what's the dif btween slowmo in cam & shooting 60

Doing slow motion in-camera does not result in any resolution loss. Do it in post limits you to certain speed (usually 50% slow down) and requires the footage to be deinterlaced, costing you resolution. Plus doing it in-camera saves you time.
 
jade8472 said:
Doing slow motion in-camera does not result in any resolution loss. Do it in post limits you to certain speed (usually 50% slow down) and requires the footage to be deinterlaced, costing you resolution. Plus doing it in-camera saves you time.
Deinterlacing is not an issue with the HVX or any other camera that records a progressive image.

The big advantage with shooting 60P and doing slow-mo in post is that it gives you a choice between a smooth image + sound, and slow-mo if and when you need it. Trying to do the slow-mo in the camera means having to switch framerates, which disrupts your recording and requires planning, making it less attractive for live events.
 
I think he was asking about shooting 60i for slow-mo, not 60p. I don't know of any way to take progressive footage and make good looking slow-mo out of it in post except when you've already shot under-cranked in the first place.
 
THoff said:
Deinterlacing is not an issue with the HVX or any other camera that records a progressive image.

The big advantage with shooting 60P and doing slow-mo in post is that it gives you a choice between a smooth image + sound, and slow-mo if and when you need it. Trying to do the slow-mo in the camera means having to switch framerates, which disrupts your recording and requires planning, making it less attractive for live events.

Exactly right.

However, if all you want is top quality slo mo (don't care about the sound or time to plan for it) shooting in slow mo will give you the best visual results.
 
No, he wasn't asking about 60i.

The only way to do slow-mo in the camera is in 720P, so that's progressive already. For the post method, he specifically asked about "60 p", so that's progressive as well.

In your previous response you wrote that in order to do slow-mo in post you had to deinterlace, and now you say you can't get good-looking slow-mo from progressive footage.

If you start out with a 60P progressive image, you'll certainly get better results than you would from 60i with or without deinterlacing.
 
Agreed, but what I mean is that unless you've already shot under cranked (36p, 48p etc.) then you can't get good slow mo out of progressive footage (i.e. 24p or 30p) in post because there's no way to generate extra frames from progressive.

The basis behind slow-mo in post is to some how extrapolate as many frames as is required to reach a certain speed. Good result is only possible with interlaced footage because it gives you the most motion information at the cost of resolution, so doing slow-mo in cam is always preferred.
 
If you shoot 60P and deliver in 30P or 24P, you can get 2.0x or 2.5x slow-mo without any interpolation. You would create real-time from the 60P using pulldown (throwing away frames).

The notion that you can only generate good looking slow-mo from interlaced source goes against what you wrote earlier, where you said that 60i source must be deinterlaced. When you deinterlace 60i, you either throw away vertical resolution as you stated (and you gain NOTHING in terms of image quality), or you interpolate and make up information that's not there. Either way, it's not a good basis for subsequent processing where you may have to make up entire frames.

If you can get 60P from the camera instead of 60i, then the latter becomes irrelevant. I realize that the HVX is a slightly different beast than virtually all other cameras, but that's what we are talking about.
 
Yeah I'm with THoff - Go with the 60P and slow it in post. The HVX does a nice job of slo mo with overcranking, but I like to have the flexibility in post. I did some side by sides with overcranking and then just shooting 60P and slowing it down 40% (i think) and the difference was practically nill. Meaning there might have been a difference but I couldnt really tell.
 
I think I'm closer to my answer. I did say 60 p because the camera can shoot it, though i think the traditional dvx technique was to shoot 60 i and slo-down in post. Jdajda and co. are saying that they really think the difference is not noticable between in cam or post. If you have done tests and say the difference is almost nill, then i thank you for the information
 
THoff said:
If you shoot 60P and deliver in 30P or 24P, you can get 2.0x or 2.5x slow-mo without any interpolation. You would create real-time from the 60P using pulldown (throwing away frames).

Sigh, you're just not understanding me. If you shot 24p or 30p, how are you going to get slow-mo in post? Interpolating 24p or 30p looks like crap since the extra information isn't there.

THoff said:
The notion that you can only generate good looking slow-mo from interlaced source goes against what you wrote earlier, where you said that 60i source must be deinterlaced. When you deinterlace 60i, you either throw away vertical resolution as you stated (and you gain NOTHING in terms of image quality), or you interpolate and make up information that's not there. Either way, it's not a good basis for subsequent processing where you may have to make up entire frames.

Have I ever said that you can ONLY get good slow-mo through interlaced source? What I said was if you're going to do slow-mo in post then you HAVE to start with an interlaced source unless you're already shot undercranked footage in-camera. It's simple math, each frame of 60i is actually two half vertical frames occupying a single second, giving you extra motion information that aren't available in progressive. 60i source HAS to be deinterlaced to give you 60p; it's the ONLY possible source footage, even if you used Twixtor.

THoff said:
If you can get 60P from the camera instead of 60i, then the latter becomes irrelevant.

