[paranoia23]

jello effect and rolling shutter are 2 different things ... arent they ?!

[GibbyTheMole]

jello effect and rolling shutter are 2 different things ... arent they ?!

Jello effect is caused by a rolling shutter. The way i understnad it is the rolling (or electronic) shutter writes one line of pixels at a time. If there's camera or subject movement during the exposure, the "stacks" of pixels don't line up exactly, and it creates a warping effect called the jello effect.

[hnikesch]

I get 90% jello-free shots up to 105mm.

Nice test and nice to know the results

Hans

[paranoia23]

well if u own a k30 or a k01 ... take 2 clips with the same movement (with and without shake reduction) and compare the jello effect.

jello is caused by rolling shutter ... yes.. but the digital shake reduction in these 2 cams make it worse ... alot.

[GibbyTheMole]

jello is caused by rolling shutter ... yes.. but the digital shake reduction in these 2 cams make it worse ... alot.

Didn't know that... I'll try turning off the SR & see what happens. Thanks for the tip.

[drougge]

well if u own a k30 or a k01 ... take 2 clips with the same movement (with and without shake reduction) and compare the jello effect.

jello is caused by rolling shutter ... yes.. but the digital shake reduction in these 2 cams make it worse ... alot.

You seem to be talking about filming, everyone else here is talking about still images. Shake reduction should make jello caused by camera motion less in still images, and should have no effect on jello if it's caused by subject movement.

Jello is probably more often used to talk about filming, but I think that's because it's rarely a problem in still photography. (But focal plane shutters have the same problem, they're just much faster than the Q electronic shutter, and therefore less likely to show it for normal movement speeds.)

[snostorm]

Personally, I've seen very few shots where the rolling shutter effect has ruined a shot in and of itself that would not have been spoiled by blur anyway. I can pick out shots where rolling shutter caused slight waviness in lines that I know should have been dead straight, but these were at subject distances that could have been effected by atmospheric effects, and only were evident at very high magnification -- like 300-400%. In these shots, these anomalies did not materially negatively effect the shot -- when viewed at reasonable output size might only be seen if I was specifically looking for it -- nobody else would likely notice it.

I like that on photo fora like this we can concern ourselves with digital photographic minutiae, but it's always good to put them in context with examples and 100% crops of effected areas with exif attached and with shooting technique explained. This way, each can make their own decisions whether this should be a concern in the way they shoot.

Scott

[drougge]

Sometimes it can be a real problem. John M Flores demonstrates:




That's at 1/800, not a particularly slow shutter speed, I think it would have been decently sharp with a better shutter.

But I've never had a problem. I don't tend to shoot things that move much.

[snostorm]

Hi drougge,

I could be wrong, but the post says that this was taken with the FE, so I'm guessing the camera was mounted on the handlebar, and it's likely that he was actually driving at the time this was taken, so with road shock and engine vibration, this can hardly be considered normal shooting conditions for most.

I never stated that it can't happen, I just don't know if it's something to get overly concerned with.

This is one of the few I've shot where the effect is really visible. At 1/160, this one would have been blurred regardless. Notice the ears and the rest of the body are essentially not effected. The rolling shutter just allowed the weird distortion. If I saw that it compromised a number of otherwise good images, I'd be more concerned.

Scott

[paranoia23]

You seem to be talking about filming, everyone else here is talking about still images.

oh sorry i misinterpreted the topic then ^^

[drougge]

This is one of the few I've shot where the effect is really visible. At 1/160, this one would have been blurred regardless. Notice the ears and the rest of the body are essentially not effected. The rolling shutter just allowed the weird distortion. If I saw that it compromised a number of otherwise good images, I'd be more concerned.

But at 1/8000 (or whatever you consider enough) it would have been sharp (and the motorcycle shot too I'm sure, even though I've never tried anything like it). But not on the Q, because the shutter speed doesn't matter for this. So depending in shooting preferences it really can be a problem, even though it isn't for you or me.

[GibbyTheMole]

But at 1/8000 (or whatever you consider enough) it would have been sharp (and the motorcycle shot too I'm sure, even though I've never tried anything like it). But not on the Q, because the shutter speed doesn't matter for this. So depending in shooting preferences it really can be a problem, even though it isn't for you or me.

From what I've read elsewhere, it looks like a rolling shutter works like this: Let's say you pick 1/1000 for a shutter speed. It exposes the first row of pixels at 1/1000th, then after a certain amount of time, exposes the next row at 1/1000th, and so on. If you choose 1/8000 for a shutter speed, it exposes the first row at 1/8000, but still waits that same predetermined time period before exposing the next row at 1/8000. The time it takes to write the entire exposure is the same either way. (Somewhere between 1/48 & 1/60 of a sec.) So that's where you get the jello.

[paranoia23]

if it was like you say ... it would be impossible to take this "freezing pictures" with high shutter speeds.

[panoguy]

This is one of the few I've shot where the effect is really visible.

Nice Jello shot!

[GibbyTheMole]

if it was like you say ... it would be impossible to take this "freezing pictures" with high shutter speeds.

As far as I know it is pretty much is impossible to freeze fast motion with a rolling shutter and no wobble (jello effect). With a mechanical shutter that exposes the entire frame at once, it's a different story. The native non-toy lenses (01 prime, 02 zoom, 06 zoom) and the genuine Pentax Q to K adapter have a leaf shutter built into the lens with speeds up to 1/1000 of a sec. Jello isn't a problem with those because you're not using the electronic shutter unless you go over 1/1000. Faster than 1/1000 with those, and you're using the electronic shutter which is susceptible to the jello effect.

More info here.