Rolling Shutter CMOS Artifacts

[androbot2084]

I was watching some paparazzi footage on the Hollywood TV show Xtra Xtra and Ive never seen so much ruined footage in my life. Everytime a flash bulb popped the screen would spit up into an overexposed and underexposed mess. JVC's split screen problem palled in comparison to this defect. Now I am not saying that the Sony XD CAM is total garbage. Under controlled lighting conditions or if you can avoid these extremes I am sure it is a fine camera. But if you belong to the paparazzi the camera will be totally useless unless you are a daytime shooter.

[Stevet]

I guess you'll have to tell everyone not to use their flash. ;)

[Noel Evans]

What you describe is not what I have witnessed with the XD EX. I didnt see HUGE overexposures or unders that ran long to ruin a shot.... the effect I witnessed while not ideally what I would like was simply the flash throwing out to only partial frame and was so fast etc. I think for the ENG crowd its a small annoyance of little consequence. In the narrative though, if this occurs trying to get a particular shot with one of many things that cause a flash it may be of greater concern.

[LuckyStudio 13]

http://web.mac.com/fini1/iWeb/Site%20135/4k%20Red%20Camera.html

If you deemed the above music video is unusable, then you should stick with CCD cameras.

[3xm]

WOW :shocked:...after seeing that music video I am now VERY CONCERNED about CMOS artifacts... the "banding" that occurs with the strobes is pretty obvious to my eyes!:huh: ....damn....(and we're supposed to get our red camera in January... )

[Stevet]

Hmm...
I'm waiting for Barry to jump on this thread.

Barry does bring up a good point to understand what you're getting into before you buy something. Having said that, I crossed my fingers with the EX1, and bought one.

So far, I have not done all the tests, but nothing has bothered me so far.
I will try some flashes using different frame and shutter speeds.

Also, I believe RED has really tried hard to minimize any issues with the rolling shutter.

[Jason Wilson]

http://web.mac.com/fini1/iWeb/Site%20135/4k%20Red%20Camera.html

If you deemed the above music video is unusable, then you should stick with CCD cameras.


damn thats not cool at all....I know if I were the client I would not be pleased with that banding at all... its sad a camera that nice has that effect

[semila58]

Holy sh** that's bad. So there may be a way to sync strobes with the RED, but is there any way to do that on the EX1? I'm assuming not, so what options would there be to produce and capture a satisfactory strobe effect in a controlled environment (aside from pulling out the HVX, of course)? Looks like the ENG crowd is still screwed though...

[ESTEBEVERDE]

WOW!

Holly Holly Shit!

[Barry_Green]

No surprise -- that's pretty much exactly what I would have predicted. And DXmetal's comment is exactly spot-on: if that particular effect is unacceptable to you, stick with a CCD and stay away from a rolling shutter camcorder. If you're going to be shooting where flashes are going off, and that effect doesn't bother you, feel free to consider rolling shutters.

As Steve said, Red has tried to control the rolling shutter artifacts. I suspect the effect would have been hideously worse had it been from an HV20 or other un-optimized rolling shutter cam.

[Barry_Green]

damn thats not cool at all....I know if I were the client I would not be pleased with that banding at all... its sad a camera that nice has that effect

EVERY camera with a rolling shutter is going to have that effect. And lots of other effects. If it has a rolling shutter, it's going to exhibit rolling shutter effects -- the company may be able to minimize the effect somewhat, but it's still going to happen under certain circumstances.

The longer these cameras are on the market, and the more people use them in more circumstances, the more reports we'll see of places where it worked and places where it failed.

All I've been trying to say is: educate yourself as to what the effects are, and make an intelligent, informed decision as to whether you find the effects acceptable. The rules of video acquisition have changed, and you'd better know what you're getting into.

[ESTEBEVERDE]

How bad is it when you are filming say the US National Olympic Trials where flashes will be going off like crazy at the awards and maybe even during some of the competitions!?!?!?!

Holly Crap That Sucks!!!

[BobDiaz]

The perfect technology does not exist, with CCDs, you don't have the rolling shutter problem, BUT just point the CCD camera at a very bright light and you risk getting a vertical line when the CCD column is overloaded.

Tube cameras had image lag, remember the comet trails caused by bright lights?

As long as you know the limits of any given technology, you are buying something with an understanding what to expect from the camera. If you don't like it, don't buy it.

The Panasonic HVX-200 has been around for about 2 years, so we know what it's like, but the Sony EX-1 is still new and we're only starting to discover it's strong and weak points.


Bob Diaz

[Stevet]

Bob, that was a very level headed response. I always enjoy reading your creative posts.

[Barry_Green]

How bad is it when you are filming say the US National Olympic Trials where flashes will be going off like crazy at the awards and maybe even during some of the competitions!?!?!?!
Well, it depends on how much the flash illuminates the frame. If you're talking about a stadium where flashbulbs are going off in the crowd, it'll probably be a non-issue. As long as each individual flash isn't radically changing the brightness of the scene.

Now, a boxing match, with reporters all around the ring setting off flashes, or at the awards podium, or at a press conference -- yes, that would be a worst-case example.

The point is, as I tried to spell out in my article, different sensors will have different artifacts, and while we're "used to" the vertical smear of a CCD, we're only starting to discover the impact that the rolling shutter is going to have. It *is* going to impact people's work, and the sooner you learn what it can or cannot do, the sooner you'll be prepared to make an intelligent decision as to whether a particular camera will work for your particular scenario. Perhaps the hardest part is getting people to understand that we no longer have a "one size fits all" scenario. Now we have cameras that simply can't handle certain types of shots.

I think a lot of shooters are in for some surprises, until this issue gets more well known.

[androbot2084]

Actually the paparazzi footage I saw was worse than the music video footage as far as rolling shutter artifacts are concerned.
But then we are talking about nightime footage with a barage of flashbulbs going off from the still photographers. That being said as far as the Olympics were concerned the problem with the 2004 Olympics was that the 1080i interlace footage overloaded the compression engine to such a degree that unacceptable blocking artifacts were present anytime there was fast motion. However the Sony XDCAM has progressive scan shooting modes that will reduce these artifacts. Even for 1080i networks a cameraman can choose the 1080p30 shooting mode which may not handle fast motion very well but at least the picture will not block up. Of course the Panasonic HVX-200 has a lot of progressive shooting options also.

