OM-3 stabilization, is the stabilization overhyped or is something wrong?

SimonV said:

I've never heard as much hype as with Olympus cameras, and keep hearing that the IBIS can make up for the smaller sensor by making slower shutter speeds possible. Maybe it is indeed like mentioned in this thread, that Olympus was first and thus has that kind of reputation.

Click to expand...

That is absolutely the case here. The same goes for many things in this industry when it comes to reputation across brands. Some people still think of Sony as the brand with horrible colors because of their reputation in the early A7 series days. Or that their cameras are unreliable because in the early days, those bodies were prone to overheating and had poor weather sealing. Lumix is known for horrible AF performance, despite their new cameras performing just as well as other brands in video AF (stills AF another story, but still much improved).

To some extent, some of these things still hold true. Flagship M43 cameras like the GH7 still have the best IBIS, super stable and you don't get weird warpy corners when shooting with wide lenses. But reality is, all the brands have come much closer in performance over the years. Don't expect any particular brand to blow the competition away for any particular performance metric.

SimonV said:

Lens I.S. Priority Off

Now if I was using a longer lens like a 200mm or 100-400mm, I would use the Lens Priority IS since it tends to be better with longer lenses.

Click to expand...

The Lens IS Priority item is ONLY for those Panasonic stabilised lenses that have no OIS switch, it has absolutely no effect on any other lens.

The way you have set it up, if you mount a Panasonic unswitched OIS lens then it will automatically disable the IBIS and use the OIS only. [edit: brain fade error, if you mount an unswitched lens there will be no OIS available with your setup] If you turn off IBIS in the menu with that same lens then you get no stabilisation at all. With that lens the IBIS being On acts as the master switch for stabilisation, and then the Lens IS Priority if On flips over to OIS only when it sees the appropriate lens.

If you have no Panasonic unswitched OIS lenses then the Lens IS Priority item has no function and can be left on or off.

[Edit: If you have a switch on the lens then that controls the OIS and will auto disable the IBIS as appropriate]

SimonV said:

Another thing I noticed being new to the camera is that shutter shock (I assume?)

Click to expand...

Shutter shock appears usually as double edges on any bright detail. In my tests when on a tripod the shutter shock (if it happens) will occur at the same amplitude for shot after shot. If handheld the shock effect varies from basically near undetectable (like a slight blur or misfocus) up to about 4x the tripod amount.

Traditionally the zero sec anti-shock mode (diamond icon) only works up to 1/320 sec and is disabled for mechanical shutter speeds faster than that. Some have reported shock at 1/500 sec. The worst shutter speeds (if it happens) are around 1/60 to 1/120 sec range it seems. It's where I did all my testing.

SimonV said:

affects the image quality quite noticeably in some scenarios when using Single shooting with the mechanical shutter. I haven't yet been able to test it thoroughly, but I did turn on the camera Anti-Shock setting, although I did not yet see that it would have made much of a difference, Silent Single shooting still always produced a sharper image. But that might be because of me, will have to see after more testing.

Click to expand...

My example of handheld shutter shock....


75-300mm lens at 300mm on E-PL5 handheld, shutter speed unknown in this pair. The E-PL5 has no zero sec anti-shock setting.

Compare the detail middle right side of each image, handheld it varied from just a slight blur (left image) to really darn obvious in the right image. I had picked the best and worst out of a sequence of maybe 10 shots taken. Sadly I long ago deleted all those tests back in 2014 so no more old examples to show and details of settings all forgotten now.

danart said:

HicHic said:

ahaslett said:

The A7Rvis rated the same as an OM1 mk i and the A7CR a little behind the OM5/OM3. I find the A7CR is way behind my OM5 in my hands.

TLR I think your OM3 is faulty.

Andrew

Click to expand...

OP has A7RV. The RV IBIS is way better than your RIV and CR. I actually feel the A7RV has at least as good IBIS as OM-1, if not better. You should try it yourself and see.

Click to expand...

on what basis are you rating the A7rv as being “if not better” than the OM-1?

Show me a 6 sec exposure completely handheld unassisted (not bracing against anything) with an A7RV at 24mm

Click to expand...

