Banding and sensor readout speed G100

[2ndact scene1]

I am in the process of switching to LUMIX from Olympus for a variety of reasons. I now have a G9 and I was planning to get a G100D (I need 2 bodies) very soon. I will use them for event and theater photography. While not necessary most of the time, there are occasions that I have to shoot silently. And that raises the issue of banding under LED lights.

I was using older Canon FF mirrorless (R and RP) a couple of years ago and learned that they had very slow sensor readout speeds and under LED light, banding was in every image when shooting silently. When I moved to the Olympus EM1.iii, the problem went away due the better sensor readout speed and the sensor being 50% smaller. So far, the G9 is fine. But it just occurred to me that I should check if there any issues with a G100 (or alternatively the G95).

A related issue is the maximum G100 EFCS speed of 1/500. That should be sufficient for most movement on stage in a play but not for dance or a musical. I did check that the G100 can set minimum shutter speeds in auto ISO beyond 1/250 (the limit on the EM10.iv that I regretfully had to just sell). But it didn’t occur to me that going over 1/500 puts me into ES territory on the G100 and potentially introducing banding and I guess rolling shutter.

The alternative is getting a second used G9, which is quite inexpensive these days. But I was really hoping I could get something smaller.

A more expensive option would be an OM1 which I think would also have better eye and face tracking also but is not in my budget.

Does anyone have experience to the contrary?

 

[JakeJY]

It does have a flkr decrease setting (see page 329) which may help, although many Panasonic cameras have it too:

https://help.na.panasonic.com/wp-content/uploads/2023/02/DCG100_DVQP2173ZA_ENG.pdf

You may be able to do the same at other framerates by syncing shutter speeds.

There's no confirmation, but if the G100 is using the older 20MP sensor from the GX8/GX9, then it may be the IMX269, which has a readout speed at 12-bit of 24-27 fps.

https://www.sunnywale.com/uploadfile/2022/0423/IMX269AQR_Flyer_Awin.pdf

 

[2ndact scene1]

It does have a flkr decrease setting (see page 329) which may help, although many Panasonic cameras have it too:

https://help.na.panasonic.com/wp-content/uploads/2023/02/DCG100_DVQP2173ZA_ENG.pdf

You may be able to do the same at other framerates by syncing shutter speeds.

There's no confirmation, but if the G100 is using the older 20MP sensor from the GX8/GX9, then it may be the IMX269, which has a readout speed at 12-bit of 24-27 fps.

https://www.sunnywale.com/uploadfile/2022/0423/IMX269AQR_Flyer_Awin.pdf

Thanks for the reply.

Much of the discussion about sensor readout times revolves around distortion of moving subjects. Banding is not discussed near as much becuase it happens only in stills photography under LED or Flourescent light sources. Furthermore, as readout times get faster, the problem is going away. So it is tough issue to research.

In terms of anti Flicker. For the last ten years I have mainly been a Canon shooter. So that is what I am somewhat familiar with. Anti Flicker shooting on Canon DSLRs and mirrorless historically is a process where the camera syncs the timing of the shutter release with the frequency of the lighting source.

But what I am learning today is that the technology appears to have become more sophisticated in the most recent R series cameras (I own the original R and RP which are pretty old). Based on some quick research, I found that in order to eliminate banding, the R6II, and I think the R5II, have a feature that recommends a range of shutter speeds to the operator that will synchronize the shutter speed with the frequency of the light the camera is sensing and therefore minimize banding. The R6ii and R5ii actually have pretty fast sensor readout speeds. I am not sure if that imples that sensor readout speed may not be a complete solution to banding?

What I see on page 329 of the G100 manual (thanks for linking it) is the FLKR Decrease feature for video. I am not clear if it plays a role in stills photography and I cannot see any further references to it in the menu index. So the situation seems a little ambiguous.

As to the sensor specs you linked (and thanks for linking that too becuase it is a document I would never found otherwise), I see the reference to "1 / 30 s integration" which looks like it might be the sensor readout speed. But I am out way ahead of my headlights here! There is also a table in the document listing 18 drive modes. I see that drive mode 1 produces between 24 and 27 frames per second and the other 17 drive modes produce various FPS rates.

I think I am just going to take the more conservative course and get another G9. It will give me more features than a G100 for sure. But at the cost of lugging around two fairly big, heavy cameras. At least they can share batteries and pwer cables.

 

[yslee2]

You're a bit out of luck on this one; the G100 uses a slower sensor compared to the G9. Unfortunately all the faster sensors are on the bigger cameras.

 

[James Stirling]

I am in the process of switching to LUMIX from Olympus for a variety of reasons. I now have a G9 and I was planning to get a G100D (I need 2 bodies) very soon. I will use them for event and theater photography. While not necessary most of the time, there are occasions that I have to shoot silently. And that raises the issue of banding under LED lights.

