Shake Reduction and 'action' photography
Posted 07/05/2013 - 22:32
Link
percy wrote:
Your panning technique is obviously much smoother than mine, Chris. I shall have to keep practising.
I also note that you used a K-5. Earlier in this thread I believe you commented about apparent SR differences between the K-5 and K-5 II, and that with the K-5 II and a 300mm lens you turn SR off. Other than the autofocus differences between the 2 camera bodies, I wonder what else could be going on. Is the AF linked to SR in some way?
I think you could well be correct, Karel, that a low enough pan rate is likely to be successful. I was doing my best to rotate the whole body rather than just the camera, but most of the time the pan rate was quite high - possibly higher than needed for the stock cars.
My K5II does seem to lock on slightly faster than my K5.Your panning technique is obviously much smoother than mine, Chris. I shall have to keep practising.
I also note that you used a K-5. Earlier in this thread I believe you commented about apparent SR differences between the K-5 and K-5 II, and that with the K-5 II and a 300mm lens you turn SR off. Other than the autofocus differences between the 2 camera bodies, I wonder what else could be going on. Is the AF linked to SR in some way?
I think you could well be correct, Karel, that a low enough pan rate is likely to be successful. I was doing my best to rotate the whole body rather than just the camera, but most of the time the pan rate was quite high - possibly higher than needed for the stock cars.
I forgot to mention that the photo in this thread was taken from the seated position as at Coventry Stadium you have to remain seated while the racing is on.
Regards, Chris.
Posted 08/05/2013 - 10:12
Link
About a year ago I photographed the Halfords Pro Cycle Race series event in Durham. It never occurred to me to turn off the SR. Many of my shots were panned and I was pretty pleased with the results. So when I go to this year's event next week I'll be leaving the SR on again.
David
PPG: http://www.pentaxphotogallery.com/artists/davidtrout
David
PPG: http://www.pentaxphotogallery.com/artists/davidtrout
Posted 08/05/2013 - 11:03
Link
There do seem to be a number of conflicting experiences with this.
I just had a quick scan through some pictures I took at Mallory Park 12 months ago, using a K-5 and 50-135 lens. Out of the faster panning shots, probably about half were stabilised and half not (whereas my recent experience with K-5 IIs and 300 lens was none stabilised).
So there certainly do seem to be some difference between the camera body/lens combinations.
Another forum member, Karel, put me in touch with 'ExifData' and 'ExifDataGUI' for Windows, which does give the SR info - not just whether it is active or not, but whether or not it has stabilised.
It would be interesting to find out how many of your 'fast panning' shots do have the SR stabilised, and what differences there are between camera body/lens combinations.
I just had a quick scan through some pictures I took at Mallory Park 12 months ago, using a K-5 and 50-135 lens. Out of the faster panning shots, probably about half were stabilised and half not (whereas my recent experience with K-5 IIs and 300 lens was none stabilised).
So there certainly do seem to be some difference between the camera body/lens combinations.
Another forum member, Karel, put me in touch with 'ExifData' and 'ExifDataGUI' for Windows, which does give the SR info - not just whether it is active or not, but whether or not it has stabilised.
It would be interesting to find out how many of your 'fast panning' shots do have the SR stabilised, and what differences there are between camera body/lens combinations.
Posted 08/05/2013 - 17:34
Link
This is a sample from the cycle race at Durham: K5 + DA* 50-135mm lens, with SR turned on.
PPG:
http://www.pentaxphotogallery.com/artists/davidtrout
Posted 08/05/2013 - 20:10
Link
kh1234567890 wrote:
I suspect that the success or otherwise of SR during panning is largely down to the panning technique. The camera detects movement ('shake') using three angular velocity gyro sensors (not accelerometers). The signal is high pass filtered - i.e. only angular velocities above a set value are used. Hence panning should not affect SR as long as the pan rate is low enough or if it is a translational pan rather than a rotation (the sensors do not detect linear velocity).
