Keep in mind this is an extreme test to demonstrate the rolling shutter effect, no one in their right mind would pan like this. However, there are plenty of real life situations when the jello effect suddenly becomes a problem, a fast train or car passing will have leaning windows and elliptical wheels, or watching a chase scene with moving camera and subjects will make you dizzy. Until we get global shutters, the jello effect is CMOS sensor’s biggest Achilles Heel. Screen grabs, updated with 100% crops after the break
Foundry has a plug to fix this problem in aftereffects and nuke, but it ain’t cheap.i movie 11 also offers a jello fix however it drops the bit-rate substantially.
Obviously the Sony Nex 7 is not in the same league as the Canon 5D3 but should have the edge in this comparison because of the smaller APSC sized sensor.
The Canon 5D3 was set to 1080/24P, ISO 100, Shutter Speed 1/50th, F8, WB 5600K, Picture Style Neutral, the Sony, 1080/24P, ISO 100, Shutter Speed 1/50, F8, Picture style Portrait, sharpening, contrast, saturation turned down to 0 on both cameras.
Surprise, the 5D3 kicks Nex 7’s gluteus maximus, I sure didn’t expect that from the FF sensor, either the Canon is very good or the Sony just terrible, your pick :)!
Take a look at the screengrabs and notice the moire (circled in the doubles) we get with the Nex 7? A real problem shooting with the 5D2, you will notice it in my Brooklyn bridge video shot with the 5D2. The 5D mark III doesn’t have it, this is HUGE!
stay tuned for more, I am very excited to start shooting with this baby!
100% crops (white border)
If you find this information helpful, please consider helping me maintaining this blog by purchasing your gear through my product links to B&H. It will cost you nothing and allows me to keep adding! Or consider making a direct donation using PayPal
Thanks for the upload!
I just tested my Nex-7 for panning and results varied wildly with the settings. @ any MPG4 size (12M or 3M avi) panning is jerky. Just big jerks, just not smooth. Like squeegee jerky for lack of better words. Switching to AVCHD it is interesting to see that images are still jerky at all 25fps settings (I’ve got the European version). However, switching to 50fps all jerking immediately stops and the image is sharp and nicely smooth while panning, at least for my uses and checking in the viewfinder. This is the case for both interlaced and non-interlaced settings (50i & 50p).
So unless your Nex-7 in this test was @ 50/60fps you don’t see what the Nex-7 is really capable of.
Rgds, Ulrik
I agree with DaveT.
The Sony Nex-7 has the possibility of 60fps at 1080P. I believe if you take advantage of this feature that is not found on 5D, the Sony camera should show a quite big improvement regarding jello effect.
Please also include Sony 60fps mode in this comparison, that would be the ultimate test!
Thanks
I’d like to see a test at 60p. The NEX7 shines at this frame rate. Though it may be moot since the 5d3 can’t do 1080 60p.
Pingback: sonyalpharumors | Blog | Sony sells the 24 Megapixel sensor to Nikon (new D3200 coming)
Hi Andrew, true, this test is not very scientific
and I am sure the is a difference in force, but I don’t think it has enough practical impact in terms of effort/load to impact the result significantly, the distance between the two sensors is about 5 inches and the connection is rock solid.
Thanks for the comparison and I’m no tech guy but wouldn’t the Nex-7 be negatively affected by being higher up than the Canon?
I.E. I don’t know what you’d call it, but since the Nex-7 is higher up would there not be greater input force exerted upon it as it is further away from the fulcrum point of the tripod?
Therefore the force of your tripod movement on the Nex-7 is greater than that exerted onto the Canon, so basically even thou they’re being moved at the same time, it is nonetheless like the Sony is being swung around faster and with more force than the Canon?
As I said, no physicist and I can still see the Canon being better but maybe the test isn’t quite accurate? As in they both need to be level and from an equal distance from the middle of the tripod for true accuracy?
Pingback: Rolling shutter jello comparison: 5D Mark III vs Sony Nex 7 | CanonWatch
…I am very content. We are mainly working on textile / pattern films. 5DmIII seems to be best choice. Even if GH2 / D800 / whatever might have better resolution, the 5DmIII simply does best for us. In general, I am very pleased with the output when shooting neutral style and adding sharpness + color details in post. The files from the card are a little bit soft, but when you hit the sharpness trigger in post, it is like a real revelation. And that is how it is supposed to be, the camera delivers a beautifully gradable material. Hopefully, Technicolor will release a dedicated 5DmIII profile, soon.
thanks!
nice review!
lol, thanks Cacasodo
Nice analysis, Mike. I almost puked, but nice analysis.