A couple of test shots in the given shooting situation at different shutter speeds can often mitigate the issue completely . There is usually a sweet spot where the issue will not rear its ugly head

I was using older Canon FF mirrorless (R and RP) a couple of years ago and learned that they had very slow sensor readout speeds and under LED light, banding was in every image when shooting silently. When I moved to the Olympus EM1.iii, the problem went away due the better sensor readout speed and the sensor being 50% smaller. So far, the G9 is fine. But it just occurred to me that I should check if there any issues with a G100 (or alternatively the G95).

A related issue is the maximum G100 EFCS speed of 1/500. That should be sufficient for most movement on stage in a play but not for dance or a musical. I did check that the G100 can set minimum shutter speeds in auto ISO beyond 1/250 (the limit on the EM10.iv that I regretfully had to just sell). But it didn’t occur to me that going over 1/500 puts me into ES territory on the G100 and potentially introducing banding and I guess rolling shutter.
The alternative is getting a second used G9, which is quite inexpensive these days. But I was really hoping I could get something smaller.

A more expensive option would be an OM1 which I think would also have better eye and face tracking also but is not in my budget.
Does anyone have experience to the contrary?

 

[2ndact scene1]

I am in the process of switching to LUMIX from Olympus for a variety of reasons. I now have a G9 and I was planning to get a G100D (I need 2 bodies) very soon. I will use them for event and theater photography. While not necessary most of the time, there are occasions that I have to shoot silently. And that raises the issue of banding under LED lights.

A couple of test shots in the given shooting situation at different shutter speeds can often mitigate the issue completely . There is usually a sweet spot where the issue will not rear its ugly head

I was using older Canon FF mirrorless (R and RP) a couple of years ago and learned that they had very slow sensor readout speeds and under LED light, banding was in every image when shooting silently. When I moved to the Olympus EM1.iii, the problem went away due the better sensor readout speed and the sensor being 50% smaller. So far, the G9 is fine. But it just occurred to me that I should check if there any issues with a G100 (or alternatively the G95).

A related issue is the maximum G100 EFCS speed of 1/500. That should be sufficient for most movement on stage in a play but not for dance or a musical. I did check that the G100 can set minimum shutter speeds in auto ISO beyond 1/250 (the limit on the EM10.iv that I regretfully had to just sell). But it didn’t occur to me that going over 1/500 puts me into ES territory on the G100 and potentially introducing banding and I guess rolling shutter.
The alternative is getting a second used G9, which is quite inexpensive these days. But I was really hoping I could get something smaller.

A more expensive option would be an OM1 which I think would also have better eye and face tracking also but is not in my budget.
Does anyone have experience to the contrary?

 

[JakeJY]

It does have a flkr decrease setting (see page 329) which may help, although many Panasonic cameras have it too:

https://help.na.panasonic.com/wp-content/uploads/2023/02/DCG100_DVQP2173ZA_ENG.pdf

You may be able to do the same at other framerates by syncing shutter speeds.

There's no confirmation, but if the G100 is using the older 20MP sensor from the GX8/GX9, then it may be the IMX269, which has a readout speed at 12-bit of 24-27 fps.

https://www.sunnywale.com/uploadfile/2022/0423/IMX269AQR_Flyer_Awin.pdf

Thanks for the reply.

Much of the discussion about sensor readout times revolves around distortion of moving subjects. Banding is not discussed near as much becuase it happens only in stills photography under LED or Flourescent light sources. Furthermore, as readout times get faster, the problem is going away. So it is tough issue to research.

In terms of anti Flicker. For the last ten years I have mainly been a Canon shooter. So that is what I am somewhat familiar with. Anti Flicker shooting on Canon DSLRs and mirrorless historically is a process where the camera syncs the timing of the shutter release with the frequency of the lighting source.

But what I am learning today is that the technology appears to have become more sophisticated in the most recent R series cameras (I own the original R and RP which are pretty old). Based on some quick research, I found that in order to eliminate banding, the R6II, and I think the R5II, have a feature that recommends a range of shutter speeds to the operator that will synchronize the shutter speed with the frequency of the light the camera is sensing and therefore minimize banding. The R6ii and R5ii actually have pretty fast sensor readout speeds. I am not sure if that imples that sensor readout speed may not be a complete solution to banding?

First it helps to understand how banding happens with electronic shutter, this article explains it pretty well.

https://johnplatt.com.au/electronic-shutters-banding-rolling-shutter-explained/

Basically old florescent lighting generally flickers at twice the frequency of the local electricity. So for 50 Hz electricity (like in Australia) that is 100 Hz (for US 60 Hz it would be 120 Hz). This article below talks about the GH3 which has a 1/10 readout speed (so 10 Hz), which makes it easy to understand. So at speeds with banding, that means 100 Hz flicker speed / 10 Hz readout speed = 10 intensity peaks (or bands). You can count the bands in the example pictures to see that is the case!

https://m43photo.blogspot.com/2013/03/gh3-electronic-shutter.html

In such a situation, any shutter speed that is faster than the flicker speed can show banding, given each line only gets a fraction of the cycle of light. At shutter speeds equal to or slower than the flicker speed, as per the article, by selecting a multiple of the flicker speed, it's possible to sync things such that each line gets a complete cycle or multiple complete cycles of light, so no banding occurs.

