What size SD cards are supported?  Have you considered developing the 4 bit parallel SD card interface for speed?  I had trouble porting the CRC calculation in C to basic.

[trastikata]

Awesome!  What size SD cards are supported?  Have you considered developing the 4 bit parallel SD card interface for speed?  I had trouble porting the CRC calculation in C to basic. 

Hello normet,

theoretical maximum card size is 2TB (2^32) sectors. I've tested it with up to 64GB only. I think there might a problem with small 128MB cards and FAT32 (Edit: I forgot I fixed that) but there are many combinations I still have to test and try.

From a hardware standpoint I don't see how a 4-bit SQI interface can be more efficiently applied on a PIC than the HW SPI? I think calculations show that a software based 4-bit mode solution will always be slower than HW SPI mode for the same MCU. If you have any ideas how to efficiently apply it, I am up for trying it.

Unlike the SD mode, in SPI mode the CRC is only required at the beginning for CMD8, after that you send a dummy CRC.

I think the benefits of the SD mode can be realized only with  a specialized hardware or FPGA with HW SQI and CRC.

[normnet]

Hello normet,

theoretical maximum card size is 2TB (2^32) sectors. I've tested it with up to 64GB only. I think there might a problem with small 128MB cards and FAT32 (Edit: I forgot I fixed that) but there are many combinations I still have to test and try.

From a hardware standpoint I don't see how a 4-bit SQI interface can be more efficiently applied on a PIC than the HW SPI? I think calculations show that a software based 4-bit mode solution will always be slower than HW SPI mode for the same MCU. If you have any ideas how to efficiently apply it, I am up for trying it.

Unlike the SD mode, in SPI mode the CRC is only required at the beginning for CMD8, after that you send a dummy CRC.

I think the benefits of the SD mode can be realized only with  a specialized hardware or FPGA with HW SQI and CRC.

The 4 bit mode was just something I was going to test.  Have you by chance tried a 200Mhz dsPIC33CK?  More speed for quicker screen loads. 

Great contribution!

[trastikata]

Have you by chance tried a 200Mhz dsPIC33CK?  More speed for quicker screen loads. 

Check at the end of this page: https://www.picgraphic.com/Howtostart1.html