[dvInsight]

I posted the following over at DV Forums following a private email I sent to a friend on the East coast and you would have thought I said something about their first born:

I have a very open mind regarding the Sony EX-1. I wish the HVX200 looked that good. I mentoned one thing that I saw while demoing the camera. It was not an in-depth test.

Basically the dealer had a little set (just like they do at trade shows) and I had zoomed in to grab focus on a child's ball and toys. I loved the image. Then I accidentally kicked the tripod with my toe and saw the image wobble. It was a "what the f..k" thing. So I did it again, and again, and the damn image moved like jello. The HVX sitting 1 foot away zoomed in on the same thing didn't. The HVX looked just like what you'ed expect when you kick the sticks. It just moved, not wobbled.

Now any shot that you accidentally kick the sticks on is a lost shot, it is going in the can, so it is not a deal breaker for the EX-1, just an observation. I did some slow pans, faster, and some whips. I did not see any issues, but I was by no means testing the camera, just making some observations.

The salesman, who knew that I had an HVX, and knows my interest in getting the EX-1 said that he had a customer who did some green screen work, so this info is now twice removed, and they said that the edges fell apart compared to the HVX. I didn't see it, so I really don't know (shoot me for mentioning it).

I really want to see green screen recorded to the chip at 4:2:0 compared to out the SDI at 4:2:2. I expect the 4:2:2 to look better than the HVX because of the resolution of the camera. But, when shooting to the chips, I do wonder what the 4:2:0 will do. That is about it regarding green screen.

Based on my 20 to 30 minutes with the camera, I like it, but will need more info to be sold. I do not like the buttons and their layout. Sure, a lot of this stems from muscle memory from my HVX and I can relearn. I can't expect it to be exactly like my HVX.

The thing that I disliked the most was that the focus assist button is on the right side of the camera by the record button. I would have greatly preffered it on the left near the focus ring. The buttons and switches felt more consumer than pro. I say this only in comparison to the HVX. Heck, it is a prosumer camera after all.

So besides the flash issue, I can surmise that a sharp or high frequency hit to the sticks or camera body might introduce a wobble on a static shot.

Again, not necessarily a deal breaker for some, just like the noise issue isn't for some of us with the HVX.

I am still waiting for the camera that can give me the advantages of both without the negative of either. Until then I shot my HVX if appropriate for the job and rent other camera's when the HVX is not quite right.

Rob

[dvInsight]

As a follow up to the above post, I will give new users far more slack. Sometimes all anyone is trying to do is add some info. No hidden agenda like my friend and I from the east were trying to convey over at DVforumns.

I got to see what it was like being a "junior member" some place else.

Rob

[reem12]

i guess shooting a wedding next to a photogrpher is out of the question huh?

[philip bloom]

I posted the following over at DV Forums following a private email I sent to a friend on the East coast and you would have thought I said something about their first born:

I have a very open mind regarding the Sony EX-1. I wish the HVX200 looked that good. I mentoned one thing that I saw while demoing the camera. It was not an in-depth test.

Basically the dealer had a little set (just like they do at trade shows) and I had zoomed in to grab focus on a child's ball and toys. I loved the image. Then I accidentally kicked the tripod with my toe and saw the image wobble. It was a "what the f..k" thing. So I did it again, and again, and the damn image moved like jello. The HVX sitting 1 foot away zoomed in on the same thing didn't. The HVX looked just like what you'ed expect when you kick the sticks. It just moved, not wobbled.

Now any shot that you accidentally kick the sticks on is a lost shot, it is going in the can, so it is not a deal breaker for the EX-1, just an observation. I did some slow pans, faster, and some whips. I did not see any issues, but I was by no means testing the camera, just making some observations.

The salesman, who knew that I had an HVX, and knows my interest in getting the EX-1 said that he had a customer who did some green screen work, so this info is now twice removed, and they said that the edges fell apart compared to the HVX. I didn't see it, so I really don't know (shoot me for mentioning it).

I really want to see green screen recorded to the chip at 4:2:0 compared to out the SDI at 4:2:2. I expect the 4:2:2 to look better than the HVX because of the resolution of the camera. But, when shooting to the chips, I do wonder what the 4:2:0 will do. That is about it regarding green screen.

Based on my 20 to 30 minutes with the camera, I like it, but will need more info to be sold. I do not like the buttons and their layout. Sure, a lot of this stems from muscle memory from my HVX and I can relearn. I can't expect it to be exactly like my HVX.

The thing that I disliked the most was that the focus assist button is on the right side of the camera by the record button. I would have greatly preffered it on the left near the focus ring. The buttons and switches felt more consumer than pro. I say this only in comparison to the HVX. Heck, it is a prosumer camera after all.

So besides the flash issue, I can surmise that a sharp or high frequency hit to the sticks or camera body might introduce a wobble on a static shot.

Again, not necessarily a deal breaker for some, just like the noise issue isn't for some of us with the HVX.

I am still waiting for the camera that can give me the advantages of both without the negative of either. Until then I shot my HVX if appropriate for the job and rent other camera's when the HVX is not quite right.

Rob

Are you sure it wasn't the OIS that was on. That gives a Jello effect if you bash the camera.

[David S.]

A question for those familiar with the RED camera.

With so much technology in this cam, why elect a rolling shutter configuration?

[anchoryanker]

I posted the following over at DV Forums following a private email I sent to a friend on the East coast and you would have thought I said something about their first born:

<SNIP>


I saw that, I'll never understand why people get so defensive about this stuff, you'd almost think you were insulting their favorite sports team from some of the reactions. I love my DVX, but I'm also aware of it's shortcomings and can discuss the pros and cons objectively.

This is just equipment, I see no reason to be so attached to it emotionally that you have to shout down anyone who points out any potential flaws. I'll bet that if your post had been a glowing review they would have eaten it up!.

The only reason I can see for that mindset is that some people derive part of their identity from their camera(much like some people and their cars) and any perceived slight against it is also considered a slight against them. I guess in some ways these cameras are like any other luxury "status" good, like a Rolex or Ferrari.

At least that's my 2-bit psychoanalysis of the reason why.

[ESTEBEVERDE]

And they are protecting their ego from being hurt should they feel they made the wrong decision with such a high ticket item.

