Luminous Landscape Forum

Equipment & Techniques => Mirrorless Cameras => Topic started by: Guillermo Luijk on December 23, 2014, 06:59:29 pm

Title: Fuji X-T1 1/32000 rolling shutter calculations
Post by: Guillermo Luijk on December 23, 2014, 06:59:29 pm

In another forum the following image of a X-T1 at 1/32000 has appeared. High electronic speeds such as 1/32000 are fine for exposure, but not so much to freeze moving objects because of the rolling shutter effect derived:

(http://www.guillermoluijk.com/misc/rollingshutter_calculos.jpg)

If the entire sensor is read in 1/32000th of a second, one line at a time, having 3264 lines, that means 1/10th of a second to read the entire sensor, pretty slow. But I have made some calculations (subject to several assumptions) and the result has been around 1/30th.

Does anyone know the real time it takes to read the entire sensor in the X-T1 working at 1/32000?

I did the following:

• Assumption 1: arrow diameter 8mm. This coincides with 8px in the original 1280px wide image.
• Ratio between arrow speed/sensor readout speed: 60:133
• Assumption 2: arrow speed 200 Km/h
• 200 Km/h = 55.556 mm/s. Since in the image 1px~1mm the arrow flies horizonally at 55.556 px/s
• That speed translates (60:133) as 25.063 px/s vertical readout
• Assumption 3: the original 3:2 image was cropped up and/or down, I added lines to complete 853 px height (the image was 1.280 width).
• 853px at 25.063 px/s requires 0,034s -> vertical readout speed = 1/30th of a second

What do you think? 1/10th? 1/30th?

Regards

Title: Re: Fuji X-T1 1/32000 rolling shutter calculations
Post by: David Sutton on December 24, 2014, 04:00:21 pm

Hello Guillermo. There is a test on the X100T here (http://janssico.com/2014/12/x100ts-electronic-shutter-speed-analysis/). Pretty much a similar conclusion to yours. What matters to me is can I get sharp pictures hand holding the XT1 using the electronic shutter with stabilisation off? I hope to have a look at that in the next day or two.
David
(As I mentioned in another thread, having spent just a short time with the electronic shutter, my guess is that it's close to 1/30 sec)

Title: Re:
Post by: Guillermo Luijk on December 24, 2014, 06:59:51 pm

Amazing I reached the same figure (1/30th) with so many assumptions. I wonder how can it be possible though if one line is read at a time. It seems up to 3 lines are read at a time since 1/32000 for 3264 lines shouldn't be faster than 1/10th. Just guessing.

Thanks David.