I have at some stage downloaded a dng-file with a color circle:
However, in darktable (both as thumbnail and when in darkroom), it looks like this on my lap top screen (100 % DCI-P3):
I’ve tried to no avail with different export profiles settings in output color profile.
What can be the cause?
Can you share the dng file for others to test?
It’s the one uploaded above.
Come to think that in Xviewer the file present itself like this:
Is that checkered pattern something about the image having an alpha channel or something like that which I don’t know much about (transparency, ref the black background in OP?), and that again darktable don’t handle alpha channels?
Hello, your color circle looks fine in Gimp and gThumb, but opened in ART it gives the same wrong image as you showed above.
Thanks! That could perhaps fit in with what I just suggested above, as I believe Gimp (which I don’t know much about) can handle alpha channels, while ART is a raw developer like darktable.
There is something weird going on. You can resave the file in gimp, and it will be displayed okay in darktable.
Here the porperties of the original and the resaved file in comparison:
Original:
$ pngcheck -v pngkit_green-lightsaber-fargehjul-png_940464.png
File: pngkit_green-lightsaber-fargehjul-png_940464.png (31502 bytes)
chunk IHDR at offset 0x0000c, length 13
482 x 482 image, 32-bit RGB+alpha, non-interlaced
chunk cHRM at offset 0x00025, length 32
White x = 0.28056 y = 0.29582, Red x = 0.63027 y = 0.33088
Green x = 0.28782 y = 0.5821, Blue x = 0.12664 y = 0.04978
chunk pHYs at offset 0x00051, length 9: 2835x2835 pixels/meter (72 dpi)
chunk IDAT at offset 0x00066, length 31380
zlib: deflated, 32K window, maximum compression
chunk IEND at offset 0x07b06, length 0
No errors detected in pngkit_green-lightsaber-fargehjul-png_940464.png (5 chunks, 96.6% compression).
Gimp:
$ pngcheck -v pngkit_green-lightsaber-fargehjul-png_940464_b.png
File: pngkit_green-lightsaber-fargehjul-png_940464_b.png (29567 bytes)
chunk IHDR at offset 0x0000c, length 13
482 x 482 image, 32-bit RGB+alpha, non-interlaced
chunk sRGB at offset 0x00025, length 1
rendering intent = relative colorimetric
chunk gAMA at offset 0x00032, length 4: 0.45455
chunk cHRM at offset 0x00042, length 32
White x = 0.3127 y = 0.329, Red x = 0.64 y = 0.33
Green x = 0.3 y = 0.6, Blue x = 0.15 y = 0.06
chunk pHYs at offset 0x0006e, length 9: 2835x2835 pixels/meter (72 dpi)
chunk IDAT at offset 0x00083, length 8192
zlib: deflated, 32K window, maximum compression
chunk IDAT at offset 0x0208f, length 8192
chunk IDAT at offset 0x0409b, length 8192
chunk IDAT at offset 0x060a7, length 4804
chunk IEND at offset 0x07377, length 0
No errors detected in pngkit_green-lightsaber-fargehjul-png_940464_b.png (10 chunks, 96.8% compression).
That might give some hint, where to look for the problem.
@EspE1 you mentioned a DNG file in your original post but it is a PNG file with transparency. I opened a png file of a helicopter with transparency that I have previously created using GIMP and DT doesn’t show transparency. In the case of my image it was black in the background rather than transparent when viewed in DT.
Sorry, Terry, that was a typo. The question is only about this png file.
What version of dt? There was some PNG change related to alpha that only made it to 5.0.1…
If you convert the PNGs to PAM with netpbms pngtopam --alphapam, you’ll see this for the top left corner, 3 pixels as RGBA:
original: ff d4 00 00 ff d3 00 00 ff d2 00
resaved: 00 00 00 00 00 00 00 00 00 00 00
(pngdump.py shows the same, if you do the filtering manually, because the dump doesn’t account for that)
They should be treated equal, but aren’t, even with darktable 5.0.1 (which, for me, shows the circle on a black background).
Sorry, I forgot to include that in OP.
I use dt 5.0.1.
I am on DT 5.1 windows and the problem discussed in the original post is still occurring. There are times that GIMP complements DT and RT for photo editing and this png file with transparency is one such case. I am not aware of any raw file editor that works with transparency but am happy to be enlightened if one exists.