[BobDiaz]

I saw that, I'll never understand why people get so defensive about this stuff, you'd almost think you were insulting their favorite sports team from some of the reactions. I love my DVX, but I'm also aware of it's shortcomings and can discuss the pros and cons objectively.

This is just equipment, I see no reason to be so attached to it emotionally that you have to shout down anyone who points out any potential flaws. I'll bet that if your post had been a glowing review they would have eaten it up!.

The only reason I can see for that mindset is that some people derive part of their identity from their camera(much like some people and their cars) and any perceived slight against it is also considered a slight against them. I guess in some ways these cameras are like any other luxury "status" good, like a Rolex or Ferrari.

At least that's my 2-bit psychoanalysis of the reason why.

I agree and it's too bad that people have such extreme views about cameras, as if Camera_A is perfect, but Camera_B is total junk. Anyone who questions this goes to digital hell...

I've seen outstanding work created with the HVX-200 and I'm starting to see outstanding work created with the EX-1. Both cameras have strong points and both cameras have weak points, STILL users are able to produce great results with either camera. It's all a matter of person needs and wishes that defines which camera it best for each person.

A camera is a tool and just because someone selects a different tool from others, that does NOT make them a better or worse person than someone else.


Bob Diaz

[stephen_nugent]

If it has a rolling shutter, it's going to exhibit rolling shutter effects -- the company may be able to minimize the effect somewhat, but it's still going to happen under certain circumstances.

Barry do you have any idea how RED is dealing with the rolllng shutter problems? Jannard's answer was that "we use a rapid reset sensor system which is much faster than a typical CMOS rolling shutter. It is a bit slower than film so it is easier (not much) to induce skew with RED than with film. But it is MUCH better than a standard rolling shutter."

I am still learning about the rolling shutter issues but reading this link (http://www.reduser.net/forum/showpost.php?p=35514&postcount=159) from your article (http://dvxuser.com/jason/CMOS-CCD/) I understood that one way (with trade-offs) to reduce the rolling shutter artifacts on a CMOS sensor was to use more transistors per pixel to emulate a CCD's typical global shutter so that "pixels can be integrated and read out in the same time". Seeing tests (http://www.reduser.net/forum/showthread.php?t=7500) carried out by RED owners at this point seems that the shutter is still rolling, as in presumably that is not the direction RED is following.

Is it a mysterium or does anyone have any more technical details/links to share on RED's partial solution? sorry to post this on sony's thread

[Barry_Green]

I have no info on what Red is doing about it. I do know that they're very much aware and if something can be done about it, I'd wager they'd be the guys to do it.

That said, I don't know whether they can overcome it or not; I look forward to seeing the results of Jim Arthurs' testing.

A global shutter is a possibility, but not without drawbacks; more transistors on the chip reduce the "fill factor" of the CMOS sensor, which results in losing a stop or two of sensitivity. I'd gladly give that up, but many people wouldn't, so global shutters remain rare.

[stephen_nugent]

thanks barry, some say a good ol' mechanical shutter could be another option (w/ optical VF) like on the D20. RED opted otherwise from the start partly i guess to keep the system affordable. The case is in good hands anyway

[mico]

Red can't do anything about this flash effect. If you do a search you'll find that professional people are doing post work with Knoll light factory to fix this issue. The same can be done for the EX.

[lawriejaffa]

Oh god thats rather an incovenience I'd do without!

[mico]

From what I've heard its not difficult. There are presets flashes you make adjustments too. You don't have to create flashes from scratch. You do this right in FCP or premiere. I mean you don't have to create a flash effect for 20 minutes, just a couple of frames. Heres a demo clip of Knoll.

http://www.redgiantsoftware2.com/Downloads/knollfinal.swf

If that isn't for you maybe cameras with rolling shutter are not for you. Or don't shoot flashes.

[Stevet]

Some of Jim's RED ONE Camera rolling shutter results were finished yesterday. To my surprise, it looks like it may be about the same as the EX1.


Skew:
http://ftp.datausa.com/imageshoppe/outgoing/SHUTTER_STUDIES/RED_rolling_shutter_test.mov



Strobes:
http://ftp.datausa.com/imageshoppe/outgoing/SHUTTER_STUDIES/STROBE&ROLLING_SHUTTER.mov


Thread:
http://www.reduser.net/forum/showpost.php?p=134798&postcount=1

Jim at RED said there will be about a 10% improvement in Build 13 or 14 with a settings change.

[Steve Shovlar]

No difference to the EX1 but costing 4 times more.

[Barry_Green]

Can't download the clips from where I'm at, I'll try later today. Most interested to see the results. Did he try motion tracking?

[Stevet]

I believe Jim did not, but somebody else chimed in on that thread and said it worked fine for motion tracking. Well, I guess we really would need to see this.

[Jim Arthurs]

Hi all...

I haven't seen any sort of semi serious testing of the EX1... any links? A friend has access side by side to one with their RED today, but won't have the chance to do testing on this.

I really HATE this issue in general. I've done good tracks with RED footage that people have uploaded, but none of these shots have really stress tested the concept. Here's one example I did for fun with RED footage...

http://ftp.datausa.com/imageshoppe/outgoing/RED/SHAWN_CANOPY_tracked.mov

And another...

http://ftp.datausa.com/imageshoppe/outgoing/RED/GARAGE_tracked.mov

There are certain situations where you flat out CAN'T do what you can do with a global shutter... here's an example I've done with HVX....

http://ftp.datausa.com/imageshoppe/outgoing/PAN_TILT/TAKES_BLENDED.mov

and this...

http://ftp.datausa.com/imageshoppe/outgoing/PAN_TILT/NEAR_MISS.mov


...multiple takes shot without motion control and all motion tracked and blended together... a few degrees of skew will totally disrupt the end results of the tracking. Thing is... the tracking software doesn't know there's such a thing as skew, and will happily report a successful track. The problem comes when you want to do hyper-precise matching of say, four corners of a wall to a virtual set and keep them intact.

I love the RED. But you can't seperate the EX1 and the RED on these issues... they're the same. RED gives you 35mm DOF and deep color grading and the like, but CMOS they both are.

Irony is my first post the DVXUSER in a long while is the day after I sold my HVX... now I gotta buy something new...

Regards,

[ESTEBEVERDE]

O.K.

