Image Stabilization with EP-1 & Panasonic 45-200 lens?

Discussion in 'Olympus Cameras' started by Ctein, Mar 18, 2010.

  1. Ctein

    Ctein Mu-43 Regular

    Joined:
    Mar 18, 2010
    Messages:
    37
    Location:
    Daly City, California USA
    Dear Folks,

    I just bought the Panasonic lens for use with my Olympus, and a quick handful of test photographs at 200mm has left me confused about what provides the best image stabilization.

    With long exposures (1/30 of the second and more) any image stabilization looks a lot better than no image stabilization, but I can't see any obvious differences between using the Olympus stabilization (IS1-auto focal length mode), the Panasonic in-lens stabilization, or both at the same time. It doesn't seem like any one has an advantage over the other, just that any image stabilization is better than none.

    At high shutter speeds (1/100 of a second and less) I'm seeing just the opposite; any combination of image stabilization-- in lens, in camera, or both -- seems to be slightly worse than leaving it off entirely.
    Advertisement

    To put it simply... ?????

    I could drive myself crazy trying to figure out what's going on, but I figure there have to be folks here with more experience than me who I can ask.

    Yes I'm aware of the oft-made statements to the effect that one shouldn't have both systems on the same time. It's never been clear to me whether those statements have any knowledgeable authority behind them or just assumption. In this case, I seem to be seeing that it doesn't make any difference at all, and that does seem a little odd.

    The business about getting steadier results with no stabilization at all at higher shutter speeds seems more than a little odd.

    Comments, suggestions, advice, people? Too many variables and much weirdness here for me to figure out on my own.


    ~ pax \ Ctein
    [ Please excuse any word-salad. MacSpeech in training! ]
    ======================================
    -- Ctein's Online Gallery http://ctein.com
    -- Digital Restorations http://photo-repair.com
    ======================================
     
  2. BBW

    BBW Super Moderator Emeritus Charter Member

    Joined:
    Jan 13, 2010
    Messages:
    6,284
    Location:
    Near "Playland" outside of NYC, NY, USA
    Ctein, you've posed some interesting observations here. I'm sure as the day/evening wears on that people with solid technical knowledge and experience will post with some helpful feedback for you.

    And, by the way welcome to mu43.com! I didn't realize this was your first post here. Do drop by the Welcomes and Introductions forum, if you have the time. Glad to have you aboard!:biggrin:
     
  3. Brian Mosley

    Brian Mosley Administrator Emeritus Charter Member

    Joined:
    Dec 15, 2009
    Messages:
    3,018
    Hi Ctein, and welcome to the forum - it's quite an honour for us to have you here... I've enjoyed reading you on TOP :hail:

    As for stabilisation, I would switch off OIS to conserve power (it's always chugging away if enabled on a Pen) and rely on IBIS unless you're shooting video.

    As for the two systems conflicting - I think it used to be a problem with the earlier Olympus 4/3rds cameras and the PL OIS lenses but with the newer Pens they seem to cooperate better. If in doubt, switch it off.

    I think there's an element of 'personal holding style' which impacts the effectiveness of any stabilisation system - if you're getting worse results at 1/100s and faster - just switch off stabilisation.

    Just out of interest, can you describe how you hold your camera + lens?

    Cheers

    Brian
     
  4. bonycolin

    bonycolin New to Mu-43

    Joined:
    Mar 12, 2010
    Messages:
    1
    Location:
    Hong Kong
    The operation manual does recommend turning off the image stabilization when using a tripod. Of coz this has to do with being steady on a tripod and using longer exposure the image stabilization would in fact create subtle vibration to the sensor itself. But in your case this logic doesn't seem to apply ><
     
  5. Amin Sabet

    Amin Sabet   Administrator Subscribing Member

    Joined:
    Apr 10, 2009
    Messages:
    8,587
    Location:
    Boston, MA (USA)
    Hi Ctein, welcome to our forum! I don't have an Olympus body, but what you're reporting certainly is bizarre. On Panasonic bodies, I have never noticed a negative impact of keeping the image stabilization on at any shutter speed, but I haven't formally tested it either.

