I have a client that potentially wants to do a video testing a new product. It would involve submitting the product to explosive forces like improvised explosive devices (IED's) and gunfire.

How might I expect the EX series cameras to perform under these circumstances, taking into account the rolling shutter?

Thanks,

Forrest

Well the EX1 is tougher than a tape based camera but it's not bullet proof or hardened. Anything major would probably wreck it. Have you considered a more specialised industrial/medical camera system? There are also camera mounts and rigs specifically designed to survive bullets and bombs...

Noah

Noah,

I think you may have misunderstood my post, or I didn't make myself clear enough. The idea is to document how the client's product would withstand an explosion or being hit by gunfire.

The EX1 or EX3 would be a safe distance away, even behind plexiglass if need be. The test would be performed during the day. I'm just wondering how the camera's rolling shutter would perform in this scenario.

Forrest

Fast moving subjects don't lend themselves well to rolling shutters. You might want to go for a Progressive camera with a traditional shutter.

Something that shoots 720p60 would be great for that type of thing. It would give you great slo mo shots.

Daniel Weber

Mentioned this in another sub forum here but worth repeating given the topic.

We rented an EX1 to a client for documenting a car bomb and suicide bomber.
Footage was shot at 720p60. Looking at it frame by frame the car / dummy was there in one frame and gone the next. Almost no flash or flames were visible. Modern explosives are incredibly fast and need a huge number of fps to really catch what's going on, even 1,000fps might not be enough depending on what you're trying to show. If all you want to show is effectively before, the event and what remains then the EX1 would be plenty fast enough.

On the other aspect, our EX1 didn't suffer and it was fairly close to the action. If the operator is safe so should the camera.

I would recommend trying to rent a phantom camera in all honesty. It is going to be a difficult shoot no matter what.

Thanks, Bob.

The intent is to show the before and after as part of a product marketing video, basically to show the success of the product after being exposed to explosives. I've done these kinds of shoots several times with CCD based HDV cameras with fairly good success, but I don't have own those cameras any more, just CMOS cams.

My concern is the "half flash" effect from the rolling shutter that I've seen from video samples that contain flash photography. My thoughts were in fact to shoot it at 720 60p at fairly high fps.

Forrest

I would recommend trying to rent a phantom camera in all honesty. It is going to be a difficult shoot no matter what.

I agree. A Phantom V10 can shoot 2000fps at 720p or 1000fps at 1920x1080. I can give you a contact if you like.

I shot an action short film with the EX1 and it was a pain to see the muzzle fire of the blank guns. It took several takes. Sometimes there was no muzzle flash at all. I guess you need to be lucky that the shot/gun discharge happens at the exact right timing with the rolling shutter. But after several takes we did manage to record the muzzle flash in camera.
We were rolling 1080 at 24p. I haven't tried 60p, but I will next time around to see if it does any better.

Most definitely get a Phantom HD. You can certainly make an ok video about explosions with the EX1 but if you really want to show the true impact in a clear way, there's nothing better suited to 1000fps with a Phantom. Rentable easily in NYC and LA at Abel Cine Tech, and elsewhere.

Noah

Thanks, Bob.

The intent is to show the before and after as part of a product marketing video, basically to show the success of the product after being exposed to explosives. I've done these kinds of shoots several times with CCD based HDV cameras with fairly good success, but I don't have own those cameras any more, just CMOS cams.

My concern is the "half flash" effect from the rolling shutter that I've seen from video samples that contain flash photography. My thoughts were in fact to shoot it at 720 60p at fairly high fps.

Forrest


Looking through the explosion footage I really didn't notice anything wierd happening to the image at all. The very fast stuff is so fast that the motion blur from the shutter would mask it anyway. Also with modern explosives there's very little flash at all, checkout footage at Cordin to see what I mean, even at 1,000,000fps you don't see much if any flash.
The same goes for guns, most if not all of the powder burns before the bullet leaves the barrel.
What could happen though is the blast wave rocks the camera causing a few jello frames. Then again, the jello frames could add a certain impact that might look quite effective in this instance.

Looking through the explosion footage I really didn't notice anything wierd happening to the image at all. The very fast stuff is so fast that the motion blur from the shutter would mask it anyway. Also with modern explosives there's very little flash at all, checkout footage at Cordin to see what I mean, even at 1,000,000fps you don't see much if any flash.
The same goes for guns, most if not all of the powder burns before the bullet leaves the barrel.
What could happen though is the blast wave rocks the camera causing a few jello frames. Then again, the jello frames could add a certain impact that might look quite effective in this instance.

Thanks for the info, Bob.

Forrest

You need to have very specific information about the entire setup. The camera, EX1 or other, needs to have protection or distance from the blast. Zoom is good in this case. Long focal length, way far away. Inherent in the EX1's sensor is anti-blooming, that will help. I will suggest bright daylight, direct sunlight, and the fastest shutter possible to get a good exposure.

Exactly what is it that is your main interest in the scene. How fast is it traveling and how far away is it. How long will it take to pass your field of view. What is the resolution you need on the subject. All these issues need to be considered.

The EX1's 1/2000 is not the greatest for fast moving things. 1/10,000 is typical.





I have a client that potentially wants to do a video testing a new product. It would involve submitting the product to explosive forces like improvised explosive devices (IED's) and gunfire.

How might I expect the EX series cameras to perform under these circumstances, taking into account the rolling shutter?

Thanks,

Forrest

You need to have very specific information about the entire setup. The camera, EX1 or other, needs to have protection or distance from the blast. Zoom is good in this case. Long focal length, way far away. Inherent in the EX1's sensor is anti-blooming, that will help. I will suggest bright daylight, direct sunlight, and the fastest shutter possible to get a good exposure.

Exactly what is it that is your main interest in the scene. How fast is it traveling and how far away is it. How long will it take to pass your field of view. What is the resolution you need on the subject. All these issues need to be considered.

The EX1's 1/2000 is not the greatest for fast moving things. 1/10,000 is typical.

As I posted earlier, I've done several of these kinds of shoots in the past, most often with multiple cameras, so I'm very aware of protection, distance, shutter speeds, etc.

I'm really just interested in the performance of the rolling shutter on an EX1/EX3 in this particular scenario since I may be purchasing one in the near future.

Forrest

Doesn't the Long GOP file structure of our EX1's have as much to do with missing the "peak of action" issue as does the rolling shutter? I may be completely off-base on this, but I'd swear that I saw or read an explanation about Long GOP and how it has to "fill in" info between I-frames and if the, e.g. explosion, happens between I-frames, then the camera might not have the image? Somebody with a big brain....please set me straight on this. ;)

might one want to shoot with no shutter to catch fast flashes/explosions? i would imagine the chance of the camera completely missing a flash would be greater if there is a longer space of time between end of exposure of one frame and the beginning of the next. of course this also has a specific effect on the images, so its not appropriate if you really need to use a high shutter to avoid motion blur. but with the shutter open, the beginning of one frame and end of the next should be nearly the same time.

i really dont see how long gop would affect the cameras ability to render an explosion except that it would be tough on the encoder since the explosion would suddenly change the scene, making it harder to compress. it would just be more affected by this fast/noisy motion than an i-frame only codec.