So, rotating your whole body rather than just the camera should have a better chance of success.
Interesting - where did you find the information about the gyro sensors? I'd be curious to find out more if a more technical breakdown of the SR system has been published somewhere than the usual marketing guff that just tries to dazzle you with buzzwords.
I suspect that the success or otherwise of SR during panning is largely down to the panning technique. The camera detects movement ('shake') using three angular velocity gyro sensors (not accelerometers). The signal is high pass filtered - i.e. only angular velocities above a set value are used. Hence panning should not affect SR as long as the pan rate is low enough or if it is a translational pan rather than a rotation (the sensors do not detect linear velocity).
So, rotating your whole body rather than just the camera should have a better chance of success.
Posted 08/05/2013 - 22:51
Link
If you switch on live view and pan the camera (K-5) smoothly, you'll see that the image only seems to 'jump' when you start or stop panning. Suggesting that as long as you start early and follow-through afterwards, you might get benefit of SR (although it might affect the effectiveness of smoothing vertical movement - if that makes sense).
Generally SR seems to be more effective with higher frequency vibration than lower.
Generally SR seems to be more effective with higher frequency vibration than lower.
Posted 08/05/2013 - 23:18
Link
Mandrake wrote:
Interesting - where did you find the information about the gyro sensors? I'd be curious to find out more if a more technical breakdown of the SR system has been published somewhere than the usual marketing guff that just tries to dazzle you with buzzwords.
Patents are usually a good start, if you can get your head round the patent agent speak
Interesting - where did you find the information about the gyro sensors? I'd be curious to find out more if a more technical breakdown of the SR system has been published somewhere than the usual marketing guff that just tries to dazzle you with buzzwords.
Posted 08/05/2013 - 23:25
Link
johnha wrote:
If you switch on live view and pan the camera (K-5) smoothly, you'll see that the image only seems to 'jump' when you start or stop panning. Suggesting that as long as you start early and follow-through afterwards, you might get benefit of SR (although it might affect the effectiveness of smoothing vertical movement - if that makes sense).
Generally SR seems to be more effective with higher frequency vibration than lower.
Remember that SR functions differently when in live view, in that the sensor is being moved all the time - when it is about to hit the movement limits or the camera movement stops it re-centers. Hence the jumps. That is also why live view with SR on uses much more power. In viewfinder view mode the sensor is centered at mirror up time and is only moved during the actual exposure.
If you switch on live view and pan the camera (K-5) smoothly, you'll see that the image only seems to 'jump' when you start or stop panning. Suggesting that as long as you start early and follow-through afterwards, you might get benefit of SR (although it might affect the effectiveness of smoothing vertical movement - if that makes sense).
Generally SR seems to be more effective with higher frequency vibration than lower.
Posted 08/05/2013 - 23:42
Link
There's also the extra power being used to hold the mirror up in live view.
When not in live view there's power being used to hold the sensor in the centre position, and when SR is off.
When not in live view there's power being used to hold the sensor in the centre position, and when SR is off.
[i]Bodies: 1x K-5IIs, 2x K-5, Sony TX-5, Nokia 808
Lenses: Pentax DA 10-17mm ED(IF) Fish Eye, Pentax DA 14mm f/2.8, Sigma 17-70mm f/2.8, Pentax-A 28mm f/2.8, Sigma 30mm F1.4 EX DC, Pentax-A 50mm f/1.2, Pentax-A 50mm f/1.4, Pentax-FA 50mm f/1.4, Pentax-A 50mm f/1.7, Pentax DA* 50-135mm f/2.8, Sigma 135-400mm APO DG, and more ..
Flash: AF-540FGZ, Vivitar 283
Lenses: Pentax DA 10-17mm ED(IF) Fish Eye, Pentax DA 14mm f/2.8, Sigma 17-70mm f/2.8, Pentax-A 28mm f/2.8, Sigma 30mm F1.4 EX DC, Pentax-A 50mm f/1.2, Pentax-A 50mm f/1.4, Pentax-FA 50mm f/1.4, Pentax-A 50mm f/1.7, Pentax DA* 50-135mm f/2.8, Sigma 135-400mm APO DG, and more ..