    Btw - I removed the asterisks from your urls. Your links are welcome here.
     
  6. Brian Mosley

    Brian Mosley Administrator Emeritus Charter Member

    Joined:
    Dec 15, 2009
    Messages:
    3,018
    Amin, which OIS mode do you use on your Panasonic body? if you're using mode 1, there should be no difference between an Olympus with IBIS disabled, and a Panasonic body.

    Cheers

    Brian
     
  7. Amin Sabet

    Amin Sabet   Administrator Subscribing Member

    Joined:
    Apr 10, 2009
    Messages:
    8,587
    Location:
    Boston, MA (USA)
    Yes, I generally use Mode 1. Does Mode 2 function on Olympus bodies?
     
  8. Amin Sabet

    Amin Sabet   Administrator Subscribing Member

    Joined:
    Apr 10, 2009
    Messages:
    8,587
    Location:
    Boston, MA (USA)
    This weekend, I'll try to do a 45-200mm handholding test at 200mm and various shutter speeds on the G1 with IS off, in Mode 1, and Mode 2. If anyone is up for doing the same with their Pen, throwing in body IS as another variable, that would be terrific.
     
  9. Brian Mosley

    Brian Mosley Administrator Emeritus Charter Member

    Joined:
    Dec 15, 2009
    Messages:
    3,018
    No, mode 2 is only functional on Panasonic bodies...

    Mode 1 is where the lens is constantly active - Ctein, I'm sure you're aware that OIS mode 1 requires a half second to 'lock on' to your subject and settle into stabilising the image.

    Mode 2 is much more similar to IBIS, in that stabilisation is only applied for the duration of the exposure.

    Cheers

    Brian
     
  10. dcisive

    dcisive Mu-43 Veteran

    Joined:
    Feb 19, 2010
    Messages:
    460
    Location:
    Salt Lake City, Utah
    Real Name:
    Lee
    the shot I posted earlier in the week of the PL1/45-200OIS combo was handheld with IBIS OFF in the PL1 and OIS ON on the lens, it worked best of the combination's I tried.
     
  11. bilzmale

    bilzmale Mu-43 All-Pro

    Joined:
    Mar 2, 2010
    Messages:
    1,743
    Location:
    Perth, Western Australia
    Hello and welcome Ctein - I raised this question here a while back and got some interesting answers.
     
  12. Ctein

    Ctein Mu-43 Regular

    Joined:
    Mar 18, 2010
    Messages:
    37
    Location:
    Daly City, California USA
    Dear folks,

    OK, I am really pleased to see so many quick and intelligent responses (too many times I will go to a technical forum and post a question about a problem I'm having and hear ringing silence and the soft sound of chirping crickets in the night).

    But, I'm not much less confused now than when I started. Well, a little less confused; the photographs on dpreview (thanks for the link, bilzmale and Boyzo) are vaguely consistent with what I got: not much difference between the two stabilization systems or having both on at the same time. On the other hand, the hypothesis on dpreview that having lens stabilization on overrides the in-camera stabilization is clearly wrong; in my longer exposure tests I can feel the camera vibrate from the internal stabilization when both are on or just internal stabilization is on, and not when internal stabilization is off.

    The manual, of course, says don't use both at the same time. Manuals, of course, are not always to be trusted (a shocking discovery, I know).

    Turning off lens stabilization would make sense, but it's very nice to have a steady image at the longer focal lengths. I've got a spare battery pack, so I guess I'll have to run some tests and see how fast the system drains if I leave lens stabilization on. Anyone here got any experience with that? In typical use, twice as fast? Five times as fast? I know everyone's mileage will vary; I'm just asking about your ordinary walk-around experience.

