search

pimoroni / inky

Combined library for V2/V3 Inky pHAT and Inky wHAT.
https://shop.pimoroni.com/?q=inky
MIT License
578 stars 121 forks source link

Try #104 again now that tests are fixed - better red colour on older Red HT pHAT #106

Closed jerbzz closed 3 years ago

jerbzz commented 3 years ago

https://user-images.githubusercontent.com/28573604/113060221-439fa000-91a8-11eb-8a47-83daa4192b7f.jpeg

Makes a big difference!

jerbzz commented 3 years ago

https://github.com/pimoroni/inky/pull/104 is the original

Gadgetoid commented 3 years ago

Actually finding boards to test this on is going to be a challenge! You might have to bear with me a little.

One concern I have with this change is that the "ink" is very temperature sensitive and thus different LUT values can produce different results for different people- depending on their environment. It's possible this improves things all 'round, but I'll have to try sticking the board in a fridge and on a radiator (don't try this at home, haha) to be sure there aren't any regressions.

(Edit: But in the meantime thank you for this PR and the others! really appreciated)

jerbzz commented 3 years ago

You're welcome - I don't really know what I'm doing but I recognise the importance of "Talk is cheap - show me the code", so any small contributions I can make are hopefully useful. I have no issue with sticking my pHAT in the freezer but less sure about the radiator! Maybe the airing cupboard? 😆

donbing commented 3 years ago

I have 2 red inkywhats and one yellow here. if someone can tell me how to perform the required testing, i'd be happy to do so!

is it just a matter of loading this branch up and displaying an image, putting one in the freezer and one on a radiator, and then checking what colour the things are after 30 mins or so? (i'd upload images obvs)

Gadgetoid commented 3 years ago

Just tested the before/after and I don't see quite as drastic a difference. Presumably due to it being really rather hot in here.

Stuck it outside to cool down and still looks fine. I think we can merge this! Thank you for asking though @donbing!

In retrospect a freezer might be a little drastic!

siddhantvirus commented 3 years ago

Hi ! I just got one of those high-temperature inky phats and it seems like the changes haven't had an effect on the output. It seems like rolling back version 0.0.4 did fix the problem, as suggested here #66 . Maybe we could add an if condition where if these particular boards are found [ which is definitely being read from the EEPROM and then labelled as Detected Red pHAT (High-Temp) in the output ] it makes the approach a little easier for everyone?

siddhantvirus commented 3 years ago

@Gadgetoid @jerbzz so I studied the changes between v0.0.4 and v1.1.1 here and seems like we are definitely checking the display type and then assigning the lut type. Also, the luts for red_ht have changed, can we revert the same, the changes which were made here https://github.com/pimoroni/inky/commit/0b29e83ca639902d447a7a69946eef3bb6bcee88 seem to be responsible
I can upload videos and/ or images showing the stark difference in colours if needed.

P.S. I apologize for the notoriously large number of edits in my comments but it as I discovered changes across versions it made sense to update the comment accordingly.

Gadgetoid commented 3 years ago

Try grabbing v1.2.1 - looks like the change has been sitting in GitHub for a while, unreleased!

siddhantvirus commented 3 years ago

Thank You ! I'll give it a shot, I did clone the repository separately and try last night , but I'll check again and put an update here