Here's my question.

Why wouldn't this same effect be seen in fast moving objects.

I know motion blur will mask most of the effect but would it not be the case with objects such as runners, bikes, and cars moving quickly horizontally?

[3xm]

I maybe mistaken but the problem is not with fast moving objects but rather with a fast moving camera....such as a pan...that's when the wobble occurs....

[Stevet]

Jim you are a great asset! Thanks for your execellent work.

If you find yourself in the Phoenix AZ area, I'm only about 50 minutes out of Phoenix (Casa Grande). Look me up and we can perform the EX1 rolling shutter tests. Man, just from the examples you've posted, I'm willing to bet the EX1 performs just about the same.

Depending on when Joel (Joelnet at reduser) gets his camera (lives in the area and I've worked with him before, due to get his camera any day now)), we might be able to arrange a direct rolling shutter comparison.

[Lenilenapi]

I haven't gotten my EX-1 yet so haven't been able to test it, but does anyone know if the half frame type exposures from at least single burst flash guns are always in the same place in the frame. That would make it much easier to fix in post.

- Lenny

[Stevet]

I doubt they are Lenny.

[ESTEBEVERDE]

I maybe mistaken but the problem is not with fast moving objects but rather with a fast moving camera....such as a pan...that's when the wobble occurs....


right... with the exception of the fast moving object itself being subject to wobble/skew.

The reason there is wobble/skew is because of a rolling shutter.


If you pan past a pencil fast it will skew right?

My hypothesis would be that if you move the pencil instead of the camera at the same relative rate through the frame it will have the exact same amount of skew as if you panned fast.

The whole frame wouldn't wobble but the pencil would skew due to the partial exposures inherent with a rolling shutter.

[Stevet]

My hypothesis would be that if you move the pencil instead of the camera at the same relative rate through the frame it will have the exact same amount of skew as if you panned fast.

The whole frame wouldn't wobble but the pencil would skew due to the partial exposures inherent with a rolling shutter.

Of course, for sure.

[matthew77]

It has nothing to do with camera movement, per se. It is only as to do with the movement of the subject relative to the camera. I could take a photograph (or a pencil) and move it in front of a static camera and the distortions would be the same.

[Stevet]

Yes, that's just what ESTEBEVERDE was saying.
It' s movement across the image plane regardless what's moving, the senor relative to the image, or the image relative to the sensor.

[matthew77]

Yes, Steve. I think we posted our responses at the same time.

[ESTEBEVERDE]

I still like this cam!

I mean shoots.

If red is good enough for the big boys, shouldn't the little ol EX1 be good to me?


My last hold up is the loss of res when panning.


Guys if you have the cam... how about working it out a bit and posting some footage of pans and what nots?

If it's not to much trouble. :beer:

[Stevet]

The only way would be for someone to post the same pan using a different camera. Ideally the other camera would have a global shutter like the HVX200.

Even then, since some may feel it's the codec creating a loss of detail. I certainly don't believe the EX1 35mbps HQ codec would fail. The softness that was refered to by Barry was real time off the head.

I still go back to - If it's a fault the image processing not having the response speed, then why does it improve with shutter speed.

[matthew77]

I am becoming more convinced that there is absolutely nothing to the "resolution loss while panning" silliness. I'm preparing a test that I will try to post tomorrow - It just needs daylight to complete so I'll get to it in the morning.

[Barry_Green]

Why wouldn't this same effect be seen in fast moving objects.
It would.

[matthew77]

One interesting thing to observe on the EX1 (and this may be true of other cameras - I don't know) is that peaking finds vertical lines of contrast. If you pan, the peaking goes away because motion blur smears the vertical lines and reduces their contrast.

If you *tilt* on the EX1, you will see that peaking does not change - no "resolution loss" occurs.

Also, if you tighten the shutter to 1/2000, peaking is unaffected by normal panning, ergo, no "resolution loss" is likely to be occurring.

[Barry_Green]

Interesting. Now that you mention it, I do recall noticing that the peaking only disappeared on the horizontal. We noticed the res difference on the horizontal, not on the vertical. Motion blur should be equally applicable to both the horiz and the vert.

[matthew77]

I would guess that motion blur is the same for both horizontal and vertical (except for minor differences due to vertical shutter scan) but that it is the peaking that is inconsistent. Peaking seems to find contrast in the vertical features and not horizontal features.

If you place a resolution chart in front of a static EX1 you will see peaking all over the vertical stripes but no peaking whatsoever on the horizontals.

[ESTEBEVERDE]

O.K.

I fully admit I am the slow one on the short bus.

If it's not to much trouble, could a few of you describe a bit more about the peaking. I don't quite get it.

Do you mean peaking as far as hot exposures in the frame or???

[matthew77]

Peaking is a focusing aid built into the EX1 (and similar to the focus assist function in the JVC cameras) that finds contrasty edges in the frame and applies a color to them. It allows you to see things "snap" into focus more easily.

On the EX1 it finds vertical contrasty edges (abrupt changes in luminance along the horzintal raster lines) and does not seem to find horizontal edges.

[ESTEBEVERDE]

Cool... so it's just a focusing aid as opposed to zebras that are for the exposure hot spots?

[matthew77]

Yes, correct. Like zebras it shows up on the viewfinder but not on the recording.

[ESTEBEVERDE]

Right... of course... it's just metadata that the camera displays.

[Barry_Green]

I would guess that motion blur is the same for both horizontal and vertical (except for minor differences due to vertical shutter scan) but that it is the peaking that is inconsistent. Peaking seems to find contrast in the vertical features and not horizontal features.
It should find it on vertical as well. Even so, the res loss that we were noticing was on the waveform monitor, not the cam's LCD. I brought up the peaking only to show people that it's obvious to see that something's happening, and that same effect doesn't happen on other cameras that I've tried (at the same pan rates, etc... obviously if you use an excessively short shutter speed or move excessively quickly you can force it).

[matthew77]

But Barry, the EX1's peaking clearly only works on verticals - if you don't believe me try what I said - use a res chart and see if peaking appears on the groups of horizontal lines. It doesn't.

This is with a static camera.

[matthew77]

Here's a test I did to check for resolution loss when paning the camera. It is not a perfect test but in any case reveals no change at all in a panning or still camera.

