waveshareteam
Tweak refresh performance of E-Ink Spectra 6 (4.0': 10sec, 7.3': 6.7sec)
The E-Ink Spectra 6 is likely not the fastest EPD, however it can be tweaked in case on knows the commands and arguments. With a bit of research, I determined the following: Especially the PLL CONTROL setting makes a huge difference, while the AUTO command can speed up the PON->DRF->POF(->DSLP) with pure hardware.
To give an impression, what the PLL CONTROL setting can do:
- E-Ink Spectra 6 - 4.0'
D7 D6 D5 D4 D3 D2 D1 D0
- - - - [ FRS ] -> Frame Rate Setting (in Hz)
0x00 -> 37662ms (= 0x10)
0x01 -> 30038ms
0x02 -> 22091ms
0x03 -> 19083ms (= 0x08 ~ 0x0F)
0x04 -> 14809ms
0x05 -> 12887ms
0x06 -> 11448ms
0x07 -> 9819ms
The default was 0x08 :-( With the new setting the 4.0' is below 10 seconds!
- E-Ink Spectra 6 - 7.3'
D7 D6 D5 D4 D3 D2 D1 D0
- - - - [ FRS ] -> Frame Rate Setting (in Hz)
0x00 -> 25365ms (= 0x10)
0x01 -> 19863ms
0x02 -> 14944ms
0x03 -> 12932ms (= 0x08 ~ 0x0F)
0x04 -> 10075ms
0x05 -> 8798ms
0x06 -> 7832ms
0x07 -> 6746ms
The default was 0x03 :-( With the new setting the 7.3' is achieving 6.7 seconds!!!
- E-Ink Spectra 6 - 13.3'
D7 D6 D5 D4 D3 D2 D1 D0
- - - - [ FRS ] -> Frame Rate Setting (in Hz)
0x00 -> 76357ms (= 0x10)
0x01 -> 38650ms
0x02 -> 19681ms (= 0x08 ~ 0x0F) // seems default (compared to timing)
0x03 -> Picture just dark
0x04 -> 12789ms, but not always correct
0x05 -> Picture not finished
0x06 -> -''-
0x07 -> not working
The default was 0x02 :-) I guess, in case of the 13.3', the Waveshare adapter solely supplies a 3.3V with ~330mA (if I am not mistaken). That might be insufficient to refresh faster.
Note: all EPDs are temperature sensitive. This could lead to slower or even better performance in your case, depending on the temperature given.
In case someone is interested in a much cleaner implementation, that can reach a bit faster speeds: https://codeberg.org/psiegl/E-Ink_Spectra_E6/ Hoping someone has the full spec, instead of my reversed engineered one.
