Duplicate frames in avi recordings using Planetary mode

[PhilBlack]

Backyard version 2.1.2 trial, Camera Nikon D7500

Trial user of BYN getting to grips with all the options before purchase.

Whilst processing avi files produced in planetary mode using Registax software I have noticed that many of the individual frames are identical, typically only every 3rd to 5th frame is unique. I am guessing this is a Nikon issue in the way they process the liveview image but wondered if it was possible to slow down the capture rate so that more of the frames recorded are unique. Frames seem to be captured in the 80-100 fps range, although with a busier image during daytime testing this seemed to drop to around 50fps but with similar result re unique frames. This doesn't seem to be creating any issues other than larger files than should be required for the given number of unique frames.

Any thoughts or ideas would be appreciated

[admin]

This is odd.  How do you confirm there are identical?

[PhilBlack]

Initially just a visual observation scrolling through avi frames in Registax, but since tested by recording of digital stopwatch showing hundredths of a second, and then reviewing/scrolling the individual jpegs in photo viewer. 

 

[astroman133]

With my Canon T5i I get about 15 fps in Planetary or Frame & Focus modes. I would be surprised if Nikon delivered unique images at 6 times that rate. My guess is that BYN thinks that there is a new frame and downloads the old frame again to get that kind of rate. I am sure that Guylain will figure it out.

[admin]

BYN, just like BYE, receives a notification of a new image being available in the live stream and an in-camera memory pointer is provided to grab that image. If this is truly the case, it would mean that the Nikon SDK would be publishing an event of an image being ready more than once for the same image.

[PhilBlack]

Further observations, I have now looked at about 10 sets of images with a range of BYN settings. Altering the throttle value appears to have no noticeable effect on unique frame gathers, in the range 20ms to 100ms. Neither does changing shutter speed or iso value.

Changing the Lv dial on the camera from still to video distorts the onscreen image in BYN (width squeezed slightly) but when recording image corrects both onscreen and in the recordings. Significantly the number of unique frames has doubled, so whilst in my most recent tests BYN  gathered at 60fps I have just manually counted 30 unique frames within a 1 second period (video of my digital stopwatch) Obviously still something strange going on but I am happy with that.

Other observations.

Shutter and ISO changes work instantaneous. no need to exit from Lv

The LV image recording size is 1024 x 768

5x mode works? in that it zooms in massively, should it match the area of the zoom box? it is much smaller

Thanks for you help so far 

 

[PhilBlack]

A bit more info and experimenting,

I have attached a screenshot of planetary capture files (quality isnt brilliant as had to compress but just good enough), highlighting one second of jpegs approx 65 images, of these 25 are unique the remainder are duplicates. The images are of a stopwatch with the final figure being in hundredths of a second, I would have expected the final digit to progress by one or two hundredths in each subsequent image, but this does not happen. Instead duplicates are recorded then after a few images a new unique image is recorded. Are these actual duplicates? I think so, looking in very close detail at three successive duplicates the images are identical right down to the detail within the image noise, unique images either side have different noise patterns.

I also noticed that watching the stopwatch in BYN Planetary mode liveview, the image looks as good as looking at the stopwatch direct, ie progression of the time is smooth and uniform with the final digit constantly changing almost a blur, however, after pressing the capture button there is a noticable stuttering to the progression of the final digits.

I can attached further images or send a log file if that would help