Motion Cadence

[sam]

Yes it does.  So does the 1dc. 

To give a better idea of the distance and speed the pendelum travels, the length of the pendelum is roughly 6ft, and the outside black lines 6ft apart. So for most of the clips, the pendelum is only moving about 3ft. (1.5 ft on each side of the center) Also consider it basically comes to a stop at the peak of it's arc. 

In other words, natural human movements, like walking at a casual pace, picking up a coffee or an actor turning their head quite easily have the potential to be effected.   Maybe subconsciously people notice this. Or not ?  

[fuzzynormal]

On 4/15/2018 at 2:25 PM, sam said:

I set up these cams ... Details and settings are in the YT description.  

The way I tweak motion cadence is to just slow down the shutter a bit more than "normal."  I'm not convinced that the 180 degree rule is terribly important for hybrid video cams, y'know?

[TwoScoops]

Surprised how bad the 1DC looks tbh. That was the 4K motion jpeg too right? 

[sam]

Yes. 4k clog mjpeg. I went through a check list for each cam to make sure each setting was as it should be, double checked right before recording. 

Fuzzy, Did you check out the link to the cml thread? Or the tessive filter? If it was as simple as changing your shutter why would world class Dp's be discussing this topic?  Why would the tessive filter exist? But I do agree with you about not always needing to adhere to a 180 degree shutter.  Love the look and feel of the clips you posted fairly recently by the way! 

 

   

[fuzzynormal]

1 hour ago, sam said:

Fuzzy, Did you check out the link to the cml thread? Or the tessive filter? If it was as simple as changing your shutter why would world class Dp's be discussing this topic?  Why would the tessive filter exist?

World class DP's are not wrong.  They've earned the right to squint hard at their images and navel gaze about 'em.  They're upper echelon craftspeople that pixel peep because they can/should and have opinions that are relied upon for big budget productions.

I, on the other hand, am using a consumer camera that cost less that the Twizzler supply at their craft services table.  And, even then, the camera I'm using creates images that look ridiculously amazing.  The gap between what I have on my computer desk and what Roger Deakins is using is so close that even I sometimes have a hard time telling the difference.  And even if I could readily tell the difference, would it really matter anyway at my level?

For me the answer is "no."  I'm blessed with great cheap gear.  If it's not perfect it doesn't bother me all that much, honestly. I just work with it and make it the best I can.  Slowing the shutter is one trick I've used that I really like.  I dunno.  I'm a guy that likes noticeable motion blur.  Which, BTW, the tessive filter appears, to my eye, to allow.  So, a little more motion blur from a slower shutter just looks kind of "analog" to me --and that's neat-o.

I suggest that if anyone is bummed about how their camera renders motion cadence at a 180 shutter, try slowing it down a touch.  You might appreciate the results.  You may not.  But experimenting is allowed!

[Don Kotlos]

7 hours ago, fuzzynormal said:

I suggest that if anyone is bummed about how their camera renders motion cadence at a 180 shutter, try slowing it down a touch.  You might appreciate the results.  You may not.  But experimenting is allowed!

Yep I too find that using a slower than 180 shutter works great. 1/40 for 24p looks fine for fast motion.

[sam]

That all sounds nice..... happy where I'm at.... Roger Deakins.....cameras good enough....mmmm twizzlers...

8 hours ago, fuzzynormal said:

 If it's not perfect it doesn't bother me all that much, honestly.

But actions mean more than nice words on the internet.   Did our old cameras break? Why do we keep upgrading? Participating in this pixel peeping camera obsessed forum for literally thousands of hours because we can't see any difference?  

 if I applied your "fix" to the images from the test, the only thing that would change is the images would be even more blurry.  Only 3ft. of movement at relatively slow speed in absolutely perfect conditions, and there are clear differences.   

But I guess......now that every videographer, newsshooter, and youtuber =filmmaker,  and every non fixed lens video cam  = cinema camera, "good enough" is the new mantra.  But I"ll stop lamenting and get back to my good enough American life of eating my reheated mcdonalds, posting on facebook, while lounging in my lazy boy in camo sweatpants, content to know the well informed denizens of the internet have deemed my gh5 alexa worthy.

 

[Don Kotlos]

@sam first of all thanks for the very informative test. To me it seems that there are occasional "hiccups" where S8 and 1DC in lesser degree miss a step and skip to the next. So motion cadence (defined as the quality of motion?)  is affected not only by shutter speed,  shutter type &  codec but also consistency of frame times. 

But there is no need to get defensive... fuzzynormal just stated his opinion, and to a large extent I agree with him since shutter speed to me affects the motion cadence the most. Even in your test at the beginning I could not tell the difference, and only when I slowed down to 1/4 the speed I could point to the delays. I might not be the most sensitive person to the motion cadence, but I doubt the majority of people would notice either especially when not comparing it with other cameras. For example has anyone complained about the motion cadence of 1DC ?  

Now the tessive shutter article that you posted is also very interesting. I didn't know about it, but that at least partly explains our subjective opinion that a slower shutter than 180 can help with the motion blur. 

Let's keep this place on the positive side, and I welcome any tests that you do! 

Cheers

[fuzzynormal]

1 hour ago, sam said:

That all sounds nice..... happy where I'm at.... Roger Deakins.....cameras good enough....mmmm twizzlers...

But actions mean more than nice words on the internet.   Did our old cameras break? Why do we keep upgrading? Participating in this pixel peeping camera obsessed forum for literally thousands of hours because we can't see any difference?  

 if I applied your "fix" to the images from the test, the only thing that would change is the images would be even more blurry.  Only 3ft. of movement at relatively slow speed in absolutely perfect conditions, and there are clear differences.   

