O
OlyPent
Guest
I'm experimenting with the Sigma 150-500mm and so far, its I.S. seems to work better than the K-S2 in-body. But that is preliminary. I do not know if or when the two systems might ever work in-tandem though.
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Priscilla Turner
Senior Member
I have done no comparisons, but I do know that with a long zoom, when the reciprocal changes all the time, it's truly effective to set SR on, Av, Auto ISO and let speed be determined automatically and flexibly by the cam according to the focal length in use at any given moment.
I can walk about making hundreds of images with these settings and get no shake at all. The only adjustment that may need to be made is of aperture if I run very short of light, or want a shallow depth of field. Otherwise I set the sharpest, usually mid-aperture with modern lenses, for anything requiring plenty of depth of field.
It's only with a single focal length that this method with SR on becomes unnecessary.
I can walk about making hundreds of images with these settings and get no shake at all. The only adjustment that may need to be made is of aperture if I run very short of light, or want a shallow depth of field. Otherwise I set the sharpest, usually mid-aperture with modern lenses, for anything requiring plenty of depth of field.
It's only with a single focal length that this method with SR on becomes unnecessary.
MightyMike
Forum Pro
IS in the lens has the added benefit of you seeing a stable image and therefore you don't correct as aggressively compared with not seeing a stabilized image. this could suggest in lens stabilization is better, however in camera stabilization has added benefits such as rotational stabilization as well as pitch and yaw stabilization as seen in the latest cameras. More important than all that is the stabilization spool-up time... how long from activation does it take to actually be ready to stabilize during the exposure. This is not a commonly mentioned issue but likely leads to a lot of complaints about ineffective stabilization when its really the user who jumps the gun, however you could blame the manufacturer for not providing guidelines in this regard either.
Petroglyph
Veteran Member
- Messages
- 6,096
- Solutions
- 9
- Reaction score
- 1,384
My 150-500 OS has a panning OS setting and that was always better for panning with a moving subject (since my K5 would turn SR off when it detected panning). Each generation of Pentax body has gotten better with the SR and they seem to have 5-axis on the latest cameras so I couldn't tell for sure if Pentax SR with a Pentax lens is better than in-lens stabilization. I'd guess that Pentax SR would work less well with a 3rd party lens though. The effectiveness of all types of stabilization varies with focal length and CIPA will test at a fixed length, probably 50-80mm and that would be where the newer cameras get the best numbers for SR effectiveness.
Eurostar
Well-known member
I have K-5 and Sigma OS HSM 50-500. So far, my findings is that SR and OS are both able to stop handshake, with Pentax SR maybe noticeable steady. Where the lens OS is better is that on a less shaky image coming from the lens, the AF sensor works better. I guess part of the better AF performances form Canon and Nikon come from the in-lens optical stabilization .
petrochemist
Senior Member
With longer lenses the lens IS has an advantage, for shorter lenses in body is at least as good. If using in lens IS make sure to turn off the camera stabilisation. There are a few cameras that can use both systems in tandem but practically anywhere there's both available they don't communicate & both work to cancel the same movement effectively giving exactly the opposite of the unstabilised shake. I've done that more than once :-(
Leon Wittwer
Forum Pro
All IS and VR systems correct for pitch and yaw. These are the dominant effects in causing blur. I've yet to see anyone show where roll has had any effect on an image. The translational modes of motion are almost always insignificant.
Holger Bargen
Veteran Member
It depends on the situation:
For long tele focal range lens I.S. works better. I think that in camera I.S. above 200 mm is very limited help.
Lens I.S. may disturb bokeh - but I have seen very nice bokeh from lenses with lens I.S. turned "ON".
In-camera I.S. has problems with panning. Some lenses offer lens I.S. that allow panning.
In-camera I.S. needs very low amount of energy while lens I.S. may reduce the number of photos per battery significantly.
In-camera I.S. works only for the first photo of a series - it always needs a little time to prepare for proper work.
I have not much experience so far - but I could imagine that meanwhile in camera I.S. for the focal range wide to light tele may be more effective than lens I.S.
Best regards
Holger
For long tele focal range lens I.S. works better. I think that in camera I.S. above 200 mm is very limited help.