Flash: AF-540FGZ, Vivitar 283
Posted 08/05/2013 - 23:49
Link
Smeggypants wrote:
When not in live view there's power being used to hold the sensor in the centre position, and when SR is off.
According to the timing diagram in Fig.8 of one of the patents the sensor is only driven in the interval between mirror up and mirror down.
When not in live view there's power being used to hold the sensor in the centre position, and when SR is off.
Posted 09/05/2013 - 00:28
Link
kh1234567890 wrote:
That would make perfect sense from a power efficiency POV. I suppose it's difficult to imagine the sensor assembly being shifted to centre position from mirror up to shutter opening with all the inertia and settling involved. This centering off the sensor would also have to happen with SR Switched off as well.
Smeggypants wrote:
When not in live view there's power being used to hold the sensor in the centre position, and when SR is off.
According to the timing diagram in Fig.8 of one of the patents the sensor is only driven in the interval between mirror up and mirror down.When not in live view there's power being used to hold the sensor in the centre position, and when SR is off.
[i]Bodies: 1x K-5IIs, 2x K-5, Sony TX-5, Nokia 808
Lenses: Pentax DA 10-17mm ED(IF) Fish Eye, Pentax DA 14mm f/2.8, Sigma 17-70mm f/2.8, Pentax-A 28mm f/2.8, Sigma 30mm F1.4 EX DC, Pentax-A 50mm f/1.2, Pentax-A 50mm f/1.4, Pentax-FA 50mm f/1.4, Pentax-A 50mm f/1.7, Pentax DA* 50-135mm f/2.8, Sigma 135-400mm APO DG, and more ..
Flash: AF-540FGZ, Vivitar 283
Lenses: Pentax DA 10-17mm ED(IF) Fish Eye, Pentax DA 14mm f/2.8, Sigma 17-70mm f/2.8, Pentax-A 28mm f/2.8, Sigma 30mm F1.4 EX DC, Pentax-A 50mm f/1.2, Pentax-A 50mm f/1.4, Pentax-FA 50mm f/1.4, Pentax-A 50mm f/1.7, Pentax DA* 50-135mm f/2.8, Sigma 135-400mm APO DG, and more ..
Flash: AF-540FGZ, Vivitar 283
Posted 09/05/2013 - 21:10
Link
kh1234567890 wrote:
Cheers for that. Interesting stuff.
Mandrake wrote:
Interesting - where did you find the information about the gyro sensors? I'd be curious to find out more if a more technical breakdown of the SR system has been published somewhere than the usual marketing guff that just tries to dazzle you with buzzwords.
Patents are usually a good start, if you can get your head round the patent agent speak Interesting - where did you find the information about the gyro sensors? I'd be curious to find out more if a more technical breakdown of the SR system has been published somewhere than the usual marketing guff that just tries to dazzle you with buzzwords.
Posted 12/05/2013 - 21:13
Link
Posted 13/05/2013 - 00:15
Link
While I can see that this is clearly aimed at newcomers to Pentax SR, Pentax would be much better off writing a techie white paper on their SR, which is something that they should be proud of. As always, unless you understand how something works you are going to have a hard time figuring out why it sometimes doesn't
Add Comment
To leave a comment - Log in to Pentax User or create a new account.
425 posts
16 years
Herefordshire
I also note that you used a K-5. Earlier in this thread I believe you commented about apparent SR differences between the K-5 and K-5 II, and that with the K-5 II and a 300mm lens you turn SR off. Other than the autofocus differences between the 2 camera bodies, I wonder what else could be going on. Is the AF linked to SR in some way?
I think you could well be correct, Karel, that a low enough pan rate is likely to be successful. I was doing my best to rotate the whole body rather than just the camera, but most of the time the pan rate was quite high - possibly higher than needed for the stock cars.