    The business about apparently getting sharper results if I turn both stabilization systems off still bothers me and doesn't make any sense. Turning off both stabilization systems, in practice, is kind of a nonstarter; there's no way I'm going to remember to turn stabilization on and off depending on which lens I use and which shutter speeds I have set. It would be hard enough even if there were an easy way to set up custom modes with this camera (which there doesn't seem to be). And besides, that just seems like a wrong result. I need to figure out what's going on. I'm hoping when other people here run their informal tests, we'll start to be able to make some sense out of this... or at least they'll see evidence that my results are truly anomalous.

    Good question about how I hold my camera, Brian. I am conveniently nearsighted, so I can hold it pretty close to my head. The screen is 8-10 inches from my eyes. With a long lens like this, I'll hold the body in my right hand and cradle the lens in my left hand with my upper arms and elbows braced against my side; very similar to the position one would use for available light work with an SLR, except I don't actually hold the camera to my eye. It feels pretty steady to me, but that doesn't mean it's optimal. Especially since I'm used to working with much heavier equipment. A friend of mine in Minneapolis, David Dyer-Bennet, found that a very different strategy worked for him for stabilizing his DSLR. So if other people have different ways of holding cameras that work better for them, I'd be happy to give them a try.

    ~ pax \ Ctein
    [ Please excuse any word-salad. MacSpeech in training! ]
    ======================================
    -- Ctein's Online Gallery http://ctein.com*
    -- Digital Restorations http://photo-repair.com*
    ======================================
     
  13. Ctein

    Ctein Mu-43 Regular

    Joined:
    Mar 18, 2010
    Messages:
    37
    Location:
    Daly City, California USA
    Dear folks,

    OK, ran some more tests in the 1/250th sec range at 200 mm, and I can't convince myself in this series that image stabilization is actually worse than no stabilization, but it certainly is not any better.

    Which raises a couple of questions in my mind, which some of you may know the answer to:

    1) What is the frequency response of image stabilization systems? Normally we rely on them for relatively long exposures; when you're talking about exposures that are only 4-6 ms long, maybe the response time of the system is simply too large a fraction of the total exposure time? Anyone here know the answer or know of a technical website that discusses this?

    2) In a related vein, am I simply being too demanding? I've never had a chance to compare stabilization and no stabilization with a long lens like this. I'm expecting results to be sharp down to the single pixel level-- no visible motion blur. Am I simply demanding too much of the system? Maybe it can't do better than a human being down to the single-pixel level for long focal length lenses.

    I honestly don't know the answer to either of these questions. In other words, I don't know the genuine limits of the equipment I'm working with. I should; if it turns out that I'm asking more of the equipment than current technology allows, I will simply adapt. It's a strategy that's worked for me for 40 years so far.

    It's just, if I can have things work BETTER, I could live with that, too.

    And just to be clear, at slow shutter speeds the superiority of image stabilization is utterly apparent. With my body braced, but not using a tripod, image stabilization will get me an acceptably sharp photograph, about one in five times, at one quarter second at 200 mm. That's hugely impressive.


    ~ pax \ Ctein
    [ Please excuse any word-salad. MacSpeech in training! ]
    ======================================
    -- Ctein's Online Gallery http://ctein.com*
    -- Digital Restorations http://photo-repair.com*
    ======================================
     
  14. squeegee

    squeegee Mu-43 Veteran Charter Member

    Joined:
    Jan 26, 2010
    Messages:
    403
    With the older 4/3rds cameras this was definitely true - at least for some portion of the cameras. People had posted videos of the e-5xx series using one or the other, or attempts at using both at once etc.

    Would it be possible for some one to run a sound test and post the results? I don't have an OIS lens. What I mean by sound test is...
    1) turn off OIS, turn on IBIS, set the shutter to 1.5 seconds (make sure it's only 1.5 seconds). Take a picture, you should clearly hear a grinding rumble. That rumble is the IBIS.
    2) turn on OIS, turn off IBIS, take a picture, I'm assuming you shouldn't hear a rumble (I'm told it's more of a whirl...)
    3) turn both OIS and IBIS on, and take a picture. Do you hear both a whirl, and a rumble?

    On another note, I also feel the camera vibrate with IBIS, it "worries" me that it's sometimes doing more harm than good... but apparently I worry too much about things.
     
