[gazonk]

Staff note: This post may contain affiliate links, which means Pentax Forums may earn a small commission if a visitor clicks through and makes a purchase. If you would like to support the forum directly, you may also make a donation here.

---

I know, those three are indeed very small but are, compared to other mirrorless mounts, slow (for their sizes I mean)

If you compare to m4/3, you can't really compare apertures 1:1, since APS-C has both shallower DoF and better noise at the same aperture.

[torge]

Brilliant idea Douglas!
Thinking out of the box - be different!
Love it!

[thibs]

If you compare to m4/3, you can't really compare apertures 1:1, since APS-C has both shallower DoF and better noise at the same aperture.

I know but that's the point really.
APS-C has already 1 EV loss as of DOF concern.
Losing one more? Brrr

[froeschle]

If this thread is open for more or less crazy ideas, I have one.

As I understood it, this thread was (at least in the beginning) supposed to be a collection of facts and ideas about the forthcoming camera . By now, with already 11 pages, imho the main theme got a bit lost. But...

[...] A thin Pentax EVIL with a mirror. A fixed mirror. [...]

...is an interesting idea. Though, it might be not so easy to realize. There would e.g. be a lot of empty space inside the camera (rotated APS-C sized optical path, which also restricts the "thinness"). The main obstacle of this concept within this thread, however, is that "Pentax is going to release a mirrorless camera" .

EDIT:
According to http://photorumors.com/2011/01/28/pentax-d-fa-645-25mm-f4-lens-medium-format...nnounced-soon/

The rumored Pentax NC-1 EVIL camera will not be announced in February.

---

Last edited by froeschle; 01-28-2011 at 09:37 AM.

[Parliament]

If this thread is open for more or less crazy ideas, I have one.
[...]
If a fixed mirror is used to reflect the picture upward to a sensor placed where the focus screen are on SLRs, the registration distance will no longer determine the camera thickness.
[...]
-Include the mirror in the SR system and correct motions with both sensor and mirror. This should help cutting down the difference between in body and in lens SR.
[...]
The less modest version (with interchangeable viewfinders) you can find on my blog.

i really like your idea! especially the FF one on your blog, which also would work with smaller wideangle-lenses=)
(see M 20mm/4 vs. DA 21mm/3.2, which almost have the same dimensions)


*the sensor doesn't necessarily have to be at the same spot/have the same orientation as the focusing screen, you could put it nearer to the mount for example to save room behind the sensor.
I think it is optimal to maximise the yellow area in my drawing, under the condition that the sensor doesn't obstruct the light.
Or you could mount the mirror sideways and put the sensor stuff into the grip, thus saving more space for an EVF

*i've read once that the shutter should be as near as possible to the sensor to avoid diffraction at high speeds=small slit between shutter curtains. but i have no idea at what speeds/distances to sensor that effect is really relevant

*Tilting the mirror and shifting the sensor would do the same thing in terms of shake reduction, with different "side effects":
whereas shifting the sensor doesn't alter the focus plane, tilting the mirror does (as you've mentioned in your blog) As before, i have no idea how big these effects would be quantitatively, i.e. if you could even see in the photo that the focus plane was slightly tilted.
I think that, because of the lightness of the mirror and the "leverage"(move the mirror a tiny bit equals big shift of reflected light to sensor), tilting the mirror could be far more effective in SR-terms than shifting the sensor.
maybe it would make sense to implement mirror-SR only, as the current sensor-SR systems seem to add a lot of width and length to the sensor unit, which would make the whole camera bigger.

Now i hope that one day i would really have a FF-EVIL based on your idea, it is indeed very different and potentially better =)

Alex


EDIT:

As I understood it, this thread was (at least in the beginning) supposed to be a collection of facts and ideas about the forthcoming camera . By now, with already 11 pages, imho the main theme got a bit lost. But...

...is an interesting idea.

hmm i think there once was a thread dedicated to interesting ideas for the upcoming pentax "mirrorless" camera, maybe we should move the posts to this thread?

[Douglas_of_Sweden]

If you compare to m4/3, you can't really compare apertures 1:1, since APS-C has both shallower DoF and better noise at the same aperture.