The chart is attached to the matte box. DoF was not enough to get it all in perfect focus.

These 1/4 size are crops but have not been resampled (except for jpg). The majority of the frame (not shown) is a panning image.

If there is any way I can improve this test and reshoot it, I would be happy to do so. Suggestions welcome.

[stephen_nugent]

what's this test for matthew? I'm not sure i understand.

[matthew77]

It's a test for possible resolution loss in the EX1 when panning.

[ESTEBEVERDE]

It's one good component of a the test.

I would also pan by a series of verticles like a picket fence to check as well.

That way we can see the hard edges of the fence as well s the background in between the fence.

[matthew77]

ESTEBEVERDE,

When you nodally pan a camera, everything in the frame will motion blur equally, both the "pickets" and the stuff between them. Perhaps you mean dollying past a fence? This would apply more motion blur to the foreground objects than the background material. But then I'm not sure what that would prove vis a vis the assertion that there is resolution loss while panning.

[ESTEBEVERDE]

No... just the fence.

The foreground will be moving across the frame at a slower relative speed and because it is a picket fence it would be easier to see how much blur we are getting.

If this is all blur the picket fence should blur less than the back ground.

On a dolly move the relative speed would be the same as it is linear and not an arch...

[matthew77]

ESTEBEVERDE,

Motion blur is a 2D phenomenon. As measured in pixels, a nodal pan will create equal length streaks (at a given position in frame because of lens distortion) on everything in frame. With a lateral dolly move (no camera roation), foregound objects will streak a certain amount while obects approaching infinity will appear to stand still - hence no motion blur.

[ESTEBEVERDE]

Hmmm....

O.K.

Let's look at it like this.


Do a "reverse pan" where the subject moves and not the camera.

Put the camera in the center of a gigantic cd.

Now, step three feet away.

Start at the right side of the frame and run just to the left.

Measure your distance and speed.

Now... step 200 feet away and do the same thing.

Start at the right side of the frame and run just to the left.

Measure your distance and speed.

You have to move much faster and farther the further you are away from the camera.


And.. on second thought... it's actually the same with the dolly move.

Foreground objects have slower relative speed but cause more blur....

[ESTEBEVERDE]

ESTEBEVERDE,

Motion blur is a 2D phenomenon. As measured in pixels, a nodal pan will create equal length streaks (at a given position in frame because of lens distortion) on everything in frame. With a lateral dolly move (no camera roation), foregound objects will streak a certain amount while obects approaching infinity will appear to stand still - hence no motion blur.



Another way to look at it is will an airplane traveling at 500 miles per hour 2 miles away cause more motion blur or one 2 feet away?

[ffaf07]

What about slow-moving shots?

Think of movies such as Elephant, or Magnolia.

Some of those shots are at the pace of someone normally walking, for a rather soothing effect. Some of the shots are level, then "dip" in to dutch angles, then come back.

Would a rolling shutter have a serious effect on these types of shots?

(Sorry to temproarily hijack the thread, BTW)

[ESTEBEVERDE]

Those are great questions! not a hijack at all!

[Barry_Green]

The rolling shutter becomes most noticeable if you reverse direction. Gliding smoothly will not reveal much in the way of artifacts, if anything. A quick reversal will make the artifacts very obvious. A handheld shot on an HV20, for example, is much jigglier than the same handheld shot on a DVX. As the handheld motion results in frequent reversals (camera is not gliding, it's more "bouncing") then the effects of the rolling shutter become more obvious.

If your camera movements are highly controlled and smooth, the rolling shutter artifacts will be masked. If the camera is moving in a more uncontrolled way, "Cops"-style, the rolling shutter will be much more problematic. The ultimate nightmare scenario for a rolling shutter is a high-vibration scene.

[Jim Arthurs]

I've added tilt test results showing expansion and compression of the image to that first post in this thread...

http://www.reduser.net/forum/showpost.php?p=134798&postcount=1

It would be a real asset if anyone with an EX1 could do similiar sort of work...

For instance, it's deductable from my first pan test that an image subject in motion moves from one screen edge to the other in approx. 18 frames will skew 3% of the 4K width. Charting out an EX1 would give us more information about skew in general and help compare the RED to EX1 on equal terms.

Regards,

Jim Arthurs

[Stevet]

Wow, that's an eye opener more than any other artifact. So have you determined the minimum speed an object has to pass between to points to cause noticeable artifacts, tilt, or pan?

Also, Jim, can you send me details on how I can set up this experiment with the EX1, or if you're ever down in Phoenix AZ area?

[Jim Arthurs]

Hi Steve... all you need is a sheet of black foam core board from a Michaels or Hobby Lobby... then some white chart pack tape 1/16" inch or rip off a thin strip of white grip tape.

I wouldn't use the dots I used on the RED test, upon reflection, I'd just do a series of long thin strips an inch apart running up and down the whole length of the board. Then shoot it both horizontal for pans and vertical for tilts. Do one pan very slow with the tilt locked off to be a base reference. Then do another where you whip through frame each direction. I'd let each pan motion go completely off the target in each direction and try and keep the speed constant through the pan. Do different speeds.

Then repeat the same for tilts; a base tilt recorded first, very slow, and then your faster ones.

If you wish to shoot something like this, I'd be glad to do the post on it for you so we can share the results... let me know PM and I'll set you up some FTP space to upload the files.

Oh, and you can kind of figure out how much skew at different speeds... if 3% width with a pan at 18 frames from left to right, then 1% skew with a pan that's 54 frames... so a 3 second or longer pan time for an object crossing the frame (at least for the RED) shouldn't have very much skew at all.

[Stevet]

Thanks Jim.
I should be able to get to this on Friday.

[Stevet]

So Jim, that example was at 4K, right?
I take the results be half as bad at 2k?

[Jim Arthurs]

Yes, full recorded 4096 wide. 2K would be half as bad, maybe less if I understand that the reset is not directly proportional... don't quote me on that.

BTW, here's an EX1 test that has just come to light by Eric Pascarelli.

http://dvinfo.net/conf/showthread.php?t=112362

While there's some slop in my test, his is very cut and dried with the outcome when you measure the frames in Photoshop... 2.2% width of frame skew in 17.6 frames for a subject to cross the frame.