  15. Ctein

    Ctein Mu-43 Regular

    Joined:
    Mar 18, 2010
    Messages:
    37
    Location:
    Daly City, California USA
    Dear Squeegee,

    Lens stabilization is essentially noiseless; well, I can't hear it, anyway. In-body stabilization does make a whirring sound and makes the camera vibrate. Easy to tell when in-camera's on (at low shutter speeds), not easy to tell when in-lens is.

    The vibration is the way it should be. The in-camera stabilization works by jerking the sensor around very quickly, to counterbalance image movement. What you're feeling is the recoil, good old Newton's "for every action there is an equal and opposite reaction."

    Indeed, don't worry, be happy.

    pax / Ctein
     
  16. mauve

    mauve Mu-43 Top Veteran

    Joined:
    Mar 9, 2010
    Messages:
    892
    Location:
    Paris, France
    Could this be linked to what you're experiencing ? http://www.imaging-resource.com/PRODS/EP1/EP1BLUR.HTM

    I know they related the 'bug' to the kit zoom, but if it's shutter vibration interfering with in-body IS for long focal lenses, then it might apply to other tele as well. The 14-42 being really wobbly doesn't help, but at 200 every lens wobble a bit internally.

    Cheers,

    Mauve
     
  17. Amin Sabet

    Amin Sabet   Administrator Subscribing Member

    Joined:
    Apr 10, 2009
    Messages:
    8,587
    Location:
    Boston, MA (USA)
    Okay, this was very quick and dirty. I set my G1 to f/6.3, 1/250s, and auto ISO (should have set the ISO manually). Pointed it at my computer display and took 30 hand held images with the 45-200 at 200mm.

    Here are the 100% crops.

    The top row is with OIS off. Middle row is OIS mode 1, and bottom row is OIS mode 2. Would have been much better to do this with more light, ISO 100, and a target other than a display, but this was enough to convince me of the following:

    • None of these methods give me motion-blur free results every time
    • OIS mode 1 gave me the most consistently sharp results
    I found these results disappointing, but I don't know if I am being too demanding either. My impression in using Tamron's vibration compensation 28-300 and 18-270mm lenses was that there was reproducibly no motion blur whatsoever at these types of shutter speeds, but I never tried to test it this way.
     
  18. Ctein

    Ctein Mu-43 Regular

    Joined:
    Mar 18, 2010
    Messages:
    37
    Location:
    Daly City, California USA
    Hey guys,

    Thanks for all the help. Much food for thought and I'm going to have to mull over the test photos and reports some more.

    mauve, that's a great reference and the beautiful bit of technical research. I don't believe it's the problem I'm seeing, because my recollection is that my assortment of test photos didn't have the blur oriented vertically (I didn't save them), but I'm going to go back and recheck this. Resonances are not unheard of-- the Pentax 67 had a really serious shutter resonance at 1/15 second. I could handhold it with the 105 mm lens at 1/30th of a second and get sharp photographs. But even on a monopod with the mirror flipped up, I couldn't get sharp results at 1/15th of a second. In fact, it would take a heavy-duty tripod, in the 20 pound weight class, to keep transient vibration from clobbering photos made at 1/15 the second. But the problem would mostly go away at 1/8 of a second and it was invisible at one quarter second. Sound familiar, huh?

    I'm going to have to run some more Olympus experiments of my own, but it probably won't be today. Maybe tomorrow, maybe Tuesday.

    Amin, I appreciate your test; it indicates that I'm not crazy... or that we are all collectively so. Your choice.

    E-mail comments to me from Mike Johnston, Oren Grad, and Carl Weese strongly argue that the results I'm seeing, while unexpected for a naif like moi, are not anomalous. I was kind of hoping for a "set it and forget it" answer, but it looks like it's not going to shape up that way. I'm still interested in the technical issues behind the problems; I may very well turn this into this week's TOP column, in the hopes that one or more of their readers know something about the inner workings of stabilization systems. I have a definite feeling that frequency response is going to turn out to be a big issue, but that's just instinct, not knowledge, talking.