I am not suggesting that you can't do this but without controlled repeatable tests of the same equipment by multiple people it is just a personal take of what you can do . I can guarantee that I would be lucky to get anywhere near this long a shot not braced and expect sharp results

danart said:

https://www.dpreview.com/galleries/1260162627/photos/4470110/_8200065

Or a 2 second unassisted at 150mm equivalent

https://www.dpreview.com/galleries/1260162627/photos/4470124/

These exposures are very close to or hitting the actual quoted rating.

I’d be impressed if you could show me an actual example of the A7RV hitting an effective 5 stops…

Click to expand...

All such comparisons would show you are the respective hand holding abilities of the users not the absolute metric . There are other factors that come into play personal toleration for what is considered sharp. Also 60mp will show sharpness issues more than 20mp

I am leery of CIPA's tested results as well particularly with regard to battery life which does not seem to be accurate for mirrorless cameras. However the CIPA rating for stablilsiation is a standardised test done in a controlled and repeatable manner. This is the number that every maker lists on spec sheet so one would assume they give it some value. Users of whatever camera can say they can hand hold for x amount of time and claim sharp results but which may well be the case . But it is not a verifiable ,data point not repeatable in a controlled manner that applies to multiple people it is your personal take on it . Just as my take on it is mine

--
Jim Stirling:
"To argue with a person who has renounced the use of reason, is like administering medicine to the dead." - Thomas Paine
Feel free to tinker with any photos I post

James Stirling said:

Also 60mp will show sharpness issues more than 20mp

Click to expand...

Yep, this is actually one of the most important factors that people often overlook. People nowadays never really stop and try to critically analyze the reasons for things. For instance, why is the reciprocal rule for shutter speed 1/(whatever focal length)? Where did this rule come from? It's because a higher focal length puts more pixels on the subject, where any blur from shaking will be magnified and revealed. That means shooting a 50mm FoV at 1/50s SS with no IS will differ greatly, depending on whether you're shooting a 10MP sensor, or a 100MP sensor. On the 100MP sensor, the success rate of sharp photos will be noticeably lower.

It's also why Canon reinforced the camera body and tripod base plate when they introduced their first super high MP body - the 5DSR. It's because a higher MP body (more pixels on subject) will reveal micro vibrations much more readily than a tiny 20MP sensor. Again, this is all quite basic photography foundation, but in a forum open to hobbyists of all maturity levels, it is understandable that this is not common knowledge for some.

HicHic said:

James Stirling said:

Also 60mp will show sharpness issues more than 20mp

Click to expand...

Yep, this is actually one of the most important factors that people often overlook. People nowadays never really stop and try to critically analyze the reasons for things. For instance, why is the reciprocal rule for shutter speed 1/(whatever focal length)? Where did this rule come from?

Click to expand...

It's from olden days referring to 35mm film cameras. For me it has proven true when testing my own hand-holding ability using digital sensors from 1/2.3" up to 4/3" size. The rule proved to be the slowest reliable shutter speed is 1/(FF equivalent focal length). Though an old Kodak white paper I came across years back stated that the old rule needed to be at 5x faster shutter speed for true reliability. So the minimum 1/50 at 50mm FF should be 1/250 minimum for 100% reliable results.

My own extensive tests some 10 years back proved to me that I fitted the old rule at all focal lengths. So I was "normal", others may do better or worse. Ten years later I do worse at wide angle but OK at tele.

The old rule plus all the DOF charts and tables assume that the full frame result from whatever size film or sensor will be printed to 8x10 inch and observed by someone with normal vision from about 12 inches.

Computers and monitors of course make pixel peeping such an easy exercise so all those old rules don't make much sense now.

HicHic said:

It's because a higher focal length puts more pixels on the subject, where any blur from shaking will be magnified and revealed. That means shooting a 50mm FoV at 1/50s SS with no IS will differ greatly, depending on whether you're shooting a 10MP sensor, or a 100MP sensor. On the 100MP sensor, the success rate of sharp photos will be noticeably lower.

Click to expand...

If it was 5MP or 500MP then when printed to 8x10" size and viewed from 12" then you would not see any difference and the old rules would still follow.