My RED test was approximately 3% skew in 18 frames. My pan speed isn't perfectly constant like Erics. It's clear even with some error that these two cameras are very very close. It's possible that the EX1 has a little less skew than the current firmware build on RED.

[Stevet]

Thanks Jim, When I first saw your thread and results from your tests, that was my immediate thought. It sure looks just like the EX1.
My thought is there really is not much they can do about this issue via software.


If there was a way, they surely would of implemented it from the start.

I did hear about RED's upcoming 10% improvement.

So, based on those results from Eric, It appears when shooting at 24 FPS, a 3/4 second constant pan speed measure from edge-to-edge will yield 2.2% frame skew.
So, I imagine if you kept this speed to at least 2 seconds or more, there may be little to no visable skew. I'd like to see this also done for the tilt axis.

[Sumfun]

Thanks for the reference to Eric's test.

Aside from the skew, the pictures at high shutter speeds look pretty sharp. Barry Green mentioned that the picture lost sharpness during pans, but maybe it was just due to motion blur?

[Barry_Green]

I'd have to get more time on it to find out. There are several posts raising questions about what happens in motion; Noel Evans just posted something. I am not sure what was going on, but something seemed "off". There was definitely some motion-induced res loss happening, and it *seemed* to be moreso than would be attributed to motion blur. But I'd have to get some more time on an EX1 to get to the bottom of it.

[Jared Meyer]

Barry do you have any idea how RED is dealing with the rolllng shutter problems? Jannard's answer was that "we use a rapid reset sensor system which is much faster than a typical CMOS rolling shutter. It is a bit slower than film so it is easier (not much) to induce skew with RED than with film. But it is MUCH better than a standard rolling shutter."


I was cruising around on reduser and saw that Macgregor just posted some examples of really horrible skew from the Red. I haven't seen any of the rolling shutter issues addressed much over there but this thread might be an interesting one to follow:

http://www.reduser.net/forum/showthread.php?t=8612

[Barry_Green]

Rolling shutter is rolling shutter, I've never encountered a rolling-shutter camera that didn't have the exact same issues. Maybe a slightly lesser degree here or there, but -- if you buy into the rolling shutter you *have* to expect this type of stuff to happen.

[Jared Meyer]

Right. I'm actually surprised I've seen so little mention of it over there...I mean, it's not fair to say that skew or other rolling shutter artifacts are the Red's achilles heel because it's not the only pro camera out there that uses one. But it sure continues to seem like a significant reason not to invest in Red, or any other rolling shutter design.

I think Jim might have made it sound as if they had lessened the problem to a great degree but honestly the amount of skew exhibited in Macgregor's test footage is enough to turn me off of the camera. Handheld, fast motion, vibration and whip pans are part of cinema today. No way around it.

EDIT

It's interesting to me that several of the commentors in that thread seem to have been completely unaware of this phenomenon. The misinformation is kind of startling. One person asked if maybe this was just limited to a certain number of faulty cameras...

[Stevet]

But it seems like it would be a huge reason not to invest in Red, or any other rolling shutter design.


Apparently not huge enough ;)
All of these sub 10K cameras have their good and bad. If this is not good enough for you, wait it out for something new. Maybe Panny's got something under their sleeves.

[Jared Meyer]

Apparently not huge enough ;)
All of these sub 10K cameras have their good and bad. If this is not good enough for you, wait it out for something new. Maybe Panny's got something under their sleeves.

I completely agree. Hope I don't sound like a whiner. Just thinking it all over. :)

[Stevet]

No... It's legit man.
Rolling shutter artifacts are not pretty.
I stay away from whip pans on my EX1.

[aschocken]

show that footage to someone who is not a cinematographer, and ask if they notice anything strange. they'll mention that the camera is flying all over the place, and a dude is jumping around. but they will not notice the skew. i think most of the situations where skew comes into play, it's only something noticed by professionals.

[mico]

Well that was shot at 1/2000 shutter so no motion blur which would make it less noticeable. The other thing is it has been talked about many times over there ever since the Peter Jackson short showed a bent wheel on a whip pan over a year ago. The people there know it, discussed it, and accepted it as part of the deal. Jannard himself said he could still lessen it by about 10% .
And personally after all the complaints sent to Roger Ebert and other sites about the camerawork making them nauseous in Cloverfield and Bourne, that type of shooting may be on its way out.

[Noel Evans]

i think most of the situations where skew comes into play, it's only something noticed by professionals.

Most, yeah I agree. Nut if this small scene happened to appear on a big screen, many people would pick up on it a little, it looks unatural. But I dont think they would care or give it more than a passing thought.

One thing though, I really dont think the EX is any worse than this.

And Mico, Bourne is horrible agreed. But these shots werent that bad I didnt think. Pretty typical of action components in a lot of TV, where I think a lof Reds market will be.

[Blaine]

And personally after all the complaints sent to Roger Ebert and other sites about the camerawork making them nauseous in Cloverfield and Bourne, that type of shooting may be on its way out.One can only hope...:thumbsup:

[mico]

And Mico, Bourne is horrible agreed. But these shots werent that bad I didnt think. Pretty typical of action components in a lot of TV, where I think a lof Reds market will be.

So lets see it at 1/48th. The LART red test movies are available for all to see on Vuze. With lots of TV like action shots and I couldn't see any noticeable skew. TV and film action is not shot at 1/2000.

[BrenMyster]

Just a quick question, The issues with the rolling shutter and the flash,

I know nothing can completely solve this problem but it looks like it may be possible to make it less apparent with some post correction.

Maybe someone will come out with a FCP plugin or filter that can counter act it.

[Barry_Green]

Depends on the situation. If you're talking about one flash here or there, people have suggested overlaying a white transparent image over the whole frame to minimize the effect. But if you're talking about constant strobes like the cop car footage, ain't nothin' gonna fix that.

[BrenMyster]

Yeah, I mean, I'm buying the Ex and want to shoot music videos, live shows, short films etc,
Obviously any place were there are crowds your going to get cameras.

I'm not going to be shooting Police Cars and Fashion shows, so I think it will work out ok. Sooner or later there will be a good plugin that will counter act the effect to some degree.

[Noel Evans]

Sooner or later there will be a good plugin that will counter act the effect to some degree.

With Graham Natress on the team, theres a pretty good chance. BTW his FCP plugins are great.

