Denniscasey
Well-known member
I know this is one of those "it depends" or "personal preference" questions, but let me give a few parameters: (1) subjects are rapidly moving race cars; (2) objective is to minimize blur of car without making motion appear "unnatural" (i.e. jerky).
This is part of my ongoing quest to do a better job capturing race car action as I become more critical of my previous efforts. Races are few and far between so I have tried experiments filming cars on freeways. However, I think motion is too slow to see differences I would expect. For instance, only setting where motion appeared unnatural to me (panning with cars) was 1080p30 with both 1/60 and 1/500 shutter. Surprisingly to me, 1/500 shutter at both 1080i (60 fields) and 720p60 while panning looked to have natural motion. When I did not pan (cars moving at right angles) car motion was blurry at both shutter speeds (just tried extremes of shutter speed) although individual frames did show much sharper definition at 1/500. I have also experimented on tree leaves blowing in wind (not conclusive) and ceiling fan motion (conclusive that 1/30 blurs more than natural eye sight, 1/250 gave jerky action, but is this a good simulation of race cars with panning??).
I have thought more about forum comments and read more articles essentially about spatial and temporal resolution so understand more about theory of frame rate and shutter speed impact on "sharpness" and "human eye perceived motion." So, theory is getting clearer, but I don't have experience level of forum members and races are few and far (distance) between so chances to experiment are limited. I am going to drive a ways to a race type I don't really care about to do more experiments but would appreciate some guidelines from experiences you may have had--for instance, 1/500 will definitely give impression of jerky motion (maybe I did not see this because of relatively slow speed of cars and way I was panning?). As another example, in Barry's HMC book page 20 it says "for a stuttery/choppy action sequence (..."Saving Private Ryan" effect) try a fast shutter speed such as 1/250 or 1/500."
I know I am just touching one factor (shutter speed) and there are many factors like panning, frame rate, etc that are going to impact tradeoff between "natural" motion and sharpness. If you say definitely 1/500 should not be used, but 1/250 might be used I will not include 1/500 in my range of experiments.
Based on past forum postings as well as additional reading I think I have settled at least on using 720p60 for frame rate (one factor is that my output is Blu-ray discs) so that will be one constant in my range of experiments. Thank you, Dennis
This is part of my ongoing quest to do a better job capturing race car action as I become more critical of my previous efforts. Races are few and far between so I have tried experiments filming cars on freeways. However, I think motion is too slow to see differences I would expect. For instance, only setting where motion appeared unnatural to me (panning with cars) was 1080p30 with both 1/60 and 1/500 shutter. Surprisingly to me, 1/500 shutter at both 1080i (60 fields) and 720p60 while panning looked to have natural motion. When I did not pan (cars moving at right angles) car motion was blurry at both shutter speeds (just tried extremes of shutter speed) although individual frames did show much sharper definition at 1/500. I have also experimented on tree leaves blowing in wind (not conclusive) and ceiling fan motion (conclusive that 1/30 blurs more than natural eye sight, 1/250 gave jerky action, but is this a good simulation of race cars with panning??).
I have thought more about forum comments and read more articles essentially about spatial and temporal resolution so understand more about theory of frame rate and shutter speed impact on "sharpness" and "human eye perceived motion." So, theory is getting clearer, but I don't have experience level of forum members and races are few and far (distance) between so chances to experiment are limited. I am going to drive a ways to a race type I don't really care about to do more experiments but would appreciate some guidelines from experiences you may have had--for instance, 1/500 will definitely give impression of jerky motion (maybe I did not see this because of relatively slow speed of cars and way I was panning?). As another example, in Barry's HMC book page 20 it says "for a stuttery/choppy action sequence (..."Saving Private Ryan" effect) try a fast shutter speed such as 1/250 or 1/500."
I know I am just touching one factor (shutter speed) and there are many factors like panning, frame rate, etc that are going to impact tradeoff between "natural" motion and sharpness. If you say definitely 1/500 should not be used, but 1/250 might be used I will not include 1/500 in my range of experiments.
Based on past forum postings as well as additional reading I think I have settled at least on using 720p60 for frame rate (one factor is that my output is Blu-ray discs) so that will be one constant in my range of experiments. Thank you, Dennis