Question regarding mechanical shutter vs. e-shutter

[pdk42]

This is actually how I figured out that the rolling shutter is 50ms, by counting the bands and knowing that the light on a 50Hz grid flickers at 100Hz.

Anyway, the only way to avoid flicker using Electronic shutter is to capture the image at a shutter speed which is a multiple of 1/100 (or 1/120 if the powergrid runs at 60Hz), so 2/100, 3/100, 4/100 are all fine as you are always capturing a full cycle of the light.

The undocumented special sauce John Vickers is referring to is probably flash band compensation: https://eww.pavc.panasonic.co.jp/hdw/oi/AG-X2/html/AG-X2-X20_DVQP2773_eng/0099.html. It is a feature usually only implemented on camcorders and eliminates the uneven exposure you may get when flash photographers are shooting away in your presence or other abrupt bright flashes of light. This has nothing to do with the regular light cycle of AC powered lights. From the side effects you can deduce it simply repeats partial frames, taking the darker part of two different frames and stitching them together. Only applies to video capture though.

Note that on Panasonic cameras there is a feature which enables you to set up the camera in such a way that when you shoot video while the camera is set to a photo mode, it will automatically override your set shutter speed and switch to a multiple of 1/100 or 1/120, depending on what system frquency you have set (PAL or NTSC). Again, this only applies to video recording.

Thank you Michael. That's a helpful response. I wasn't aware of the flash band compensation, but since it's video only it's not likely that I'll ever need to worry about it since I don't do video.

As you say though, AC lighting will cause banding with the S5's electronic shutter when shooting stills at shutter speeds faster than the reciprocal of the frequency x 2, as my example demonstrates. There is no special sauce to fix it, apart from switching to mechanical shutter.

 

[dirk]

Had you been paying more attention on DPReview.com, you would understand the issues.

As I said above, I am not prepared to explain the issues to you here.
If you wish detailed explanations, ask at DPRevived.com. At DPRevived.com, there are experts who understand this material, who will not hesitate to correct me (or yourself) if I or you talk nonsense.

You are not a "details" person. You should avoid attempting to explain technical matters, and certainly in the absence of peer review.

@John Vickers

I do not know whether this is a personality issue here or that you are used to that kind of bad tone at DPrevived. No matter what the reason is, I want to make one point very clear:

We do not want to have this kind of discussion culture in our forums. I know that at DPrevived there is the wild west and every user can misbehave as he wants to and nobody cares.

But you are here at LMF, not at DPrevived. We do this VERY different. We do care about the way how we treat each other and how we communicate with each other. We are a friendly community and you better adapt to a better, more friendly communication style than you are used to at other forums.

As I said above, I am not prepared to explain the issues to you here.

But you are here at LMF. You commented in a thread at LMF and therefore you should discus it here instead of linking somewhere else.

 

[John Vickers]

Paul -

Oh dear.

This is actually how I figured out that the rolling shutter is 50ms, by counting the bands and knowing that the light on a 50Hz grid flickers at 100Hz.

I used a signal generator driving LEDs. I get 51.07ms on S1. It's similar on S5.

Anyway, the only way to avoid flicker using Electronic shutter is to capture the image at a shutter speed which is a multiple of 1/100 (or 1/120 if the powergrid runs at 60Hz), so 2/100, 3/100, 4/100 are all fine as you are always capturing a full cycle of the light.

That's the right idea.

There's a potential problem though. (There are a several, really, which have been discused at length on DPReview, mainly in a video context).

Suppose you are recording images under lighting with ~60Hz mains flicker. What actual shutter speed do you expect if your "shutter speed" setting is 1/60 ?

When shooting stills ?

When shooting video ?

When camera settings are changed ?

How accurate do you think the shutter speed needs to be ? Under what conditions ?

The undocumented special sauce John Vickers is referring to is probably flash band compensation: https://eww.pavc.panasonic.co.jp/hdw/oi/AG-X2/html/AG-X2-X20_DVQP2773_eng/0099.html. It is a feature usually only implemented on camcorders and eliminates the uneven exposure you may get when flash photographers are shooting away in your presence or other abrupt bright flashes of light. This has nothing to do with the regular light cycle of AC powered lights. From the side effects you can deduce it simply repeats partial frames, taking the darker part of two different frames and stitching them together. Only applies to video capture though.

No.

Note that on Panasonic cameras there is a feature which enables you to set up the camera in such a way that when you shoot video while the camera is set to a photo mode, it will automatically override your set shutter speed and switch to a multiple of 1/100 or 1/120, depending on what system frquency you have set (PAL or NTSC). Again, this only applies to video recording.

S1

---

and S5[ii] have [Flicker Decrease(Video)] which allows a preset 1/50, 1/60, 1/100, or 1/120 "shutter speed" setting when shooting video from PASM modes.

S1

---

have [Flicker Decrease(Photo)], which should fix Paul's flicker problem with mechanical shutter - the one he doesn't think he has - but I cannot make it work on S1. It's supposed to have the same effect as - say - Flicker Reduction (Photo menu) on some Nikon cameras or Anti-Flicker Shooting on Canon. It's easier to make it work when shooting with the viewfinder on a DSLR.

 

[pdk42]

Paul -

Oh dear.

What does that mean John? And are you going to comment on the two example shots I posted?

 

[John Vickers]

John, please feel free to post the explanation here. I will understand what you post and would not hesitate politely correct it if appropriate. For reference I am a physicist with substantial optical system experience and holder of the Royal Photographic Society Progress Medal. Chuck

Good.