Thanks you for pointing this out, so I didn't have to. I'm really not at all interrested in a m43 camera EVIL. A APS-C would be OK, FF would be a dream.

Brilliant idea Douglas!
Thinking out of the box - be different!
Love it!

Thanks!

As I understood it, this thread was (at least in the beginning) supposed to be a collection of facts and ideas about the forthcoming camera . By now, with already 11 pages, imho the main theme got a bit lost. But...

I simply took this thread as the best of the currently running Pentax-EVIL threads.

...is an interesting idea. Though, it might be not so easy to realize. There would e.g. be a lot of empty space inside the camera (rotated APS-C sized optical path, which also restricts the "thinness"). The main obstacle of this concept within this thread, however, is that "Pentax is going to release a mirrorless camera" .

Yes, but doesn't that really mean a camera without a mirror that move syncronised with the shutter, that is not a SLR?
Mmh, if we use a pentaprism instead of a fixed mirror, it will work without a mirror and actually really be a penta-x!

i really like your idea! especially the FF one on your blog, which also would work with smaller wideangle-lenses=)
(see M 20mm/4 vs. DA 21mm/3.2, which almost have the same dimensions)

I'm all with you!

*the sensor doesn't necessarily have to be at the same spot/have the same orientation as the focusing screen, you could put it nearer to the mount for example to save room behind the sensor.
I think it is optimal to maximise the yellow area in my drawing, under the condition that the sensor doesn't obstruct the light.
Or you could mount the mirror sideways and put the sensor stuff into the grip, thus saving more space for an EVF

I understand that, but starting with a 45 degree solution was simple to explain the concept. Having the sensor in a grip is an interresting idea.

*Tilting the mirror and shifting the sensor would do the same thing in terms of shake reduction, with different "side effects":
whereas shifting the sensor doesn't alter the focus plane, tilting the mirror does (as you've mentioned in your blog) As before, i have no idea how big these effects would be quantitatively, i.e. if you could even see in the photo that the focus plane was slightly tilted.
I think that, because of the lightness of the mirror and the "leverage"(move the mirror a tiny bit equals big shift of reflected light to sensor), tilting the mirror could be far more effective in SR-terms than shifting the sensor.
maybe it would make sense to implement mirror-SR only, as the current sensor-SR systems seem to add a lot of width and length to the sensor unit, which would make the whole camera bigger.

You are right that the mirror-only-SR would probably be more effective. But without being an expert, intutionally, and I am at least a physicist, a system moving both sensor and mirror should be abel to correct more complex camera motions. It should be like when several lens-elements move in a Canikon lens. With the advantage that there is no deviation from the optimal optical solution.

Now i hope that one day i would really have a FF-EVIL based on your idea, it is indeed very different and potentially better =)

He who lives will see.

[falconeye]

But there may be a way around this. A thin Pentax EVIL with a mirror. A fixed mirror.

Douglas,

your idea is very good. It shows how Sony failed miserably to make their SLT series small cameras.

However, I wouldn't be me if I wouldn't point out some weak points with your idea too :=)

First, I think the mirror would have to be slightly larger than SQRT(2)*16mm if it must support fast lenses as well. The crucial ray is at the bottom for an f/1.4 lens.

Second, one of the real appeals of mirrorless camera is the short registration distance. Part of the reason why the Leica M series is in a hall of fame is the stunning performance from some of its short focal lenses. With mirrorless, the sharpest lenses would be around 20mm rather than 40mm.

[Douglas_of_Sweden]

Douglas,

your idea is very good. It shows how Sony failed miserably to make their SLT series small cameras.

Thanks!

However, I wouldn't be me if I wouldn't point out some weak points with your idea too :=)

First, I think the mirror would have to be slightly larger than SQRT(2)*16mm if it must support fast lenses as well. The crucial ray is at the bottom for an f/1.4 lens.

I'm aware of that, and considered making that more obvious in the figure, but for sake of simplicity and not to obscure the basic idea that there is a way around the registration distance. I'll add a note in the blog. It has to be larger also for it to be usefull for SR and for the ide to turn every lens into a tilt lens to work. But I think that there is enough space in the design for that, and if it would be necessary to add a few mm, it is still a small camera.