    ~ pax \ Ctein
    [ Please excuse any word-salad. MacSpeech in training! ]
    ======================================
    -- Ctein's Online Gallery http://ctein.com*
    -- Digital Restorations http://photo-repair.com*
    ======================================
     
  19. Ctein

    Ctein Mu-43 Regular

    Joined:
    Mar 18, 2010
    Messages:
    37
    Location:
    Daly City, California USA
    Dear folks,

    OK, had a chance to run some more 200mm comparisons today, shutter speeds from 1/30 of a second through 1/250, with no image stabilization, in body image stabilization, in lens image stabilization, and both.

    Confirmed that what I'm seeing is not mechanical vibration or resonance; blurring varies in orientation randomly from frame to frame. This time, the in-lens stabilization was consistently better than the in-body stabilization. I hadn't seen that previously; in fact, in one round of tests I did, the in body stabilization looked slightly better. So I think that's random, not a meaningful result. I'm still going to stick with the overall opinion that they are comparable in quality, until demonstrated otherwise.

    Having both systems on didn't perform any worse. That agrees with what I saw before.

    At 1/125 of a second and above, image stabilization is not obviously better than no stabilization. It is most definitely not foolproof; I would say only half of the frames were pixel-level sharp.

    At 1/60 of a second and lower, image stabilization is clearly advantageous, although not by as much as one would think. It looks like the lower the shutter speed, the better image stabilization does relative to ordinary handholding; I've had utterly spectacular results when I've used shorter focal length lenses, handheld in the 1/8 -1/15 second range. (This was handheld with the Panasonic 20 mm lens at f/1.7 and 1/7 of the second -- the frames are consistently sharp : http://ctein.com/contributor/Moon_Palms.jpg ). Of course, with a 200 mm focal length, low shutter speeds are going to be problematical (understatement).

    What I have to conclude is that image stabilization is of modest benefit with this lens used at its longest focal length, but it's not really a game-changer. That's somewhat disappointing. Most notably, getting consistently tack-sharp photographs at 200 mm it's going to require be to use exposure times even shorter than 1/250 of a second, unless I can improve my handholding skills. The methods and stance I've used in the past worked extremely well, but they may not be optimal for this camera and lens combination.

    I'm going to write this up for my next column for The Online Photographer (which will appear Thursday). Hopefully some of the readers there will have more expertise in this area than we do and may even be able to answer some of the technical questions I've raised, like the frequency response of these systems.


    ~ pax \ Ctein
    [ Please excuse any word-salad. MacSpeech in training! ]
    ======================================
    -- Ctein's Online Gallery http://ctein.com*
    -- Digital Restorations http://photo-repair.com*
    ======================================
     
    • Like Like x 2
  20. Amin Sabet

    Amin Sabet   Administrator Subscribing Member

    Joined:
    Apr 10, 2009
    Messages:
    8,587
    Location:
    Boston, MA (USA)
    Ctein, thank you for sharing your results with us. I look forward to your column and the responses to it!

    I have to say, I'm a bit disappointed in what we both have found in our testing. I have a general sense that Canon and Nikon's latest VR do a better job with these narrow angle of view lenses than what we are finding with our Pana 45-200, and I thought that Tamron's VR was even more effective than the Canon and Nikon image stabilization systems I had tried before.

    Though I haven't tested it, my sense is that the OIS of the Pana 14-140 is less effective than that of the Pana 45-200, and the OIS on my Pana-Leica 45mm lens is not impressing me that much either.

    I could be wrong about all of this, having never tried controlled testing of Canon/Nikon/Tamron stabilization compared to the Panasonic. These are just my impressions.

    I don't know anything about how these systems work, but using the Tamron VC, the lens is rigid, motionless, until it isn't anymore, and then it jerks. Seemingly not suitable for video application, where vibration must be neutralized without eliminating the ability to move/pan. Could it be that the Panasonic OIS is less effective for still photography because it is somehow optimized for video?
     

Share This Page