Faster readout speeds help a bit by reducing the amount of bands that are visible, but even your E-M1 III with 1/60 readout speed does not completely eliminate the problem, just that it makes it so in the most typical examples (100 or 120 Hz) you get 2 or less bands, which is less noticeable.

What I see on page 329 of the G100 manual (thanks for linking it) is the FLKR Decrease feature for video. I am not clear if it plays a role in stills photography and I cannot see any further references to it in the menu index. So the situation seems a little ambiguous.

As mentioned, you can simulate the same thing by smartly choosing your shutter speeds.

As to the sensor specs you linked (and thanks for linking that too becuase it is a document I would never found otherwise), I see the reference to "1 / 30 s integration" which looks like it might be the sensor readout speed.

No, that is only the speed at which that sensitivity measure was taken, not the readout speed.

But I am out way ahead of my headlights here! There is also a table in the document listing 18 drive modes. I see that drive mode 1 produces between 24 and 27 frames per second and the other 17 drive modes produce various FPS rates.

Yes, that is what I am referring to.

I think I am just going to take the more conservative course and get another G9. It will give me more features than a G100 for sure. But at the cost of lugging around two fairly big, heavy cameras. At least they can share batteries and pwer cables.

As another mentioned maybe a test can be done? If you still have the old cameras that you know have flicker, find which lights cause the flicker. Then test the camera under the same lights. It's best if you can find a portable light that does this so you can perhaps bring it to a store to test (and also something that can enclose it to block ambient light from affecting results).

 

[Tom Caldwell]

Brings up a few points - the only banding I have seen in Australia with electronic shutter was in a son's house with old fluorescent lights. LED have not been a nuisance.

I don't know the type of theatre lights that I encounter but it is probably immaterial as I am not really using electronic shutter and my distances from the stage make the quiet mechanical shutters effectively soundless.

Not an expert on the effect of Hertz but the Australian Hertz does not match US Hertz and may be easier to tune out electronic banding here than there. Thrown in for what it is worth. There are worked instances on the web.

I don't think I have actually used the G100 in theatre yet - I may have and will check in due course.

 

[Tom Caldwell]

I am in the process of switching to LUMIX from Olympus for a variety of reasons. I now have a G9 and I was planning to get a G100D (I need 2 bodies) very soon. I will use them for event and theater photography. While not necessary most of the time, there are occasions that I have to shoot silently. And that raises the issue of banding under LED lights.

A couple of test shots in the given shooting situation at different shutter speeds can often mitigate the issue completely . There is usually a sweet spot where the issue will not rear its ugly head

I was using older Canon FF mirrorless (R and RP) a couple of years ago and learned that they had very slow sensor readout speeds and under LED light, banding was in every image when shooting silently. When I moved to the Olympus EM1.iii, the problem went away due the better sensor readout speed and the sensor being 50% smaller. So far, the G9 is fine. But it just occurred to me that I should check if there any issues with a G100 (or alternatively the G95).

A related issue is the maximum G100 EFCS speed of 1/500. That should be sufficient for most movement on stage in a play but not for dance or a musical. I did check that the G100 can set minimum shutter speeds in auto ISO beyond 1/250 (the limit on the EM10.iv that I regretfully had to just sell). But it didn’t occur to me that going over 1/500 puts me into ES territory on the G100 and potentially introducing banding and I guess rolling shutter.
The alternative is getting a second used G9, which is quite inexpensive these days. But I was really hoping I could get something smaller.

A more expensive option would be an OM1 which I think would also have better eye and face tracking also but is not in my budget.
Does anyone have experience to the contrary?

That maybe true but it’s a chicken and egg situation. I don’t own a copy of the camera and even if I did, I don’t have easy access to the venues. If I wait until I have an event to shoot, it is a bit late to find out it doesn’t work. And Panasonic dealers where you could experiment with a copy are few and far between in the USA. But as I think about it, I am planning to go to New York for a few days in late March. So a trip to B&H Photo will probably be in my itinerary anyway.

I cannot give you a link but there is a good thread on electronic banding on the web - I found and then re-found it easily enough. It discusses tuning out the banding shutter speed - which may or may not be useful. The US Hertz rates are different from the Australian ones which are easier to work-around shutter speeds.

But frankly it has been years since banding bothered me on my GM5 with similar shutter restrictions and I have not noticed it on any of my G100 bodies so far. But the emphasis is on "not noticed" of course. I certainly was not going looking for it.

 

[2ndact scene1]

Brings up a few points - the only banding I have seen in Australia with electronic shutter was in a son's house with old fluorescent lights. LED have not been a nuisance.

I don't know the type of theatre lights that I encounter but it is probably immaterial as I am not really using electronic shutter and my distances from the stage make the quiet mechanical shutters effectively soundless.

Not an expert on the effect of Hertz but the Australian Hertz does not match US Hertz and may be easier to tune out electronic banding here than there. Thrown in for what it is worth. There are worked instances on the web.

I don't think I have actually used the G100 in theatre yet - I may have and will check in due course.