Posts: 5
Threads: 2
Joined: Jan 2022
Posts: 5
Threads: 2
Joined: Jan 2022
I have a few older timelapse sequences that I shot in aperture priority mode (for various reasons) and I have not been able to successfully remove all of the flicker using LRTimelapse.
Today I discovered that, even if two photos have the exact same reported shutter speed, aperture, and ISO values, one can STILL look noticeably brighter or darker—even in cases where the brightness in the scene did not change. That didn't seem like it should be possible, so I started digging into the EXIF data of the two photos to see if there were any significant differences between the two, and I discovered that they did indeed contain different values for the `SonyExposureTime` entry.
For example, a dump of the EXIF data for one of the files shows the following, notice that the shutter speed is reported as 1/30 but the "exposure time" is listed as 1/35.
```
Exposure Time : 1/30
Sony Exposure Time : 1/35
```
For the other photo in question, the exposure time and `SonyExposureTime` values were the same:
```
Exposure Time : 1/30
Sony Exposure Time : 1/30
```
I couldn't find any documentation on the `SonyExposureTime` value, but I'm assuming that my camera will only display values in 1/3 of a stop intervals, but internally the camera will try to use more granular values to get as close to the light meter reading as it can.
To test my hypothesis I computed the exposure difference between 1/30 and 1/35 shutter speeds:
```
log2((1/30) / (1/35)) = 0.22239 stops
```
And when I adjusted the photo with the 1/35 `SonyExposureTime` value by +0.22 stops of exposure in Lightroom, the two photos looked identical.
Anyway, you may already be aware of the `SonyExposureTime` entry, but if you're not I thought it might be worth looking into, as it could be a more accurate way to deflicker photos shot in aperture priority mode, or at least it could be a better starting point for the current deflicker algorithm. That is, better than starting with the reported shutter speed values, which may be off by some amount.
Also, obviously this only applies to Sony cameras, but I wouldn't be surprised if there isn't an equivalent EXIF entry for other camera models as well. I only have access to Sony files, so I can't confirm that.
Thanks for your consideration!
Today I discovered that, even if two photos have the exact same reported shutter speed, aperture, and ISO values, one can STILL look noticeably brighter or darker—even in cases where the brightness in the scene did not change. That didn't seem like it should be possible, so I started digging into the EXIF data of the two photos to see if there were any significant differences between the two, and I discovered that they did indeed contain different values for the `SonyExposureTime` entry.
For example, a dump of the EXIF data for one of the files shows the following, notice that the shutter speed is reported as 1/30 but the "exposure time" is listed as 1/35.
```
Exposure Time : 1/30
Sony Exposure Time : 1/35
```
For the other photo in question, the exposure time and `SonyExposureTime` values were the same:
```
Exposure Time : 1/30
Sony Exposure Time : 1/30
```
I couldn't find any documentation on the `SonyExposureTime` value, but I'm assuming that my camera will only display values in 1/3 of a stop intervals, but internally the camera will try to use more granular values to get as close to the light meter reading as it can.
To test my hypothesis I computed the exposure difference between 1/30 and 1/35 shutter speeds:
```
log2((1/30) / (1/35)) = 0.22239 stops
```
And when I adjusted the photo with the 1/35 `SonyExposureTime` value by +0.22 stops of exposure in Lightroom, the two photos looked identical.
Anyway, you may already be aware of the `SonyExposureTime` entry, but if you're not I thought it might be worth looking into, as it could be a more accurate way to deflicker photos shot in aperture priority mode, or at least it could be a better starting point for the current deflicker algorithm. That is, better than starting with the reported shutter speed values, which may be off by some amount.
Also, obviously this only applies to Sony cameras, but I wouldn't be surprised if there isn't an equivalent EXIF entry for other camera models as well. I only have access to Sony files, so I can't confirm that.
Thanks for your consideration!
-->