HicHic said:

It's also why Canon reinforced the camera body and tripod base plate when they introduced their first super high MP body - the 5DSR. It's because a higher MP body (more pixels on subject) will reveal micro vibrations much more readily than a tiny 20MP sensor. Again, this is all quite basic photography foundation, but in a forum open to hobbyists of all maturity levels, it is understandable that this is not common knowledge for some.

Click to expand...

If pixel peeping had not been invented then that reinforcing would not have been necessary.

Guy Parsons said:

If it was 5MP or 500MP then when printed to 8x10" size and viewed from 12" then you would not see any difference and the old rules would still follow.

Click to expand...

Yes, because you're equalizing the details in both photos by printing small, hence proving my point more.

I recall a few OM3s had IS issues especially with longer lenses and had to be returned. Somewhere else in the forum this is reported.

HicHic said:

Guy Parsons said:

If it was 5MP or 500MP then when printed to 8x10" size and viewed from 12" then you would not see any difference and the old rules would still follow.

Click to expand...

Yes, because you're equalizing the details in both photos by printing small, hence proving my point more.

Click to expand...

I, personally, wouldn't call 8x10" "small".

I know we here all love to pixel peep, but most people view most pictures at smaller than 4x5" these days.

mmartel said:

Sam Bennett said:

This is okay as a general guideline and illustrates the relative difference between focal lengths, but ultimately what matters most is magnification. The above guideline is only really useful if you’re taking “normal” photos from “normal” distances. From there the differences are due to your subject being magnified. But the same applies if you’re moving closer to your subject. That’s why you can run into problems even with a 25mm (50 equivalent) if you’re very close to your subject, where you can throw that 1.5 sec figure out the window.

Click to expand...

Excellent point and caveat to add to the list!

Click to expand...

I get how magnifying the output invalidates the rule of thumb. I'm not seeing why distance from the sensor to the subject would have an effect. Just trying to think about the geometry of the situation, it doesn't seem like it would matter. Especially if it doesn't matter from 3 feet to infinity, and I've never heard that differences in that range do.

I expect that you guys are speaking from experience - I haven't tried any macro yet so I have none - can you explain it to me like I'm 8?

bobnn said:

HicHic said:

Guy Parsons said:

If it was 5MP or 500MP then when printed to 8x10" size and viewed from 12" then you would not see any difference and the old rules would still follow.

Click to expand...

Yes, because you're equalizing the details in both photos by printing small, hence proving my point more.

Click to expand...

I, personally, wouldn't call 8x10" "small".

Click to expand...

I said 8x10 simply because that is the size that all the old rules are based on. Wander away from that and anything can happen.

bobnn said:

I know we here all love to pixel peep, but most people view most pictures at smaller than 4x5" these days.

Click to expand...

Yup, 99.9999999999999999% of all photos taken these days are on those abominations called smartphones. Pity that the users are not also smart.

Not anti smartphones, simply just hate the things.

Guy Parsons said:

bobnn said:

HicHic said:

Guy Parsons said:

If it was 5MP or 500MP then when printed to 8x10" size and viewed from 12" then you would not see any difference and the old rules would still follow.

Click to expand...

Yes, because you're equalizing the details in both photos by printing small, hence proving my point more.

Click to expand...

I, personally, wouldn't call 8x10" "small".

Click to expand...

I said 8x10 simply because that is the size that all the old rules are based on. Wander away from that and anything can happen.

Click to expand...

Yes, I understood that you were talking about the "old rules" I was responding to HicHic.

Guy Parsons said:

Guy Parsons said:

I know we here all love to pixel peep, but most people view most pictures at smaller than 4x5" these days.

Click to expand...

Yup, 99.9999999999999999% of all photos taken these days are on those abominations called smartphones. Pity that the users are not also smart.

Not anti smartphones, simply just hate the things.

Click to expand...

Got it

bobnn said:

I, personally, wouldn't call 8x10" "small".

I know we here all love to pixel peep, but most people view most pictures at smaller than 4x5" these days.

Click to expand...

Maybe back in the 16MP days, I would consider it not small. 10-15 years ago?

James Stirling said:

danart said:

HicHic said:

ahaslett said:

The A7Rvis rated the same as an OM1 mk i and the A7CR a little behind the OM5/OM3. I find the A7CR is way behind my OM5 in my hands.

TLR I think your OM3 is faulty.

Andrew

Click to expand...