But I guess......now that every videographer, newsshooter, and youtuber =filmmaker,  and every non fixed lens video cam  = cinema camera, "good enough" is the new mantra.  But I"ll stop lamenting and get back to my good enough American life of eating my reheated mcdonalds, posting on facebook, while lounging in my lazy boy in camo sweatpants

How did you know I had camo sweat pants?

At any rate, anyone can do whatever they wanna.  Some like to strive to acquire what they consider the best gear for them, and perhaps they'll use motion cadence results to determine that.  Your tests are good for illustrating those details.

I'm not here to say what's right or wrong.  

Bottom line:  Gear cheap.  It make pictures good.  Many ways things be done.

[fuzzynormal]

1 hour ago, sam said:

the well informed denizens of the internet have deemed my gh5 alexa worthy.

I also got to say, when the major IQ difference is 3 stops of DR between the best camera on the market and a consumer camera, a Gh5 kind of is worthy of being considered a legitimate imaging device.  Not because it should be used in place of an Alexa, but because it could be used in a budget production --and it wouldn't be the weak link.

You can qualify my opinion because all the movies I grew up watching and loving typically were shot with film stock that only had, realistically, 12 stops of DR and most of that went out the window once crappy prints got distributed to the crappy movie theaters I used to go to. 

So that's my subjective background.  Know that I'm a nostalgic fan of beat-up analog stuff.  Therefore, modern gear amazes me and I tend to look at everything these days in consumer cameras as being well beyond good enough.

And FWIW, I'm shooting a short next month on S16 and Vision3 stock.  Not because it's going to have the best IQ, but because it's going to have a unique IQ.

[sam]

Hey! Not defensive.  Thats why I used the words " me" and "our" Not " you" 

This is the problem with the internet. No visual cues, tone or body language to go along with words.   

I was merely using sarcasm and hyperbole to show what happens when "good enough" becomes the norm.   As they say "ignorance is bliss"

[fuzzynormal]

49 minutes ago, sam said:

Hey! Not defensive.  Thats why I used the words " me" and "our" Not " you" 

This is the problem with the internet. No visual cues, tone or body language to go along with words.   

Don't worry about it.  That happens all the time.

[jahwah]

On 4/15/2018 at 2:25 PM, sam said:

I set up these cams (although more carefully aligned in the actual video) Details and settings are in the YT description.  

Thanks for the test, I really appreciate the time you took to do this.

I think it would also be helpful to see a scene with two actors walking, etc. with some decent lighting but keep the cameras anonymous (put the makes at the bottom of the YouTube description), I know it's not scientific, but I still think it helps keep your mind open while you judge the scene. Although in this case, it's pretty obvious that something is up. But who has time for that? 

[sam]

41 minutes ago, jahwah said:

Although in this case, it's pretty obvious that something is up. But who has time for that? 

Not sure I follow.   I own all these cams and tried to be fair to each  and all.   I also designed the test so anyone can replicate it themselves on the cheap. Take a 6ft. length of rope. Measure 2 marks on a wall 6ft wide. Fasten a weight, rock, etc.. to the rope and hang it from something like c stand or a ladder.  Set your camera to the same settings and dictance as described in the yt video.  (I would choose something large and round at the end of the rope so you can easily see whats happening).  Now let gravity do its thing.   

[sam]

https://frames-per-second.appspot.com

Anyone wanting to see how Fuzzy's recommendation of changing shutter speeds from the norm for film, can do so here.

Change shutter, fps, rate of motion and blur.  I chose the soccer ball on autumn background as it seemed easiest to see.  Then I chose 24fps at 100 pixels/second. I set one ball to realistic and the other to 1.5 (more blur).

By definition, not cadence, but an easy way to get a quick visual without a camera.

 

[fuzzynormal]

21 minutes ago, sam said:

recommendation of changing shutter speeds from the norm for film, can do so here.

I think maybe it's better to just check on whatever camera one has to see how changing up shutter speed renders the motion image.  There's a lot of other variables that determine how a motion image looks from camera to camera, so just experimenting with the gear one has is the best bet.  Eyeball it and do what looks right to you.

[Don Kotlos]

Yep, also it depends on the type of shot. For relatively static shots you can get away with almost any shutter angle. Faster camera movements can become quite distracting and then testing few shutter angles would be optimal. Sometimes you want the extra smoothness/dreamy look, others you want movement to be sharp. No reason to stick to one shutter angle unless you don't enjoy experimenting. 

[sam]

Of course.  This is obviously just a quick way to get a basic understanding so one can make an informed choice.

[kye]

I've just watched the video and unfortunately I detected a couple of potential issues with the test.

The fact the camera is panning is a problem because it means that the motion can look uneven (as if it's speeding up and slowing down) by having an uneven pan.  Ideally the cameras would all be stationary.

Also, I noticed repeated frames on both the S8+ and the 1DC.  It also looked like there was a skipped frame before the first repeated frame on the S8+.  Please note that it wasn't a repeated frame in the video file (which I downloaded from YT) - on both the examples of repeated frames the other cameras had non-repeated frames, so the YT video file appeared to be fine.

I wonder if the 1DC file was fine but the error was introduced in post by a slight time stretch?

The S8+ file (which seemed to combine a skipped frame with a repeated one) seems more likely to be related to the camera itself - maybe something was happening in the background and the camera app didn't get CPU preference at that moment.

It's a pity as the test seems like a good idea and would have taken quite some time to setup and edit and upload.

[jonpais]

It seems to me that @sam is confusing rolling shutter with whatever motion cadence is supposed to be.