What email address or phone number would you like to use to sign in to Docs.com?
If you already have an account that you use with Office or other Microsoft services, enter it here.
Or sign in with:
Signing in allows you to download and like content, and it provides the authors analytical data about your interactions with their content.
Embed code for: Percentage of Sharp Images vs Shutter Speed
Select a size
This article details the recreation of the curve of the percentage of sharp images vs shutter speed from KenRockwell.com.
Percentage of Sharp Images vs Shutter Speed
by Vincentius Timothy, on 23 November 2016
This post marks the first article in the photography section! Yay!
Today I stumbled upon a very interesting article in KenRockwell.com which discusses the importance of vibration reduction (VR) in Nikon lenses and image stabilization (IS) in Canon lenses. The advantages of VR or IS are generally discussed around the following curve:
The curve discussing the percentage of sharp images vs shutter speed, obtained from http://www.kenrockwell.com/tech/images/vr/vr-graph.gif. This image belongs to KenRockwell.com.
In the section "Where VR Shines", it is stated that:
"VR and IS give huge improvements in the sweet spot where the two curves separate. Shoot at about 1/2 - 1/15 of a second with normal lenses and you'll see the biggest night-and-day difference. Shoot faster and both are sharp, shoot slower and even IS gets blurry."
Naturally, I wondered: What is the likely percentage of obtaining sharp images given any shutter speeds? Of course, my nerdy guts told me that I've got to recreate the curve in MATLAB. I know I will not be satisfied unless I do that. And so the process begin.
Here is the specification of the curve.
The most likely function to model the curve is sigmoid function
The x-axis is reversed
The x-axis is logarithmically scaled with base 2
The x-axis ranges from ~1/512 s to 4 s
There are two curves: the one with VR or IS, and the other without VR or IS
With that specification in mind, the MATLAB code to recreate the curve is as follows:
traw = -9:0.01:2;
t = 2.^traw;
withvr = 1./(1+exp(log2((3.6.*t).^1.4)));
withoutvr = 1./(1+exp(log2((42.*t).^1.4)));
plot(t,withvr,'r',t,withoutvr,'k','LineWidth',1.5); grid on;
title('Percentage of Sharp Images vs Shutter Speed');
xlabel('Shutter Speed (seconds)');
ylabel('Percentage of Sharp Images');
legend('With VR or IS','Without VR or IS','Location','best');
Result & Discussion
The resulting curve from the MATLAB code above is as follows. You can see that the graph closely resembles the one form KenRockwell.com. Therefore we are successful in recreating the curve. Yay!
The curve discussing the percentage of sharp images vs shutter speed, recreated in MATLAB. Small version.
The curve discussing the percentage of sharp images vs shutter speed, recreated in MATLAB. Big version.
We can extend the code above to obtain the percentage of sharp images given any shutter speeds. The result is tabulated as follows. We can see that VR and IS lenses do really shine in the range of 1/2 to 1/15 seconds of shutter speed, as discussed in KenRockwell.com.
We have been successful in recreating the curve discussing the percentage of sharp images vs shutter speed from KenRockwell.com, as well as obtaining the percentage of sharp images given any shutter speeds. Bravo!