It's good for this forum that there are folk around - such as yourself - who can analyse complex technical issues, and their engineering solutions.

But your medal is for the invention of stereolithography. Not for photography.

Have you thought much about the various issues related to avoiding artefacts in photography related to (still or moving) photography under lighting which varies periodically ? There are quite a few subtleties.

The problem here is Paul's misinformation, and his confidence in it. I have a big problem with misinformation. Unfortunately, many people take personal offence at being contradicted. So, contradicting misinformation (or worse, disinformation, not the case here) can lead to forceful intervention by moderators who would rather keep the peace and move on than figure out what's going on.

It would be helpful if you were to intervene when misinformation appears.

I have partially explained my reluctance to post explanations here. I have some frustration at being asked to re-litigate issues that have been resolved under Paul's nose on DPReview.com.

 

[pdk42]

The problem here is Paul's misinformation, and his confidence in it. I have a big problem with misinformation. Unfortunately, many people take personal offence at being contradicted. So, contradicting misinformation (or worse, disinformation, not the case here) can lead to forceful intervention by moderators who would rather keep the peace and move on than figure out what's going on.

Can you explain exactly what misinformation I'm guilty of propagating?

 

[John Vickers]

What does that mean John? And are you going to comment on the two example shots I posted?

The "Oh Dear" was some kind of mistake - it wasn't supposed to be there. I apologise for that.

And yes, I am.

 

[CharlesH]

There are quite a few subtleties.

Can you point me to the discussion in DPRevived? I don't follow that forum. Also can you let me know your background?

 

[pdk42]

I have partially explained my reluctance to post explanations here. I have some frustration at being asked to re-litigate issues that have been resolved under Paul's nose on DPReview.com.

John, I don’t know what exactly you are referring to here on DPReview, nor why it was resolved “under Paul’s nose”. I have not engaged much in DPR except for the L mount forum and I don’t recall any specific discussion I was involved in about electronic shutter and AC lighting. Perhaps you’re confusing me with someone else?

 

[dirk]

I have partially explained my reluctance to post explanations here.

Sorry, but this is your personal problem. If you have problems in discussing things in a thread in which you decided to participate, then simply do not post in this thread

For me this sound like a cheap excuse not to discus it here either because you have no clue what you are talking about or you only want to troll here around. Simple as that.

Can you point me to the discussion in DPRevived? I don't follow that forum. Also can you let me know your background?

Do not waste your time. John mentioned DPrevived only to put a link to that forum. There was no direct link to a specific thread, which is dealing with the issue of this thread. Otherwise he would have linked directly to it.

He mentioned only, that this was discussed in detail at DPReview, without linking to a specific thread neither. Since we do not know yet, whether the DPReview forum will be deleted or not, it is a waste of time to refer to a site, which very likely does not exist anymore in 4 weeks.

@John Vickers

I think you are just trolling here. You come here and write in a very arrogant way, claiming you would be a lot smarter than the rest of the world without any proove. Kind of risky, if you do not know who is sitting on the other side of the table and whether you can hold a candle to him.

I would say I am not Mr. Einstein, but even me know, that people who always claim that they are smarter than others are in reality not as smart as they think they are.

To make a long story short:

Stop trolling here around. Stop hiding behind other sites. Put the facts on the table and then we can continue to discus. If this already was discussed at DPreview, it should be very easy for you to copy & paste your argumentation into this thread. Please the whole text. Not only a link because of the above mentioned reasons.

If you do not want to do that, I am fine with this. But then you should not engage anymore in such threads to avoid to make a fool of yourself.

 

[jalywol]

For photography I stick with the mechanical shutter... but that might just be habit!

You might want to try EFCS or E-shutter if you are shooting longer lenses or at shutter speeds between 1/80s and 1/320s. Shutter shock is a real thing, and both EFCS and E-shutter will eliminate it. (Some lenses are more prone to it than others, and if you shoot wide, you may never encounter it...)

 

[pdk42]

You might want to try EFCS or E-shutter if you are shooting longer lenses or at shutter speeds between 1/80s and 1/320s. Shutter shock is a real thing, and both EFCS and E-shutter will eliminate it. (Some lenses are more prone to it than others, and if you shoot wide, you may never encounter it...)

Back in my early days with m43, I found shutter shock to be a real problem. The EP5 I had was terrible with some lenses. Olympus eventually added EFCS and then full electronic shutter and since then that’s what I’ve always used and have had no problems. I’ve not tested the S5 for shutter shock, but I would not be surprised if it suffered from it.

I now use electronic shutter most of the time, but switch to EFCS if shooting in artificial light to avoid banding. I know that EFCS can cause bokeh modifications when shooting at very wide apertures, but the fastest lens I have is the 50/1.8 and I’ve not seen any such effect with it.

 

[AlanC]

Shutter shock can definitely be a problem with the S5 and 70-300 when using full mechanical shutter, being at its worst around 1/125sec. But EFCS effectively eliminates it.

 

[jalywol]

Shutter shock can definitely be a problem with the S5 and 70-300 when using full mechanical shutter, being at its worst around 1/125sec. But EFCS effectively eliminates it.

I never even think of using either of my long lenses with full mechanical shutter. After my M43 experiences over the years, it became very clear that EFCS or Eshutter were the way to go with them (even the models with the newer style, reduced SS, shutter) and that's been what I do with the S5 also. Seems to work well for my purposes.