OP has A7RV. The RV IBIS is way better than your RIV and CR. I actually feel the A7RV has at least as good IBIS as OM-1, if not better. You should try it yourself and see.

Click to expand...

on what basis are you rating the A7rv as being “if not better” than the OM-1?

Show me a 6 sec exposure completely handheld unassisted (not bracing against anything) with an A7RV at 24mm

Click to expand...

I am not suggesting that you can't do this but without controlled repeatable tests of the same equipment by multiple people it is just a personal take of what you can do . I can guarantee that I would be lucky to get anywhere near this long a shot not braced and expect sharp results

Click to expand...

no, but as the user is the control that makes the comparison with two different cameras one can determine the effectiveness of one camera vs another in any particular example and this is the constant. Yes IBIS results will vary between users, but it is the user is the constant that can determine a reliable comparison.

James Stirling said:

James Stirling said:

https://www.dpreview.com/galleries/1260162627/photos/4470110/_8200065

Or a 2 second unassisted at 150mm equivalent

https://www.dpreview.com/galleries/1260162627/photos/4470124/

These exposures are very close to or hitting the actual quoted rating.

I’d be impressed if you could show me an actual example of the A7RV hitting an effective 5 stops…

Click to expand...

All such comparisons would show you are the respective hand holding abilities of the users not the absolute metric . There are other factors that come into play personal toleration for what is considered sharp. Also 60mp will show sharpness issues more than 20mp

Click to expand...

My personal comparisons are relating to images downsampled to build a meaningful one- I understand very well that an image at 60mp will be less effective when presented at full scale - however - when you downsample an image to 20Mp you have a level field. It is fallacy that low MP sensors produces sharper results than a higher MP sensor. This only applies to full scale comparisons - if the higher resolution sample is downscaled it produces the same result and all becomes equal.

If you have a more effective IBIS unit in a higher resolution camera. It will show when the image is downsampled.

I cannot achieve a 6 second handheld exposure with an A7rV that matches the output of my OM-1 when downsampled in the same way I cannot match the same exposures with better performing IBIS found in other lower resolution FF cameras… yes, all cameras will perform differently based on the user. But the performance gap is the same between 2 cameras and one user.

There are plentiful examples outside of my personal representations that present a wide and varied sample to establish fact.

You yourself have noted 2 seconds for M43 and 1 sec for FF. Whilst it might simply be a stop. At this extreme it represents far better performance. The gap in performance also applies in less demanding applications.

M43 IBIS will give you sharper results more often, more reliably. There is no worming around this reality.

The reputation of of IBIS in M43 comes down to a very simple fact. A smaller sensor is easier to stabilise effectively than a larger one.

The larger the surface area being stabilised, the less effective the stabilisation.

James Stirling said:

I am leery of CIPA's tested results as well particularly with regard to battery life which does not seem to be accurate for mirrorless cameras. However the CIPA rating for stablilsiation is a standardised test done in a controlled and repeatable manner. This is the number that every maker lists on spec sheet so one would assume they give it some value. Users of whatever camera can say they can hand hold for x amount of time and claim sharp results but which may well be the case . But it is not a verifiable ,data point not repeatable in a controlled manner that applies to multiple people it is your personal take on it . Just as my take on it is mine

Click to expand...

? A questionable rational that only seems to suit or prop up a personal opinion.

Whilst battery use will vary user to user - the controlled and repeatable results are easily qualified when testing two different cameras power consumption. If the same user applies the same approach they will recognise the difference in battery life. My OM-1 battery certainly outperforms my EM1.3 battery as reflected in the ratings despite both being able to take more shots than each rating states… I

What I find far more questionable is IBIS ratings. Questionable on the basis that I am in actual agreement 100% with your statement that IBIS functionality is not independent of the user. In fact it is very dependent. Agreed.

So… Where the individual is the control. How are IBIS result so dispersed in my case between FF cameras and M43 camera? Where M43 cameras with lower CIPA rating far out perform FF cameras with better CIPA ratings? From my deductions and understanding of CIPA ratings - these ratings are based on a 50mm standard. If the CIPA rating is 8 stops as found in the A7RV this equates to a 5 second HH exposure at full resolution as the “verifiable” data point.

