Be careful with that type of analysis; the number of pixels plays a huge role in base-ISO PDR (or DxO "Print" DR). Look at their ranking in terms of pixel count, and there is a very strong correlation.
For any given type of sensor readout, the more pixels, generally speaking, the higher the base-ISO DR, because most of "pixel-level" post-gain read noise has nothing to do with the pixels themselves, and everything to do with the signal path beyond the amplifiers.
Hi John,
Sorry to say, I absolutely disagree.
I'm not convinced that you disagree that much. Perhaps you inserted an assumption when you read my post, that I meant that more pixels is
bad for base-ISO DR, and did not notice that I wrote "for any given type of readout". I have never held the belief that higher pixel counts or densities decrease base-ISO DR, that I can remember. I remember arguing for more density for more base-ISO DR with Roger Clark almost 15 years ago, when he believed that pixel IQ throttled image IQ, regardless of MPs.
Let's compare Canon 5DIII and Canon 5DIV, with the later model having 30 MP while the older having 23 MP:
The 5D4 has two things going for it; less post-gain read noise due to a better
type of readout at the pixel level,
and more pixels. More pixels always seems to help unless base-ISO pre-gain read noise is much higher than post-gain read noise, at which point, pixel qualities limit DR per unit of sensor area (which is exactly what happens at higher analog-gain-based ISOs).
Comparing Nikon's latest generation:
Again, different types of readout
and different pixel counts,
both working in the D850's favor for base-ISO DR (although a FF full of D500 pixel quality would be about 1/3 stop better yet).
As photosites get less noisy and downstream readout circuitry gets less noisy, it is still the case, generally speaking, that many current sensors are post-gain-read-noise-limited in a way that makes higher pixel density improve base-ISO DR. At some point in the future, this may prove to be less helpful as higher densities create significantly lower pixel DR in the photosites themselves. Even when input-referred read noise charts the same or very close at the lowest analog gain ISO and double that ISO, the post-gain noise may have spatially-correlated noise that doesn't show in standard deviations, but shows visibly in shadow pushes to the same high ISO exposure index.
Any time you see the horizontal trend in DR near base ISO, deviating from a straight descending line more typical at the higher ISOs, that type of readout circuitry would get more base-ISO DR with higher pixel density, as the ratio of post-gain-to-pre-gain read noise is still very high at base ISO.
