;-)
without APG it wouldn't be possible to get that amount of panos in this time!

Hellkeeper wrote:

One of my latest Gigapixels. Have fun! ;-)

http://www.gpix.at/Gpix.at-Gigapixel_gp … ,gpix.html

Nice!

best, Klaus

inMotion wrote:

Hellkeeper wrote:

;-)
without APG it wouldn't be possible to get that amount of panos in this time!

have any gigapixel workflow tips you'd like to share?

May i drop in here?

1) get a fullframe camera (you already did afaik).
2) use prime-lenses (saves time in post-processing CAs and so on).
3) use a very sturdy head and tripod.
4) use a very precise head.
5) us a precise and fast head.
6) use a sturdy, precise, fast head providing an intelligent controller - preferably detachable from the head - which writes xml files.
7) use the genial APG-xml-import. It provides very, very much very clever options . . . once you got used to it i mean . . .
8) plan your shooting! Gigapixel-shoots take their time - even using a relatively fast moving head. You will definitely find yourself facing unexpected problems . . .

9) take things easy.

best, Klaus

Last edited by klausesser (2012-05-30 17:04:43)

inMotion wrote:

Thanks Klaus! Seems like getting the panogear head was not the best idea...

Well - related to the price it´s a very clever investion! Depends on what you want to do/achieve:

For private/hobby use it´s absolutely great! The time of a shoot doesn´t count that much here.

For commercial use it might be too limited in it´s features and also too slow. Depeds on what you shoot.

Nevertheless it´s a fine head!

best, Klaus

Depending on what camera you are coming from, the D800 may not necessarily increase pixel count. The Canon 7D has a higher pixel density, and the Sony A77 higher still.

Shoot the same scene with the same focal length lens, and both those will give you more pixels than the D800 will, the Sony significantly so.

Of course the D800 won't need as many source images, but if you want to shoot with the lowest possible number of source images, then nothing gets close to the Phase One DF/IQ180 combination.

For a full spherical pano around 4GP -

D800 - over 200 shots.
5D Mk II - over 300 shots.
7D - around 400 shots.

Phase One DF/IQ180?

95.

Hi Ed -

I think we're much in agreement, but coming at this from slightly different angles. Importantly though, the following statement simply is not true, and misses what are the fundamental factors surrounding this entire discussion.

in order to get the biggest pixelcount, well use the smallest workable sensor (micro 4/3 in my book but on that you may differ) and the longest lens

Actually, only two things drive the resolution of a panorama for a given particular field of view and output projection. The focal length of the lens used, and the pixel size on the sensor.

Focal length of the lens.
Pixel size on the sensor.

That is it.

Here are a few examples of pixel size (all in microns)

1D Mk IV - 5.66
5D - 6.41
D800 - 4.89
7D - 4.34
Sony A77 - 4
IQ180 - 5.45

If the only thing you're interested in is size, then pick the appropriate combination of camera with a small pixel size (NOT necessarily a crop sensor) and availability of a lens with a small field of view (note - NOT necessarily longer focal length, I'll come onto that later).

When it comes to resolution of the output panorama, sensor size is an irrelevance.

Slap 300mm lenses on both an IQ180 and Canon 1D Mk IV, shoot the same panorama, and to all intents and purposes, you end up with the same resolution, because to all intents and purposes, the pixel size on both sensors is the same.

Example, 300mm lens shooting a full spherical panorama:
1D4 - roughly 60 gigapixels.
IQ180 - roughly 65 gigapixels.
(the relative output ratio will be equal to the ratio of the squares of the relative pixel sizes)

What sensor size determines is how many shots you need to take for a chosen field of view for the panorama. And that's it. It doesn't impact anything else.

And the reason of course is that for a given focal length, the field of view at the sensor is dependent on the size of the sensor.

In FF DSLR terms (which most people will be familiar with, so I think it makes sense to do the comparison in this way), a 300mm lens on a 1D4 will give the field of view equivalent to a 390mm lens, whereas on an IQ180, a 300mm lens will give the field of view equivalent to a 192mm lens.

The ratio of the theoretical number of images required between two systems will be equal to the ratio of the squares of the “effective” focal lengths used. 390^2/192^2 = 4.1.

