[slip]

If full frame cost is an issue, couldn't pentax meet half way and get a sensor made that is between 1.5 and FF like a 1.25 factor?
just a thought, not sure about the technical aspects of it/.

any options welcomed, of course

[Adam]

Like the Canon 1D mark IV? They could use APS-H, but I think that's even less likely than a 35mm sensor would be. Plus, all the Pentax reps have been talking about full-frames lately, so one's bound to be made available in the coming year(s).

[ismaelg]

While APS-H has its technical merits, in my opinion it will hurt the brand because it would be seen like "we FAILED to do full frame and we are a decade late for APS-H"
Just my opinion...

Thanks,

[ElJamoquio]

At one point I'd suggested doing a 24x32 4:3 (same height as full frame) or 25.9x34.6 (same diagonal as full frame). For me it's way easier to start with the 4/3 and go to prints.

[rvannatta]

While APS-H has its technical merits, in my opinion it will hurt the brand because it would be seen like "we FAILED to do full frame and we are a decade late for APS-H"
Just my opinion...

Thanks,

I just keep wondering why I want ff. The bodies-- if the other guys are being fair about the price seem to cost double or more than the half frame
and what do I get. I've spent thousands of dollars on long lenses that effectively won't be as long any more. Unless the image is morethan
twice as good as what I have, I'll have to crop half of it off to get the same image I can get now....

Sensor resolution seems to be pressing the resolving capability of the lenses we have---will a bigger sensor resolve this issue???

The 35mm is only an accident anyway. Movie film used to be 70mm, and a long time ago some wag figured out that if they split the 70 mm film
in half they could reduced costs by half, and it worked so well they did it again and made 16mm movies and then again and made 8 mm movies

I recall reading of the 35mm cameras being called 'miniature' cameras because they used chunks of 35mm movie film instead of legitmate
full frame 4x5 sheet film that would be insisted upon by an legitimate photographer.

I suppose most of the news photographers (and others as well) who believed that no decent photos could be taken with anything smaller
that a Speed Graphic with 4x5 sheet film are mostly dead by now. Indeed somewhere around here, I have the correspondence course work of the NY Institute of PHotography from the late 1940's and it is absolutely adamant that no legitimate photo can be taken with less than a 4x5 plate.

I guess I'm missing something here. Why aren't we demanding 4x5" sensors.? That's the real full frame.

Could it just be that 'what's inside the box isn't as important as IQ. I'm not sure there is any such thing as 'Full frame' --- just a bunch of historic
sizes.

[sterretje]

I've spent thousands of dollars on long lenses that effectively won't be as long any more.

That's a double edged sword; see the advantages at the wide side where one can get away with a 24mm (FF) instead of a 16mm (APSc) for the FOV.

[RioRico]

I've spent thousands of dollars on long lenses that effectively won't be as long any more.

Incorrect. I'll flip this around: I put a T2-mount 400mm lens (built for FF) onto my half-frame (HF) Olympus Pen-FT. (Half-frame is about the size of APS-C). I thought, OH WOW IT'S A 600MM LENS NOW! Wrong. It just produced 400mm images with the sides chopped off. Moving a lens from a FF to a HF cam doesn't stretch it; moving it from HF to FF doesn't shrink it. No Procrustean beds are involved.

That's a double edged sword; see the advantages at the wide side where one can get away with a 24mm (FF) instead of a 16mm (APSc) for the FOV.

When I put the Kiron 24/2 or Tokina 21/3.8 or Zenitar 16/2.8 on my ZX-M or MV1 or K1000, those lenses are WIDE! But not quite as wide as the DA10-17 or Tamron 10-24 are, on my K20D. FF allows exploitation of those legacy wides. APS-C gives us wider lenses that are MUCH cheaper than their FF equivalents were, back in the day.

[stanislav]

When I put the Kiron 24/2 or Tokina 21/3.8 or Zenitar 16/2.8 on my ZX-M or MV1 or K1000, those lenses are WIDE! But not quite as wide as the DA10-17 or Tamron 10-24 are, on my K20D. FF allows exploitation of those legacy wides. APS-C gives us wider lenses that are MUCH cheaper than their FF equivalents were, back in the day.

Well I do not have som much experience with the price of legacy wides. However there are some FF "wides" that are as wide as they get (rectilinear - Sigma 12-24, or some of circular fisheyes).

Unfortunately I do not have Zenitar 16/2.8 (although I would like to have one), but it should be a fisheye with 180 degrees diagonal coverage. So in a FF it should give similar picture as DA 10-17 @ 10mm (at least regarding FOV).

However it is a bit misleading when comparing the focal length of fish-eye lenses with their FOV. As a matter of fact I still didn't get a hold of what focal length on a fish-eye lens describes actually. On a rectilinear lenses it can be use to calculate FOV knowing the size of "sensor" or "film". Moreover on rectilinear lenses a if a "crop factor" is 1.5 (FF size / APS-C size) than it means that if you want a same FOV on both cases the focal length for APC-C has to be 1.5 times smaller compared to FF when lens is focused to infinity (i.e. 45mm on FF gives same FOV as 30mm on APS-C or 10mm on APC-C gives same FOV as 15mm on FF). Now this is an "ideal" comparison with "perfect" lenses, no distorsion etc, a reality is a bit more complex, but for ordinary usage it is a good approximation.

