SLT Contrast Loss...

[agorabasta]

Now what do we know?

1. We definitely can see some ghosting. It happens next to highlights exceeding at least 9EV levels over the shadows, and it doesn't happen in absence of blown highlights.
2. If the mirror is causing any haze to kill the contrast, that haze magnitude is definitely not much higher than the ghosting level. Even if it were 2EV higher than the level of ghosting, it's still -7EV wrt the saturated white. Otherwise the first thing to see in the shadows would be haze detail obliteration instead of ghosting.
3. There are no reasons for contrast loss other than ghosting, haze and flare related to the presence of mirror.

Now what it gives? Any contrast loss due to haze is definitely less than 1/256 of the white level - that's less 0.5% - i.e. practically inexistent. In fact, it must be even less than that, or we wouldn't be able to recover much shadow detail at all in presence of strong saturated highlights. The simple fact that SLTs simply show a DR loss of 0.3EV and not any more already means that there's no contrast loss due to the haze.

The flare caused by the mirror may and does happen at times, but that's an occasional thing and is not always present there.

[mike2008]

Yes it appears that the doubled image is <1% of the transmitted intensity. Quite an impressive job, there must be very effective coatings to smooth the refractive index difference between the film mirror and air. Anything lower spec than this would not be commercially viable as a DSLR replacement due to the IQ hit.

Perhaps all the scepticism is gradually being calmed.

[Greg Beetham]

I've still got a suspicion that it's not a 'clean' mirror, I think it's something fancy like a nano diffraction grating shorter than the standard wave maybe, either that or they've done something with the AA filter, whatever they did it works very well. The end result is quite an acceptable fun cam I think, (except for the heat probem and battery consumption and maybe EVF lag/latency), it will be interesting to see the SLT reception in the more serious end of the camera spectrum.
The line pairs I saw, A580 vs A55 seemed to indicate that there was a slight contrast loss in the A55, but how serious that translates into in real life I don't know, it doesn't seem like there is much impact in some side by side images I have seen.
But, be that as it may, I still won't have a camera (myself) that doesn't faithfully replicate reality on all occasions. (ghosting)
Greg

[mike2008]

I've still got a suspicion that it's not a 'clean' mirror, I think it's something fancy like a nano diffraction grating shorter than the standard wave maybe, either that or they've done something with the AA filter, whatever they did it works very well. The end result is quite an acceptable fun cam I think, (except for the heat probem and battery consumption and maybe EVF lag/latency), it will be interesting to see the SLT reception in the more serious end of the camera spectrum.
The line pairs I saw, A580 vs A55 seemed to indicate that there was a slight contrast loss in the A55, but how serious that translates into in real life I don't know, it doesn't seem like there is much impact in some side by side images I have seen.
But, be that as it may, I still won't have a camera (myself) that doesn't faithfully replicate reality on all occasions. (ghosting)
Greg

Displaying a lack of awareness of the limits of every lens you put in front of your nice sensor.

[Greg Beetham]

My awareness is fine.
Greg

[agorabasta]

Greg,

If anything like fancy epitaxial nanolayers have been used in that mirror tech, it should become readily apparent if the mirror is used in reversed position - i.e. putting the side normally facing the sensor to face the AF sensors instead.
A 'normal' mirror would then show almost no change in the resulting image.

Such a test could be done rather easily...

[agorabasta]

My awareness is fine.
Greg

My unawareness is finer yet

[Greg Beetham]

Greg,

If anything like fancy epitaxial nanolayers have been used in that mirror tech, it should become readily apparent if the mirror is used in reversed position - i.e. putting the side normally facing the sensor to face the AF sensors instead.
A 'normal' mirror would then show almost no change in the resulting image.

Such a test could be done rather easily...

That would be interesting.
Greg

[Greg Beetham]

[My unawareness is finer yet

Yeah well I don't claim to be 'aware' of everything either, but I do know what I require a camera to be able to do, and not to do, and ghosting isn't one of them, ever.
Greg

[agorabasta]

Actually, I'm with you, Greg, on that matter. It's outright abominable to me to even know that the unnecessary obstacle sits there between the lens and the sensor.

And that's an opportunity for Sony if they ever make an 'a77/a99'...

[Greg Beetham]

Kaf kaf I'm struggling with the mental concept of an A99.... and not just the backpack full of batteries, thermal cooling unit and memory cards that would be a requirement most likely.
Greg

[agorabasta]

OK, let's see...

1. The EVF power consumption is supposed to be less than that of the back LCD; all modern cams have some, all tiny P&S use the LCD (some have EVFs also) and their tiny batteries are no issue.
2. The a900/a850 have the SSS and don't suffer too much power drain. So that's no issue.
3. Memory cards capacities grow faster than the sensor resolution. This as well is no issue.

Well, Sony used some pretty laughable batteries in the SLT/Nex cams. That's easily correctable - use the older a200/300/350/700/900/850 batteries; so no issue as well.

So what exactly were you so joyously struggling with, Greg?

[bfitzgerald]

I don't know about contrast loss but I see some folks suggesting CA is greatly increased with SLT. If that's true or not I'm all ears or rather eyes!

[agorabasta]

Barry, the CA&PF are the same with SLTs. It's the amounts of misinformation surrounding them that's truly overwhelming.

[Greg Beetham]

I don't know about contrast loss but I see some folks suggesting CA is greatly increased with SLT. If that's true or not I'm all ears or rather eyes!

I'd say if that were true (could well be although DK hasn't said anything as far as I know), the keen eyed reviewers would have spotted it during the dilligent and intense scrutiny that they gave it, hmmm otoh they sortof missed the overheating issue, and the ghosting and they weren't what you would call exactly obscure.
Greg