[marco0782]

i think most of the situations where skew comes into play, it's only something noticed by professionals.

No offense but in that case why bother with a high end camera.

[aschocken]

because there are a lot of things other than skew that actually are noticed by your viewers. just because they don't notice rolling shutter problems doesn't mean they can't recognize nice cinematography when they see it.

[androbot2084]

Lately I am watching more papparazi footage with these cameramen sporting their new Sony XDCAM EX cameras . And the footage still looks terrible like a mechanics strobe light for adjusting your ignition timing. I guess these television shows are more impressed with the camera equipment then they are with the actual footage. It seems that big resolution numbers and big half inch chips sell the camera even if the image quality goes to hell for the intended usage.

[matthew77]

androbot,

How do you know that the paparazzi are using EX1's? Where do you get your info? Who are "these television shows?" What are you talking about?

I have certainly seen some cameraphone footage of Brittany Spears that has horrible CMOS artifacting, but that's a combination of both partial exposure and the cameraphone's inablility to deal with bright flashes in general - overexposure.

[RE1000]

Just found this one. It's pretty bad and I could consider this 100% unusable.
If I hired someone and they brought me back this footage, they would be beyond fired.

http://blip.tv/file/703191?filename=Jetset-specialWhatFilmmakersReallyThinkOfTheWeb273.flv

EDIT: i'm not sure what camera this is from, i did a search and this thread came up, i didn't even know this was in the Sony EX-1 section

[Joe Lawry]

wow... thats terrible..

[Stevet]

Good one...No where does it say the EX1 shot this..
But having said that, the EX1 partital flash exposures would of been close to that.

[StefW]

Just found this one. It's pretty bad and I could consider this 100% unusable.
If I hired someone and they brought me back this footage, they would be beyond fired.

http://blip.tv/file/703191?filename=Jetset-specialWhatFilmmakersReallyThinkOfTheWeb273.flv

EDIT: i'm not sure what camera this is from, i did a search and this thread came up, i didn't even know this was in the Sony EX-1 section


I'd say looking at the production, that one's probably not shot on a cameraphone...

Any video producer / camera person who wasn't embarrassed to present such a mess as acceptable output should consider a different vocation.

As an equipment vendor to the video industry in Australia and NZ since 1994, I would have to say that this Rolling Shutter issue rates as the most significant flaw I have seen in any professional video product, let alone camera, during my time in this industry (25+ years). Feel free to suggest other contenders.


So significant an issue that in conscience I will not sell a Rolling Shutter camera to my clients. Even if they were made fully aware of the technical problems and they still wanted it, I would suggest they bought it somewhere else. Why? Because when the Rolling Shutter issue is (inevtiably) resolved, ALL Rolling Shutter cameras will be worthless.

But there's an even more important reason - at least here in Australia:

Despite any effort by a vendor to alert the purchaser to the limitations of a product (not sure how a vendor would be expected to know about this issue if it wasnt for online communities...?), Australian consumer laws will expose the vendor's business to refund andpotential compensation claims by purchasers, who will eventually "notice" the weird flashes, banding, skew and wobble, and decide that the product is not fit for the purpose for which it was intended. If the camera owner doesn't come to that conclusion him/herself, he may do so shortly after his client rejects a job.

At the end of the day, Rolling Shutter CMOS technology (or at least its ill-conceived deployment in video and digital motion picture cameras) looks set to take a podium position among the greatest technology stuff ups of the 21st century... it's certainly winning the race for this decade. Hopefully it wont be around to be a contender in the next.

[matthew77]

At the end of the day, Rolling Shutter CMOS technology (or at least its ill-conceived deployment in video and motion picture cameras) looks set to take a podium position among the greatest technology stuff ups of the 21st century... it's certainly winning the race for this decade. Hopefully it wont be around to be a contender in the next.

You're joking, right?

Since most 35mm film cameras exhibit this tendency as well (albeit to a lesser extent), don't you think you are being a bit grave and dramatic here?

Or are Arriflexes the great technological stuffups of the 20th century?

Yes, most 35mm cameras are subject to partial exposure of flashes if the strobes are not synced properly with the shutter.

The rolling shutter sucks, but it only rears its ugly head in a small percentage of circumstances. Every camera system has its achilles heel and I don't see it as being any worse than the flaws of other systems.

And not everyone shoots unscripted footage of paparazzi flashes. Many of us use our cameras in resonably controlled circumstances where they take great pictures. Much like the controlled circumstances that are expected when shooting 35mm.

[Justyn]

would be nice to get confirmation on what shot that awards thing... cause that one montage of fash cuts with flash is just beyond terrible.


I think that CMOS camera limitations would be a serious detriment to shooting. I'd say that maybe 5 percent or more of what I shoot has some level of flash photography and rapid camera movement and it just can't handle that stuff. So I'd need to cover those occurances with another camera and that just doesn't seem prudent.

[lawriejaffa]

The Red and 35mm have not and do no exhibit the problems of a prosumer rolling shutter cam to even a vagely similar degree. So lets not make pointless justifications for it.

A good point is made, that when this is resolved it will impact the value of cmos cams quite dramatically.

All this pussyfooting about controlled circumstances to use a cam, where a minor oversight could completly ruin a job (especially an event capture) is just an unnecessary risk for me. So i can see your point of view StefW

[Code Rad]

I remember seeing some subway shots with a red and when the train went by it was skewed so bad it was laughable. The Red is not immune.

[LuckyStudio 13]

There is a clip of a music video shot with the RED ONE with Strobe light. It exhibits the same rolling shutter aftermath as the video above. And yet RED ONE cameras are used to produce top tier hollywood feature films all over the world.

You need to know your tools and then select the best one for that particular job. Saadly a lot of so called DP dont really know much beyond composition and pressing the 'rec' button.

[Jim Arthurs]

Hi all... okay, here's the deal. The EX1 and the RED have exactly the same characteristics in regard to rolling shutter skew. No difference. This isn't speculation. This knowledge gained by hands on testing.

http://ftp.datausa.com/imageshoppe/outgoing/EX1/EX1_vrs_RED_rollingShutter.jpg

And here is the section of pan that was pulled from to give you an indication of pan speed (this image was pulled from the middle of this pan)...

http://ftp.datausa.com/imageshoppe/outgoing/EX1/PAN_SPEED_red.mov

Now, as to the flash and strobe issues? Again, the same. Here's a test I did showing RED and multiple strobes going off.

http://ftp.datausa.com/imageshoppe/outgoing/SHUTTER_STUDIES/STROBE&ROLLING_SHUTTER.mov

Visually identical to EX1 footage of strobes that we've all seen. The difference being that you can physically sync a strobe to the RED, but not the EX1.

