p.1 #1 · Canon 5D4 dynamic range analyzed from RAWs
About 13.3 - 13.4 EV at ISO 100 (noise level about 3 DN).
This is about 0.1 EV less than 1DX2:
ISO 100 (793A6840.CR2): 13.38 EV (1dx2 was 13.55)
ISO 100 (793A6466.CR2): 13.29 EV
ISO 100 (793A6491.CR2): 13.29 EV
ISO 400 (793A7165.CR2): 12.88 EV (1dx2 was 12.83)
ISO 800 (793A7094.CR2): 12.32 EV (1dx2 was 12.58)
ISO 1600! (793A6463.CR2): 11.57 EV (1dx2 was 11.82)
ISO 3200 (793A7330.CR2): 10.65 EV (1dx2 was 11.28)
ISO 25600 (7A0A0098.CR2): 8.40 EV (1dx2 was 8.49, 1dx was 8.79)
About 1.60 EV above 5D mark III results, exactly just like 80D and 1DX2 which also are about 1.60 EV better than their predecessors.
The test image at ISO 1600 had no blown pixels, so the saturation level is assumed (but surely correct). All ISO levels seem to saturate at exactly DN 16383, the black level is 512 for ISO 100 (and likely ISO 200, but dpreview had not a single shot at ISO 200) and 2048 for the rest of the ISO levels.
The previous threads results (since the new Canon sensor era):
PS: I edit the initial post because the results were wrong: a saturation of 14380 (instead of 16383) was hardcoded in the tool, causing a general underestimation. I just have changed my impressions and comments in consequence... I apologise by the not-so-good initial news :-)
* * * * UPDATED * * * *
These are the final DR graphs, using the RAW samples from Imaging Resource. When there are several RAW for a given ISO level (e.g. 4 files for ISO 100) the file with the higher DR has been selected (but the differences are tiny).
Note how the 5D4 has different values for each color channel, specially at low ISO levels. This is a very rare feature:
The comparison uses the two greens channels combined on each camera. Had we used the combination of the 4 channels (RGGB) the result would almost 100% match that of the combined greens alone:
p.1 #4 · Canon 5D4 dynamic range analyzed from RAWs
arbitrage wrote:
DPReview has 5D4 in there comparison with RAWs to download at all ISOs now.
Thanks for doing this analysis.
Maybe tomorrow I'll take a look on them.
I have corrected my (wrong) initial results. The tool I use (https://github.com/lightful/hraw) is still manual (too much manual, editing by hand the values) and I made a mistake.
Fortunately, the 5D4 is just as good as the other Canons with the new generation of sensors. And the difference is consistent: +1.6 EV improvement at ISO 100 in all the new cameras.
p.1 #6 · Canon 5D4 dynamic range analyzed from RAWs
Hey good work, thank you for proving what we had hoped for. Looks like the same process being employed sensor wide at Canon.
I wonder if this is what Canon will stick to for another 5-8 years or will keep pushing the DR over the next generations. Be nice to eventually see them push into the high 14EV for DR, but for now things are looking pretty good, no complaints on the sensor other than keeping the AA filter.
p.1 #8 · Canon 5D4 dynamic range analyzed from RAWs
Thanks for the good work, although I was anticipating the DR would be slightly better than 80D and 1Dx2 being a all rounder camera, but it is a welcome improvement
p.1 #9 · Canon 5D4 dynamic range analyzed from RAWs
I think this is exactly what I was expecting.
I know it's not class leading but they have a 5 year head start, and at this level without banding I'm now happy for the very occasional shot that I would need it for.
p.1 #15 · Canon 5D4 dynamic range analyzed from RAWs
Great news... and the files look very good too (contrast / colors). I noticed something similar with the 80D files. Not sure how to describe what I am seeing, but I really like it and it seems different than the previous generation Canon sensors. Now to get my hands on a proper profile to play with the RAW files.
p.1 #17 · Canon 5D4 dynamic range analyzed from RAWs
Very good, thanks for making these measurements. I think we are at the point where LCD panels and print are now the limiting steps in photography. No doubt the pixel noise performance is much improved too. Time to upgrade I think.
p.1 #19 · Canon 5D4 dynamic range analyzed from RAWs
cgarcia wrote:
ISO 3200 (793A7330.CR2): 10.65 EV (1dx2 was 11.28)
That number surprised me.
I know you can't always tell 1/2 stop in reality, but when I compared the normalised ISO3200 studio shots from DPReview, I couldn't see a half stop difference. I even raised the exposure on both and still couldn't see a difference.