Second, one of the real appeals of mirrorless camera is the short registration distance. Part of the reason why the Leica M series is in a hall of fame is the stunning performance from some of its short focal lenses. With mirrorless, the sharpest lenses would be around 20mm rather than 40mm.

Yes, that is true. And another appeal is the possibility to add different adapters that allow people to mount old glass, and here we would be limited to what can already be adapted to the K-mount. Part of me would be happy if Pentax launched a APS-C or larger EVIL with short registration distance since it would allow me to use all my old Konica Hexanon glass. Another part would prefere them keeping the K-mount. But the point here was to demonstrate that it is possible. Thin EVIL does not equal that the K-mount must be abandoned. And Pentax is a very conservative company in love with the K-mount and the 45.46mm.

Now I don't believe this will happen, it is probably to much "out-of-the-box", but it would work.

[falconeye]

Now I don't believe this will happen, it is probably to much "out-of-the-box", but it would work.

Well, it is out-of-the-box within the world of SLR. But AFAIK some very slim P&S (and some mobile phone cameras I believe as well) use your idea.

I don't know why Sony SLT cameras aren't thinner as they use the same idea. Maybe, Sony wanted to keep the good old AF module made for a thick body. So, I tried to compute the minimal mirror depth for a finite f-stop and it may explain why SLT cameras aren't smaller. I may have made a mistake but this is what I got:

- APSC and f/1.4: 27.3mm
- FF and f/1.4: 29.5 mm

And this assumes the mirror is placed right at the mount which excludes lenses with protruding rear elements which do actually exist. So, in practice it would have to be larger than 30 mm and you win less than 15mm, probably 10mm would be the practical limit. If you now compare the SLT55 and A550 on dpreview (use the photos), you'll see that the SLT55 is about 10mm thinner indeed. It doesn't show up in the body dimensions but in the image you'll see that the eye piece protrudes in the SLT55 as it has the slimmer body. That seems to be about as much as your idea is capable to perform.

Sorry about that Douglas, but your idea didn't withstand my second look

____
Note. The formula I got is:

x = 2 R^2 / (R + 2 N R - H N)
where N = max. supported f-stop number, H = sensor height, R registration distance, x = mirror depth for a 45° mirror (mirror height is Sqrt(2) times this value). The approximation x = H holds true for N=R/H=2.8.

The camera can be made a bit thinner by tilting the mirror more than 45°. But not very much though.


P.S.
I agree this is OT in this thread. But I can confirm that the forthcoming mirrorless cameras are not being talked about at February CP+ in Japan. So, why not have some in-the-break chat

[Fogel70]

Well, it is out-of-the-box within the world of SLR. But AFAIK some very slim P&S (and some mobile phone cameras I believe as well) use your idea.

In these cases the sensor placement is not used for making the camera body slimmer, but to fit the optics within the camera so the lens will not protrude from the camera.

Using a mirror for keeping a longer register distance might be a good short term solution for making a smaller camera, but in the long run cameras with shorter register distance will probably be much more successful as wide angle lenses can be made smaller.

BTW, I think Pentax was one of the first to use folded optics in P&S cameras.

[gazonk]

Well, it is out-of-the-box within the world of SLR. But AFAIK some very slim P&S (and some mobile phone cameras I believe as well) use your idea.

The Pentax W-series cameras avoid a protruding zoom this way: The zoom is placed perpendicular to the outer lens inside the body. I think they use a prism.

[falconeye]

I still stand by my OP.

But I am not sure what to think about photorumors NC-1 rumor anymore.

Because Pentax NC-1E is an existing product. But one never knows...

[glanglois]

I still stand by my OP.

But I am not sure what to think about photorumors NC-1 rumor anymore.

Because Pentax NC-1E is an existing product. But one never knows...

This is schwarz, Maybe in blau?

[Mistral75]

There is also the new Kenko system camera, with C mount lenses and a tiny sensor:

Kenko C mount system camera

[ogl]

I still stand by my OP.

But I am not sure what to think about photorumors NC-1 rumor anymore.

Because Pentax NC-1E is an existing product. But one never knows...

NC-E 1 is not NC-1