Now, on film. Film skews, but differently. Only the attack and retreat portions of the shutter have the potential for skew. The center of the exposure time doesn't skew. Here's an animation I did some time ago to show this...

http://ftp.datausa.com/imageshoppe/outgoing/SHUTTER_STUDIES/FILM_SHUTTER_SKEW_CLASSIC.mov

And film has issues with strobes that aren't synced, as pointed out earlier. However, since the shutter is some physical distance from the film plane and out of focus, the effect isn't as nasty. Here's a couple example from the movie Pitch Black showing this...

http://ftp.datausa.com/imageshoppe/outgoing/SHUTTER_STUDIES/PitchBlack_strobes.jpg



Regards,

[Amulet Man]

Nice work. Don't forget, though, that the Red has a much larger sensor (more lines to read out), so it's actually must faster than the EX1 per line; it just has a lot more lines to read. Interesting that they'd be functionally equivalent, though.

[matthew77]

And film has issues with strobes that aren't synced, as pointed out earlier. However, since the shutter is some physical distance from the film plane and out of focus, the effect isn't as nasty. Here's a couple example from the movie Pitch Black showing this...


Jim,

Thanks for the visual aids - it's important for all to know that the rolling shutter has been with us in some form a hundred years and that it's not the worst technological mistake of the 21st century.

BTW - the vidicon tube does partial exposures, too, but it's hard to see because a flash completely obliterates the image in most cases. Perhaps all analog video cameras can share the podium with the Panaflex as the worst technology of the 20th century? But I digress.

It actually is possible to sync strobes to the EX1 - I've done it. As long as you are at 59.94p, 50p, 29.97p or 25p you can use the analog or SDI out into a phase adjustable sync box and it works great.

When the firmware update allows 24p out of the SDI, you'll be able to do it at that speed as well.

[Jim Arthurs]

Hi Matthew, that's good to know about the sync trick and the EX1. That opens up a few possiblilites and doors I thought were closed.

The irony is that the first job I was going to do with RED was a green screen shoot with a limo pulling up and dozens of press photogs snapping away... the entire brain trust at REDUSER couldn't come up with an effective workaround (on the budget we had, in any case). In the end, I went F900 and just shelved the problems.

Next up was a shoot involving a slow 25 second pan around a utility shed on green screen that I needed to motion track absolutely perfectly with a perfect geometry replacement. Here again, without extensive testing, I didn't feel comfortable with a rolling shutter.

In the long run, these were both issues that caused me to pass on the RED. I was much more comfortable spending 6500 US on the EX1 than 20,000 + on the same set of issues, knowing that there would still be the need for expensive rentals to work around the shutter situation in either case.

The key with the rolling shutter is to test and know your limitations. Hopefully, this will be a very short lived phase of imager development and wind up as nothing more than a foot note in a few years.

[lawriejaffa]

How often must we hear this ridiculous comparison between traditional film cam technology and the rolling shutter issue of cmos video cams?

Its a completely meaningless piece of rhetoric to justify a crude technological regression. That its bearable with the RED is only because of the vast host of other benefits it brings to the table, and because the issues of the rolling shutter on that model will eventually be addressed.

Should the EX do that then great, but in the meantime, theres no point denying this achilles heel.

I think too many cam makers seem to be chasing impressive sounding specs at the cost of basic functionality and trusted and proven technoloogy. :P

[chagchag]

Hey Matthew, could you please link to a sync box and/or the lind of strobe one have to use for this kind of shoot? A product-page or something, so I'll know what to look for.

[matthew77]

This is not what I used - but it's probably the most elegant solution. Strobes would need to be compatible with the accessory output:

http://www.cinematographyelectronics.com/index_gallery.html

Needless to say, this is expensive. But rentable at most 35mm camera houses.

When I did it, I used something far more cumbersome, but it definitely worked - it was a motion control computer (not connected to a motion control system - just standalone):

http://www.kupercontrols.com/

Also expensive.

[matthew77]

That its bearable with the RED is only because of the vast host of other benefits it brings to the table, and because the issues of the rolling shutter on that model will eventually be addressed.


But that's my point - the EX1 also brings a "host of other benefits to the table"

One of them is price. But there are others. The HV20 has amazing benefits, too. And its rolling shutter is worse than the RED's. The Arricam has benefits, but also has a rolling shutter to some extent (as demonstrated).

Basically, all cameras fit somewhere in the market space with their costs and benefits. If their wasn't some compelling reason to buy the EX1, the HV20 or an Arricam or a cameraphone, no one would.

For those of us who don't shoot weddings or paparrazzi flashes, rolling shutter is barely a consideration - yes, it sucks, but it's nowhere near a dealbeaker.

I only bother to say any of this because of the earlier alarmist commentary by StefW. Again, I agree it the rolling shutter is a complete suck-fest, but it is hardly the "greatest technology stuffup" of the 21st century.

[chagchag]

This is not what I used - but it's probably the most elegant solution. Strobes would need to be compatible with the accessory output:

http://www.cinematographyelectronics.com/index_gallery.html

Needless to say, this is expensive. But rentable at most 35mm camera houses.

When I did it, I used something far more cumbersome, but it definitely worked - it was a motion control computer (not connected to a motion control system - just standalone):

http://www.kupercontrols.com/

Also expensive.


Thanks!

[JVR]

I agree and it's too bad that people have such extreme views about cameras, as if Camera_A is perfect, but Camera_B is total junk. Anyone who questions this goes to digital hell...

I've seen outstanding work created with the HVX-200 and I'm starting to see outstanding work created with the EX-1. Both cameras have strong points and both cameras have weak points, STILL users are able to produce great results with either camera. It's all a matter of person needs and wishes that defines which camera it best for each person.

A camera is a tool and just because someone selects a different tool from others, that does NOT make them a better or worse person than someone else.


Bob Diaz

No but it can make you a better tradsman!