Which is why, for our hypothetical 360 degree spherical pano shot with a 300mm lens, the IQ180 will do it in less than a quarter of the number of shots compared to a 1D4.

In the example print I linked to, believe me - when it's 45 degrees outside, shooting quickly has its benefits.

But there are other panoramic applications that would massively benefit from being shot with a camera with a large sensor. Just one example off the top of my head: With an IQ180, you could shoot a full spherical panorama of approximately 30 gigapixels from the center spot of a football field while the players are off having their half-time break. You would not be able to shoot the same thing with a D800 and maintain the required level of quality, because you'd have to be taking a shot roughly every 0.5 seconds.

Regards,

Gerald.

Artisan S. wrote:

Now factor G will be my nemmesis Klaus......so I'll wait patiently until well until i'm ready to invest and have fun shooting pano's while doing so....I hope you do not object to that......no offence by the way, I guess if we were to meet, you and I would tick a lot more alike then you think Klaus (personally I love perfection and my best camera is a Fuji GX680 behemoth bought for a few 100 euro and with a stunning picture quality when scanned (drum preferably), but it is heavy and vulnarable......not an outdoor machine allthough I took into the great outdoors.

Greetings, Ed.

Hi Ed!

I agree to all you said!  I prefer to use a 5D2 instead of the H3D/39 - using a MF for panos can rise some DOF problems and others . . . In fact for me it´s not vital at all to have a lesser amount of shots - using the 5D2 on Josef´s Panoneed head definitely is fast enough and it´s a well trained team.
Because i nearly always shoot bracketed between 3 and 12 (PromoteControl) steps the greater depth the back gives me i don´t need here. So i keep both worlds apart from each other - "classical" photography and panorama-photography.

No problem to match the skills of the H3D with the 5D2 making the adaeqate amount of shots with using bracketing.

For some years i used a 20D on a Merlin - very fine combination for shooting panos even in big dimensions! But i´d sugest prime-lenses - i use old Nikon Ai lenses . . they work great on 20D or 5D2! (i don´t like zooms - being a bit old-fashioned here )

best, Klaus

leifs wrote:

I'm not a player in the large-sensor game. But since this forum is about APG and panoshooting I want to add a angle of view:
- in panos vignetting is a problem
- prime lenses also has vignetting
- large sensors have more vignetting problems
- full-frame lens on a crop-sensor reduce the problem

to examplify this: Photozone has tested the Zeiss Makro-Planar T* f/2 on a fullframe and on a APS-C sensor
FF 2.23 stop vignetting @f2
APS-C  0.6 stop vignetting @f2

http://www.photozone.de/canon_eos_ff/53 … ff?start=1
http://www.photozone.de/nikon--nikkor-a … rt?start=1

This is not surprising.
The problem is small (?) if you use camera and lens of the same make, as the vignetting is corrected in-camera before the image is saved to the memory-card.
Morale for pano-makers: check the house-lens vignetting characteristics before shelling out your cash.

leifs

Hi leifs!

Vignetting a problem? Yes. More or less . .
Prime-lenses also show vignetting? Yes. More or less . . Usually it´s not evident on primes @11/16.
Large sensors have more vignetting problems? No. It´s the lenses . .
Fullframe-lenses on crop-sensors reduce the problem? Definitely yes - of course they do. But you need to do more shots.

Basically ALL lenses "have" vignetting. Depends on the aperture you use. Good primes @11/16 show very few vignetting.
Basically (!) APG is able to compensate that completely. I said "basically" because over the years i never had issues with APP/APG
when i used my Nikon- or Zeiss-primes on the 5D2. Actually i very well have issues sometimes - depending on the light-situation. I believe that APG sometimes (!) has issues with areas of very bright and intense light in the current version when there´s delicate color-fading in skies.

Testing the same camera/lens with old version of APP on an older machine there´s no vignetting-issue at all!
If someone could also give that a try i´d be interested to know whether i´m wrong . .

Also basically you can of course compensate vignetting very effectively in modern RAW converters (don´t even need to use RAW to do so).

best, Klaus

Last edited by klausesser (2012-06-12 14:37:47)