I can simply not achieve this with any FF camera in any reliable way - but I can with M43 that have CIPA ratings less than 8 stops… I am the control and I practice more than most, and so I can get results at extremes - But I am the control is the crucial point.

Hence why I question the IBIS ratings… the only camera from the many that I have tried that can actually achieve anything close to this “verified “ data, are M43 cameras produced from 2019 > with the introduction of EM1X/3 until present.

HicHic said:

James Stirling said:

Also 60mp will show sharpness issues more than 20mp

Click to expand...

Yep, this is actually one of the most important factors that people often overlook. People nowadays never really stop and try to critically analyze the reasons for things. For instance, why is the reciprocal rule for shutter speed 1/(whatever focal length)? Where did this rule come from? It's because a higher focal length puts more pixels on the subject, where any blur from shaking will be magnified and revealed. That means shooting a 50mm FoV at 1/50s SS with no IS will differ greatly, depending on whether you're shooting a 10MP sensor, or a 100MP sensor. On the 100MP sensor, the success rate of sharp photos will be noticeably lower.

It's also why Canon reinforced the camera body and tripod base plate when they introduced their first super high MP body - the 5DSR. It's because a higher MP body (more pixels on subject) will reveal micro vibrations much more readily than a tiny 20MP sensor. Again, this is all quite basic photography foundation, but in a forum open to hobbyists of all maturity levels, it is understandable that this is not common knowledge for some.

Click to expand...

It is only a point if presenting the images at full scale. There is no issue if the image is downscaled to match the lower resolution sensor. That lower resolution sensor produce cleaner sharper images ( at high iso, lower shutter speeds) than higher resolution sensors has been debunked many moons ago. The only issue occurs when we are considering full scale comparison.

show me a 5 second exposure at 50mm that matches the CIPA ratings definition of 8 stops. You should by this measure be able to present it at full scale. If you can present it at 20mp I will sell all of my M43 gear tomorrow.

danart said:

It is only a point if presenting the images at full scale. There is no issue if the image is downscaled to match the lower resolution sensor. That lower resolution sensor produce cleaner sharper images ( at high iso, lower shutter speeds) than higher resolution sensors has been debunked many moons ago. The only issue occurs when we are considering full scale comparison.
show me a 5 second exposure at 50mm that matches the CIPA ratings definition of 8 stops. You should by this measure be able to present it at full scale. If you can present it at 20mp I will sell all of my M43 gear tomorrow.

Click to expand...

When people examine their high MP files to check for sharpness, they usually zoom in the image in the field to see - at the back of their EVF or screen. They're not gunna purposely not look, take the images home, and then downsize the images to a tiny 20MP, and then check for sharpness lol...by that time, it's already too late if the image is blurry. It's the same reason why in camera high res processing is so much better than the Sony cameras that require software. You don't know if it's sharp until you get home and combine the files to see.

Again, everyone shoots differently. You might be using your high megapixel cameras differently. If you prefer not checking sharpness in the field, and rather bring the images home and downsize before checking, then all the power to you. That's the great thing about this hobby, everyone is free to choose whatever process they enjoy! There's no right or wrong answer, just whatever makes the user happy.

HicHic said:

bobnn said:

I, personally, wouldn't call 8x10" "small".

I know we here all love to pixel peep, but most people view most pictures at smaller than 4x5" these days.

Click to expand...

Maybe back in the 16MP days, I would consider it not small. 10-15 years ago?

Click to expand...

The "most people" refers to most people and it is a truth that 99.99% or more of recently taken photos are viewed on smartphone screens, whatever size they happen to be.

HicHic said:

danart said:

It is only a point if presenting the images at full scale. There is no issue if the image is downscaled to match the lower resolution sensor. That lower resolution sensor produce cleaner sharper images ( at high iso, lower shutter speeds) than higher resolution sensors has been debunked many moons ago. The only issue occurs when we are considering full scale comparison.
show me a 5 second exposure at 50mm that matches the CIPA ratings definition of 8 stops. You should by this measure be able to present it at full scale. If you can present it at 20mp I will sell all of my M43 gear tomorrow.

Click to expand...

