Quote:
"Yes, am sure your bored with it all, but this needs to be resolved! the GH1 is not as people think 2x crop factor, well thats my theory anyway. Note, I succesfully overlapped almost pixel for pixel a 4:3 and 16:9 image shot from tripod and then overlayed in PS. If the crop factor was changing due to crop diaganol then they shouldnt of overlayed as the 4:4 diaganol is longer than then 16:9. Therefore this leads me to believe that the camera sensor should be considered at its max for the 16:9 width ie 19mm aprox and its max height, 13.5mm. 23.5mm diaganol, so more like 1.85 actually. This is possible as the 4:3 ratio then behaves as if the two sides are cut-off to provide this ratio.
The following though shows just how wide the 14mm setting is on a GH1 as compared to full frame and aps-c Canon. This was not my shoot but in 16:9 format the crop is definately better than a 17mm at 1.6 ie 27.2 mm and a little bit less than the 24mm full frame. See for yourself, and it does look as if the GH1 is around 25mm-26mm."
What do you think about true crop factor 16:9 / GH1 ?
Must be same as GH2
And what is the crop factor on the GH2? I've heard 1.86, 2.0, and when I run the math myself I come up with around 2.2...
It depends on aspect ratio. 16:9 aspect ratio on GH(1)2 is around 1.86. 4:3 is 2x.
Edit: I never did the maths myself, just gathered info from around the place.
How to calculate for 2:35:1 crop ?
This is an interesting topic. It seems like there is still a confusion about the crop factors of GH1 and GH2 and GH3 and BMC when shooting video at 16:9 aspect ratio. There is no stable information on internet ... just individual opinions.
There's a BIG difference between 1.86x and 2x . So which one is it?
It depends on how you measure it. When comparing to 35mm still images, which have a 3:2 aspect ratio, even full frame 16:9 images have to be cropped. There really isn't a "correct" way to make this comparison when aspect ratios are different. Which dimension should be used? Diagonal, vertical, horizontal? One could make the argument that for video, the horizontal dimension is most important. Typically, though, the diagonal is used (probably because it affects lens coverage). Depending on how you go about the comparison, you'll get these different numbers. For example, if you just fit a 16:9 frame within a FF sensor - is it a ratio of 1:1 (using the horizontal dimension), or some other ratio (which it would be if the diagonal or vertical dimension was used)?
I indeed tried to compare pics of the FF camera vs GH1 while using two different lenses with focal lengths in the proportion of exactly 1:2 (so 25mm vs 50mm or 100mm vs 200mm). The pics of the same, static subject were shot from exactly the same spot using tripod.
Using 3:2 ratio on the GH1, its crop seemed to be slightly beyond 2.0 mark, 1.9 or slightly less.
However, I have not tested anything in the optical laboratory and on the real test charts. The effective optical focal length of the single lenses may also have some tolerances compared to declared one (like 50mm lens could actually perform as 49,3mm or so) and besides all, @kronstadt the difference between 2.0 and 1.86 is indeed not so big- it's barely noticable, put the tripod just a touch closer and you have 2.0 crop. To be honest, I don't really care for the mathematical values, it is nearly crop 2.0 and that's it speaking only for my own.
If comparing 35mm film (24x36mm) with a GH1 or GH2 at 3:2 aspect ratio (11.9x17.9mm) the ratio comes to 2.01:1.
If you compare at a 16:9 aspect ratio 35mm FF film (36x20.25mm) with a GH1/2 at a 16:9 aspect ratio (18.8x10.6mm) it turns into a ratio of 1.91:1.
I can't figure out where the 1.86 ratio comes from - maybe it's for yet another aspect ratio (maybe 4:3)
@cbrandin What... you are correct.
I just need a number to keep in my head so that I can multiply the mm of the lenses when shooting GH1 video in AVCHD (I guess that's the 16:9 aspect ratio). 1.86 came from other threads, so I got confused- what thatnumber is 2 or 1.86
I've just tried to test it again
GH1+Canon FD 100mm f/2 vs Nikon D800+Nikkor Ai-S 200mm f/4
This time on sturdy-like-hell tripod and object quite far away.
and GH1's crop in 3:2 ratio is just slightly beyond 2.0 mark, the GH1 had again very, very slightly wider FOV than the D800 (just as when I tested it before). Something like 1.95 or higher, definitely not 1.86
@cbrandin The number 1.86 probably comes from the fact that GH1's sensor covers a slightly larger image circle than the native 4:3 image aspect ratio of an usual µ4/3 sensor. Thanks to this fact, the GH1 14 megapixel sensor is capable of recording images in native aspect ratios of 4:3, 3:2 and 16:9 with no cropping, so it also has the same angle of view and maximization of pixel count in any particular format.
So, effectively, the GH1 sensor has 14 megapixels, but it is never used to its full capacity (max. up to 12.1). It also means that the sensor of the GH1 (and also of the GH2) would reproduce in its full size 1.86x magnification, but effectively - it never does it, so the crop is also not 1.86 but quite near by 2.0
@kronstadt so, the answer is kind of long, but important for you is the fact that you can count crop 2 as rule of thumb both for the GH1 as well for the GH2.
If anybody is interested in downloading out-of-camera JPEG, send me PM please.
All m43 bodies have 1.85x for 4:3.
All m43 bodies except Gh1 and Gh2 have 2.08x for 3:2 and 16:9.
Gh1 and Gh2 have 2.01x for 3:2 and 1.91x for 16:9.
What does it mean? Not much. Those crop factors are "around" 2x.
You are right about the 4:3 ratio. The GH1/2 dimensions are 17.3x13mm for a 4:3 image. That means the ratio against a 4:3 image within a FF 35mm frame (32x24mm) is 1.85:1.
For video, it seems to me that what is most meaningful is the angle of coverage you can expect along the horizontal axis in a 16:9 aspect ratio frame. Thus the most appropriate ratio to use for calculating relative lens focal lengths for 35mm vs. 4/3rds would be 1.91:1.
A range from 1.91 and 2.08 is 9% difference. It's a glass half empty or half full.
I'd go with 2x to keep the math simple. That might sound stupid but there should be some margin of error. Those lenses labeled with same focal length have slightly different FOV. So 1.91, 2.01, 2.08 don't make much difference to me.
Exactly, there is indeed slight difference in FOV between different lenses of same focal length. Therefore, I agree with @stonebat and tend to keep mathematic thinking simple: 2x
And I would say it does not sound stupid at all what you wrote, as in the field conditions it is negligible error....put the tripod just a tad closer or further away and there we got it.
Speaking only for my own, it was good to have had this thread and the debate about the crop.
@tetakpatak @stonebat @cbrandin thank you all for taking the time to explain this in technical detail. For video, I'll keep in mind 1.91-2.08 (2X to be simple)
Those lenses labeled with same focal length have slightly different FOV.
That actually explains a lot too.
It looks like you're new here. If you want to get involved, click one of these buttons!