Lens I.S. may disturb bokeh - but I have seen very nice bokeh from lenses with lens I.S. turned "ON".
In-camera I.S. has problems with panning. Some lenses offer lens I.S. that allow panning.
In-camera I.S. needs very low amount of energy while lens I.S. may reduce the number of photos per battery significantly.
In-camera I.S. works only for the first photo of a series - it always needs a little time to prepare for proper work.
I have not much experience so far - but I could imagine that meanwhile in camera I.S. for the focal range wide to light tele may be more effective than lens I.S.
Best regards
Holger
- Messages
- 30,186
- Solutions
- 112
- Reaction score
- 12,428
Something has to move to counter camera/lens moment. In the case of in-camera stabilization...it's the actual sensor. In the case of lens stabilization... it's an element in the lens. The longer the focal length...the further the in-camera stabilization unit has to physically move.... vs the relative shorter distance the in-lens element must move. That's why in-camera lens stabilization often is less effective with long lenses.
A good summary:
Lens stabilization vs in-camera stabilization
MightyMike
Forum Pro
Alex Sarbu
Forum Pro
I wonder if that's indeed the case - the "sensor must move a longer distance" thing.
There's a variable which is often forgotten: the usable shutter speed, faster as we increase the focal length. Regardless the focal length, the stabilization - optical or mechanical - should only be good for a several pixels blur.
However, there are many other factors involved.
Alex
- Messages
- 30,186
- Solutions
- 112
- Reaction score
- 12,428
When a photographer is composing a picture in a DSLR optical viewfinder the image seen is subject to four induced movements: Roll Pitch Yaw Pistons. Canon and Nikon lens stabilization has roll, pitch & yaw gyro-accelerometers. Not sure about other manufactures. With Nikon/Canon equivalent x & y translations are used to counter Pitch Roll and Yawl. In some units, an additional coil-magnet actuator also shifts the lens yoke along the z-axis to compensate for piston movements.
And don't forget that it's a 2-D image being compensated for 3-D movement. You often only need to shift in the x and Y to compensate for a more complicated 3-D shift in the camera.
- Messages
- 30,186
- Solutions
- 112
- Reaction score
- 12,428
It is....just use a in-camera stabilization body and a long zoom lens with stabilization. Watch how one becomes more effective over the other as focal length changes. Remember...we are speaking of relative distances... optical vs sensor elements
Don't forget you are talking about relative shutter speed and not actual in many cases. The time frame the sensor is being exposed (with a hardware shutter system) is the same at 1/400s as it is at 1/2000s.
Alex Sarbu
Forum Pro
But I cannot identify the factors involved just with that simple test; all I could do is to observe the behavior, for those specific pieces of equipment.
Good point. However, the time a single pixel is exposed corresponds to the set shutter speed and not to the total shutter travel time. So would the shake.
Alex
- Messages
- 30,186
- Solutions
- 112
- Reaction score
- 12,428
Yes and one gets less optimal results...relative to the other... as focal length increases. Knowing how they are designed...makes it easy to see it's an issue with physics. There's lots out there written on it.
True.... but are we concerned (in practical terms) about just each pixel or the "sensor" seeing a stable image during the time it's being exposed?
Alex Sarbu
Forum Pro
I'd say, we are concerned about what happens when each pixel is exposed; at 1/2000 the sensor (in its entirety) does not see any light for 1/200 or whatever's its sync speed. If it doesn't see any light, there's no "shake" to be captured.
Alex
- Messages
- 30,186
- Solutions
- 112
- Reaction score
- 12,428
And if it is stable for pixels not getting any light then it is for those getting light as well. Is it better to be "stable" over the average full exposure or "stable" for some and less for others? Why do we not see noticeable Lines/Areas (equal in area to the open curtain slit) less sharp than others areas due to camera shake as shutter speed increases?
--
My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)
Last edited:
TomWret
Senior Member
One of the BIG reasons I switched to Pentax was in-camera IS and now I have payed for IS only once!!
I need IS, Tom W(ret)
I need IS, Tom W(ret)
flektogon
Veteran Member
I might be wrong but I think that "in lens" IS works continuously while the Pentax "in body" SR is activated only after the AF has achieved its goal. So, generally it takes more time to get a stable picture ready to be taken with the SR system.
Peter
Peter