When people examine their high MP files to check for sharpness, they usually zoom in the image in the field to see - at the back of their EVF or screen. They're not gunna purposely not look, take the images home, and then downsize the images to a tiny 20MP, and then check for sharpness lol...by that time, it's already too late if the image is blurry. It's the same reason why in camera high res processing is so much better than the Sony cameras that require software. You don't know if it's sharp until you get home and combine the files to see.

Again, everyone shoots differently. You might be using your high megapixel cameras differently. If you prefer not checking sharpness in the field, and rather bring the images home and downsize before checking, then all the power to you. That's the great thing about this hobby, everyone is free to choose whatever process they enjoy! There's no right or wrong answer, just whatever makes the user happy.

Click to expand...

In what is actually being discussed there are certainly more correct answers

Anyone with a lot of experience with high resolution sensors understands tiny screen/evf reviews are not going to reveal much in the way of micro vibrations as well as other flaws all that resolution captures out in the field which negatively impacts a final full res image. They understand what is required and shoot accordingly to ensure the best possible result and do so with the understanding they will not fully grasp the result until they view it on a proper monitor

60mp sensor and a 20Mp sensor shot in identical settings capture an identical image as it relates to motion blur that would impact sharpness. It’s just the high resolution sensor magnifies it. It’s not adding anything.

Therefore there is no sharpness issue a 60mp sensor presents over a 20mp sensor. If anything the higher resolution sensor has less sharpness issues once it is downsampled.

Rarely is anyone capturing a fully useful 60mp image outside very controlled shooting scenarios anyway….However if you can confidently shoot at a particular shutter setting with a 20mp sensor you can be more than sure you are going to get an equal or better image with the higher resolution sensor once you downsample it.

You will get a better results by downsampling a 60MP image even when printing beyond what the native 60mp would offer. Downsampling will remove many of the flaws highlighted by the extra resolution - micro vibrations/motion blur, noise, moiré, and other sensor-level artefacts that high res introduces. Downsampling will ultimately give you a superior final result far more often than it won’t…The only question is the degree of downsampling required to get said superior result.

So it is a false equivalence that a 20mp sensor has an advantage as it relates to IBIS over a high res sensor…or a 60mp sensor has more sharpness issues than a 20mp sensor.

danart said:

60mp sensor and a 20Mp sensor shot in identical settings capture an identical image as it relates to motion blur that would impact sharpness. It’s just the high resolution sensor magnifies it. It’s not adding anything.

Click to expand...

Yes that's exactly what I'm saying. I think we are in agreeance!

Guy Parsons said:

HicHic said:

bobnn said:

I, personally, wouldn't call 8x10" "small".

I know we here all love to pixel peep, but most people view most pictures at smaller than 4x5" these days.

Click to expand...

Maybe back in the 16MP days, I would consider it not small. 10-15 years ago?

Click to expand...

The "most people" refers to most people and it is a truth that 99.99% or more of recently taken photos are viewed on smartphone screens, whatever size they happen to be.

Click to expand...

Actually you can find softness and blur on phone screens much more easily than a 8x10 print because you have the option to enlarge the photo on a screen. Imagine doing a pinching motion with your 2 fingers, but you start with a pinch and you spread out.

bobnn said:

mmartel said:

Sam Bennett said:

This is okay as a general guideline and illustrates the relative difference between focal lengths, but ultimately what matters most is magnification. The above guideline is only really useful if you’re taking “normal” photos from “normal” distances. From there the differences are due to your subject being magnified. But the same applies if you’re moving closer to your subject. That’s why you can run into problems even with a 25mm (50 equivalent) if you’re very close to your subject, where you can throw that 1.5 sec figure out the window.

Click to expand...

Excellent point and caveat to add to the list!

Click to expand...

I get how magnifying the output invalidates the rule of thumb. I'm not seeing why distance from the sensor to the subject would have an effect. Just trying to think about the geometry of the situation, it doesn't seem like it would matter. Especially if it doesn't matter from 3 feet to infinity, and I've never heard that differences in that range do.

I expect that you guys are speaking from experience - I haven't tried any macro yet so I have none - can you explain it to me like I'm 8?

Click to expand...

Try it. Take a landscape photo with a particular focal length. Keep lowering the shutter speed until you’ve reached the limit of what you can do while getting a stable shot. Then get as close as you can to something at the same focal length and do it over again. See if you notice a difference.