As there is still a lot of confusion and misunderstanding regarding sRGB, and the EOFT that should be used for display calibration, this post should provide the definitive answer to what is/is not correct.... hopefully!

This post follows on from an old post regarding changes to Rec709 and sRGB we made to LightSpace back in Oct 2016.

We tried to explain the truth behind sRGB in that post, and the follow-up post in that thread made in Feb 2021.
But, it is worth making a specific post to explain the facts in slightly more detail, as some still seem to misunderstand the simple facts.

So, it is worth reiterating that:

When sRGB was developed, all displays were CRTs, with fixed gamma that could NOT be adjusted.

And nothing can alter that simple fact.
That means all displays were calibrated to a pure power law EOTF when displaying sRGB images.

But, it is worth adding further explanation to that simple fact.

When displaying sRGB images via any PC, the sRGB compound EOTF is applied to the sRGB images BEFORE they reach the display.
That means sRGB images are effectively 'encoded' with the compound EOTF.

Therefore, if any of todays displays are actually calibrated to the sRGB compound EOTF, the source sRGB images will be display incorrectly, obviously...

The confusion seems to stem from a lack of understanding of the workflow employed when displaying sRGB images.
Historically, the term 'calibration' appears to have originally been erroneously applied to the action of encoding any sRGB image with the compound EOTF before being sent to the display, which were in themselves always calibrated to a power law EOTF (Gamma 2.2), which is irrefutable as all display in use at the time were locked to a power law EOTF.

Another way to think about this is the historical difference between PCs and Macs with regard to Gamma/EOTF.
Macs were 1.8, and PCs were 2.2
But, they both could be connected to the exact same display...
And the display's gamma/EOTF didn't change - the signal being sent to the display did.

And that brings us back to the original statement.
At no point should any display ever be 'calibrated' to the compound sRGB EOTF.

It really is that simple.

Steve