... or did I get it all wrong?

[RioRico]

Well I do not have som much experience with the price of legacy wides. However there are some FF "wides" that are as wide as they get (rectilinear - Sigma 12-24, or some of circular fisheyes).

I recall reading various accounts that the prices of legacy rectilinear UWAs, adjusted for inflation, were significantly greater than their modern counterparts. Alas, I'm not in a position right now to find those accounts. I'll have another excuse in a moment.

Unfortunately I do not have Zenitar 16/2.8 (although I would like to have one), but it should be a fisheye with 180 degrees diagonal coverage. So in a FF it should give similar picture as DA 10-17 @ 10mm (at least regarding FOV).

And alas, I'm also not in a position to shoot comparison tests of the Zen16 on FF with the DA10-17 on APS-C. That's because I'm on a long road trip (departed yesterday) and I didn't bring the Zenitar. Even so, FF fisheye ZOOMS comparable to the DA10-17 were much more expensive.

... or did I get it all wrong?

Not really. Both frame-filling and full-circle fisheyes are tricky to compare with allegedly similar glass, partly because various designs use different geometric projections. So FEs rated about equal on FOV may project rather different images. Fisheye and ultrawide makers share a problem with mapmakers: How to adequately project a 3D world onto a 2D frame. There are many different ways to fry that fish.

[boriscleto]

Fisheye and ultrawide makers share a problem with mapmakers: How to adequately project a 3D world onto a 2D frame. There are many different ways to fry that fish.

Oooo...Thread Hijack


xkcd: Map Projections

[RioRico]

I'm a Goode Homolosine guy, although I own terrestrial, lunar, and astro globes, and a solar system orrery. I have constructed a Dymaxion (from a cardstock print) but who hasn't? The Hobo-Dyer and Gall-Peters both suck. The Mercator is vastly misunderstood; it's not a map, it's a navigational chart. Too bad the xkcd exhibit omits a south-on-top model -- maybe it's hard to stereotype. Up Australia!

[dadipentak]

Does anyone make a fold-able globe--you know, like a paper lantern?

[RioRico]

Does anyone make a fold-able globe--you know, like a paper lantern?

Yes. Don't ask me for details -- I got one many years ago, and it's hundreds of miles away at the moment (I'm on the road). But it's just like a paper lantern, with metal hoops at various latitudes.

[ElJamoquio]

Does anyone make a fold-able globe--you know, like a paper lantern?

Dymaxion pretty much is a foldable globe, no?

[stanislav]

I recall reading various accounts that the prices of legacy rectilinear UWAs, adjusted for inflation, were significantly greater than their modern counterparts. Alas, I'm not in a position right now to find those accounts. I'll have another excuse in a moment.

I guess you are right about this one. Moreover I guess that there were not many comercial UWAs back in 80's to begin with. It was first in computer era that making such lenses become both possible and feasible.

And alas, I'm also not in a position to shoot comparison tests of the Zen16 on FF with the DA10-17 on APS-C. That's because I'm on a long road trip (departed yesterday) and I didn't bring the Zenitar. Even so, FF fisheye ZOOMS comparable to the DA10-17 were much more expensive.

I am a lucky guy to have both DA 10-17 and F 17-28. What I can say is that on APS-C the covered angle is equal on 17mm. Moreover on FF the covered angle of F17-28 at 17mm seem to be equal to the DA 10-17 at 10mm. I have also seen a comparison of coverage (DA on APS-C and F on FF) and the photos are identical. I tried to find the link to the webpage where this comparison was done, but I just couldn't find it now. Even in this case I really do not know how much the F 17-28 cost when it was new, but it couldn't be cheap since there are not many out there.

Not really. Both frame-filling and full-circle fisheyes are tricky to compare with allegedly similar glass, partly because various designs use different geometric projections. So FEs rated about equal on FOV may project rather different images. Fisheye and ultrawide makers share a problem with mapmakers: How to adequately project a 3D world onto a 2D frame. There are many different ways to fry that fish.

I agree partly with you on this one. The problem of rectilinear lenses is not the projection in itself but rather the edge resolution, chromatic abearation, ghosting and flare. The wider the lens the bigger the problem is. Basically the only thing given is a projection. In theory if you want a rectilinear projection of a 180 degree circle (180 degree coverage) with rectiliner lens than the lens has to have the focal length of zero. That way the projection of angles is kept and lines parallel in reality are parellel on picture (projection) too. Regarding the fisheye lenses this is not the true, and a coverage of even more than 180 degrees is possible. However with heavy distorsion at the edges (at 180 degrees circular fisheye an infinitly long straight line appears as a half circle on projection, but that is the price to pay).