How shallow do you need to get? The 7D is darn near as shallow as S35. The 5D is nearly as shallow as IMAX at similar apertures. They're trying to shoot at F5.6 to F8 to get similar depth to S35 at F4. Been there, done that. I don't have enough talented focus pullers around me to make that work. Maybe if I lived in LA or NYC.

720/60p + post looks WAY better than 1080/24p + post for slo mo. WAAAAAY better.

What he said, it's actually the one thing that pains me about the 5D2.

Slowing 30 p or 24p in post with interpolation may not be clean (artifact free) as slowing by reducing 720 60p frame playback. Has anyone actually compared the two with actual examples anywhere ? I assumed it may not be clean, and that is why I was tickled when I got the T2i.

Yea, I have. And I shoot 60p ANY TIME I want slo-mo.

Go watch this: YouTube - FemHit Big

I can try to run some side by side tests on the weekend if I get a chance but I tested both and it was no contest.

In regards to the mark iv, 5d and 7d (since we shoot with these), while shooting at anywhere from f1.2 through f2.8, the camera's sensor size have done little to affect our depth of field. Your lenses and distance from the subject will have more impact on DOF than sensor size. I have no reason to believe that the 60D will be any different. How shallow do you need to get? I'm getting incredibly shallow shots with the 7D

If you need more low-light capabilities, then that's where the 5D will shine. But with the cost difference between the 2 cameras, you can buy a fast prime + 60D for the price of one 5D.

I agree with you Perrone out of experiences I had with try to slow the 5D stuff. But when someone makes a blanket statement like that, I was just wondering where it came from...

Chris, since I've never slowed down footage before, how slow can you actually get with interpolation? I'm guessing you have more options when working with 60p vs 24/30.

WIthout interpolation, and I work in Vegas, I merely have to go into the clip properties, and set playback at 24p. That gives you about 40% of full speed, and still having all real frames.

When you have to use the 5D for slo mo (slower than 30p to 24p conforming), I've gotten the best results by first shooting 30p and then conforming to 24p (1-1 frames), and then from there slowing down exactly 50% so either you're perfectly doubling up frames at that point, or doing some sort of simple blending or other more sophisticated algorithm, but you're doing it in a nice even multiple. But every 720p60 sample I've ever seen has blown away this type of work-around, even up-rezzed back to 1080p.

Slow Motion
 

I also find reducing the shutter speed works well too. I am not sure of the maths for the most appropriate shutter speed - others will know it better than me - but recently I shot some athletics at 1/1250th and the slow motion looked ok to me.

For my amateur untrained eye, the video footage plays back ok at real time. It is a bit strobe like, but probably ok for the subject of athletics or sports. If you stop any frame you will see the athletes' arms, feet and legs reasonably clearly. If I shot that the standard 1/48th - you get blurred arms and legs and I think this is what kills the post production slow motion effects as you end up with slow blurry video.

See the clip below - all shot at 1/1250th and the slow motion is post from Avid MC4 timewarp (first slow motion is at 5:25 sorry for the long wait - Vimeo needs the you tube Fast Forward feature.

Does anyone know the right shutter speeds to shoot when you plan to use slow motion in post - is it a simple multiplication factor? That is, if you plan to slow by 50% (multiply 2x) then shoot at double the normal rate of 1/48th going to 1/100th maybe? Would it go up from there for 75% (multiply 4x) shoot at 1/200th, for 80 or 90% maybe shoot at 1/200th?. I am not sure or maybe I am right off track

Are you kidding me?!! If that were the case I would have bought another 7D instead of forking out an additional £700 for a FF 5DM2.

Although I suppose it does depend if you are purely 'video recording' or filmmaking. For tele work, sure - the 7D is good, but the FF dynamic range of the 5D2 is just awesome and well worth the extra bucks.

Jeff, if you shoot at PAL's default shutter speed of 1/50th sec you capture pretty much everything that happens in front of your lens. If you shoot at 1/1250 th sec, you only capture 1/20th of the action - for all of the rest of the time the 'shutter is closed'. Or was that a typo - did you mean 1/250th sec? Even so, that means you capture less than a quarter of everything that happens, and if you're asking the computer to interpolate frames out of nothing, it's going to have a tough time of it.

You're right though - the faster the shutter speed the clearer each individual frame will look if you're in the habit of freeze-framing or want to print frames, but the human brain wants 'blurred arms and legs' (your words) to smooth out the slo-mo.

tom.

Wrong! the t2i has overheating issues that make it not even comparable to to the 5d Mk2!!

The human brain may want it - but the software you're using to interpolate intermediate frames will want the clearest possible image so that it can accurately identify and track the movement of individual pixels. It's much harder to do this when everything is blurred together.

Jeff's right about using the higher shutter speed for another reason - if you're trying to accurately simulate the look of something shot at a higher frame rate and slowed down you're going to need to use a shutter speed that would be appropriate for that frame rate. If you were shooting 100 frames per second it would be impossible to have a shutter speed slower than 1/100th/second, in a mechanical system such as film it'll have to be faster than that (since you need to move the film between frames) and would likely be 1/200th (with a standard 180 degree shutter). So I'd say a good rule of thumb would be to double the frame rate you are trying to simulate.

Show me GREAT (smooth) slo-mo from 1080/24p footage and maybe I'll agree with you - but so far I haven't seen any.