[thibs]

Don't trust images too much.
Until recently Pentax used istD drawings as not to unveil any novelty.

[Arto Suomi]

Don't trust images too much.
Until recently Pentax used istD drawings as not to unveil any novelty.

Exactly. The purpose of the diagram in the patent is just to clarify the textual part and show what the terms "shock-absorbing pin", " mirror-down shock-absorbing lever" etc. mean . Whether the diagram shows an APS or FF mirror box is irrelevant for the patent and you can not draw any conclusions about FF camera coming or not from that.

[Matjazz]

At first glance I thought it's shake reduction for the mirror
but this is probably just to increase frame rate and/or silence mirror flop.

[falconeye]

Sorry for that! I hope the link is working fine now! I don't understand why it's not working.

Links cannot normally copy-pasted. You first have to strip off your session cookies etc. Simplify and retest URLs before pasting.

And a better shock absorbant system is necessary for a bigger mirror.

No, shock absorbance just lowers the idle time between mirror up and curtain open. It is the effective curtain weight which must be kept under control.

IMHO, the drawing isn't conclusive wrt size. The mirror box looks a bit small for FF, but Fig.1 shows a relatively large sensor. But the original FF box supported F/1.2 which may not be a goal anymore.

Looking at the mirror box on the Z-1 I don't think Pentax could design telecentric lenses for a full frame sensor.

Not true, look again (the lens rear element is typically outside the mirror box).
But Pentax doesn make (and deosn't need to make) telecentric lenses anyway. The registration distance makes most normal lenses behave telecentric enough.

[Uluru]

Insect eyes;
I wonder would it be possible to create a DSLR that instead of one FF or an APS-C sensor, it has a cluster of 1/1.7" sensors behind the mirror.
Say 6 or 12 of them. They all gather light as the same time.
Using parallax error calculations and accessing information from all of them, I believe it could be possible to better analyse what is best possible exposure, and optimise colour and tone accuracy in the final image. Or is this just sci-fi?

[aurele]

It can probably be done for dslr, as we already do it with space telescope for a long time.

The main problème is that it requires a lot of calculation power, and a lot of precision in both optical domain, and in assembly.

I think it's'stille cheaper and easier to improve actual tech.
Have you seen the news about a new sensor made of "graphene" (check the english translation of this french word that designate a carbon material), which is theorocally 1000x more sensitive to light ?

[goubejp]

It can probably be done for dslr, as we already do it with space telescope for a long time.

The main problème is that it requires a lot of calculation power, and a lot of precision in both optical domain, and in assembly.

I think it's'stille cheaper and easier to improve actual tech.
Have you seen the news about a new sensor made of "graphene" (check the english translation of this french word that designate a carbon material), which is theorocally 1000x more sensitive to light ?

It is only a theorical sensitivity cause light is a flux of particles called photons; with current sensitivities there are only hundreds of photons (not thousands) hitting each sensor cell. You can't improve sensor sensitivity so much in a sense that it will result that only one or no photon will hit each cell.

[falconeye]

What are you all smoking?
Or have been smoking when you have been supposed to listen to your physics teacher?

Insect eyes are an early attempt at seeing by nature, superseeded by the incredibly more powerful complex eye with an iris. E.g., insect eyes destroy phase information and therefore have a severe diffraction problem.

Current sensors achieve better than 50% quantum efficiency. E.g., the K-5 sensor detects between 2.5 and 40000 electrons per sensel within its dynamic range. Which is 5 photons. Even less because much of the noise which limits dynamic range is part of the photon nature of noise.

The remaining progress in light sensitivity will be rather limited, remain below a factor 2 and the biggest progress will come from a different color separation technology, not sensors.

People promising a 1000-fold sensitivity to light appeal to investors who are known to have been smoking the same stuff

[jpzk]

What are you all smoking?
People promising a 1000-fold sensitivity to light appeal to investors who are known to have been smoking the same stuff

Probably this ... OG Ghost Train Haze !

JP

[Uluru]

What are you all smoking?
Or have been smoking when you have been supposed to listen to your physics teacher?

...

People promising a 1000-fold sensitivity to light appeal to investors who are known to have been smoking the same stuff

Maybe this: the mighty joint!

You Tube

You Tube

[Class A]

Have you seen the news about a new sensor made of "graphene" (check the english translation of this french word that designate a carbon material), which is theorocally 1000x more sensitive to light ?

As I posted elsewhere already, the "1000x" better sensitivity is with respect to previous graphene-based approaches, not to the standard that has been reached with CMOS technology already. The recently circulated news article apparently is deliberately misleading in this aspect, but if you read the abstract of the original paper it becomes clear that the breakthrough is relative to graphene-based detectors, not an absolute one.

[aurele]

As I posted elsewhere already, the "1000x" better sensitivity is with respect to previous graphene-based approaches, not to the standard that has been reached with CMOS technology already. The recently circulated news article apparently is deliberately misleading in this aspect, but if you read the abstract of the original paper it becomes clear that the breakthrough is relative to graphene-based detectors, not an absolute one.

i read your other post after posting here

So yeah, it's not as big as it seems !

[goubejp]

Could the graphene-based technology be used to manufacture a non-flat sensor ? This would be an clever approach to correct light falloff (at least partially) and field curvature