I've never disputed that doing in in-camera is the best way, but you seem to be ignoring everything I said. I will restate it again: Since the HVX is capable of recording more than 24 progressive frames, it will give you slow-motion in camera without ANY loss of quality. BUT if you want to do slow motion in post then it is BEST to start with an interlaced source and deinterlaced with a neligible loss in quality.
 
jade8472 said:
Sigh, you're just not understanding me. If you shot 24p or 30p, how are you going to get slow-mo in post? Interpolating 24p or 30p looks like crap since the extra information isn't there.
Depends on the tools and workflow you use.

I use Edius 4 Broadcast on a PC. I can shoot 60P and edit in a 24P or 30P timeline, and Edius will throw away frames as needed in realtime as I edit. And I can still output to a 29.97 MPEG2 file when I'm ready to author a DVD without having to use a funky framerate converter. But I can also tell Edius that the 60P file should be treated as a 30P or 24P file, which instantly gives me slow-motion video. And not only that, but the audio gets slowed down as well -- I'm not saying the latter part is useful, but I get to keep the sound, and using the built-in pitch shift filter, I might be able to salvage the audio.

So using 60P source, I get realtime video at any rate I choose, even if I didn't have the foresight to shoot it that way. I can even use Edius' time mapping tool and change the speed of portions of the video on the fly, or speed the whole thing up or slow it down by letting Edius interpolate frames.

jade8472 said:
Have I ever said that you can ONLY get good slow-mo through interlaced source? What I said was if you're going to do slow-mo in post then you HAVE to start with an interlaced source unless you're already shot undercranked footage in-camera. It's simple math, each frame of 60i is actually two half vertical frames occupying a single second, giving you extra motion information that aren't available in progressive. 60i source HAS to be deinterlaced to give you 60p; it's the ONLY possible source footage, even if you used Twixtor.
No, and I didn't mean to imply that.
jade8472 said:
BUT if you want to do slow motion in post then it is BEST to start with an interlaced source and deinterlaced with a neligible loss in quality.
Sorry, this I can't agree with.
 
Hey Jade 8472, i'm interested in your' theory about interlaced. For one i should say that i only asked about 60 p footage getting slowed down. So when i said progressive i meant 60 frames of it, not 24 or 30 which you came up with out of nowhere. Now there seems to be a debate of whether 60 i or 60 p is better for slowing down footage in post. The original question had to do with quality of in cam slomo, and the possibility of recreating it under optimum conditions in post, (having shot the extra frames). If you shoot 60 p it doesn't mean you are over cranking. But who says 60 i is really the way to go when aiming for post slomo? And how much better is it to have in cam slomo, when i'm starting to think that's a gadget intended for consumer reasons. If the quality is really better then i can understand what the fuss is all about, but even the varicam slomo has to be veiwed on an nle, and not in cam.
 
Hey BenB, i;m not sure that is a very good example. Was it 60 frames you slowed to 24 or 24-30 frames you slowed down. Some people are saying taking 60 frames and slowing to a 24 timeline will give similar results to shooting 60 fps and playing it back in camera at 24pn.
 
First of all, I didn't do anything in any Native mode.

Yes, shooting 60fps and conforming it to 24fps timeline without dropping or duplicating or faking frames at all is all the same thing. No matter how you do it. By tricking your NLE or using the HVX modes to do it.

Problem is, on the DVX, you can't get 60p, but 60i, so there's some image quality issues there.

Slowing down in an NLE creates "faked" frames to fill in, and so is not accurate like in-camera is.

I shot just like the caption said, 720p24/60.

1- In FCP I imported once without pulldown removal, used the FRC, got my native, frame accurate slow motion clip.

2- Then I imported a second time with pulldown removal to give me my normal, frame accurate 24p normal motion clip, then slowed it down with the Speed setting in FCP to 40% in the FCP Sequence.

Simple as that. Taking 60fps and "conforming" it to 24fps takes the first 24 frames of your original 60, and reassigns it to be the first second of footage. Then takes the next 24 frames of your 60fps original, tags that as the second second of footage, and so on. This is frame accurate, no matter how you do it.

Again, in a DVX100, you don't have 60 progressive frames to start with, you have 60 fields, so there is a real image quality issue.

TECHNICAL SPECS; I shot this while out in the French Quarter on Mardi Gras day last month from before sunrise to after sunset to document the whole day, with a mid-day break to travel the city and get shots from the other parts of Mardi Gras. I used my HVX200 set to 720p24/60, cause I didn't know what would and wouldn't end up as slow motion. I also had an FS-100 in a Bebob Box FS mounted to the bottom of the camera. I used DVXuser.com batteries in the HVX. I had two extra in my pocket, as well the original battery the FS-100 ships with. I had the extended life battery from Focus on the FS itself. I only had to change out the battery on the HVX once, all day. And I shot all freakin' day!

Weight? Not a problem! No fatigue, no cramps, no strain, nothing, was really nice.

FW cable? On the HVX side I had to unplug and plug back in probably four times all day. But I never touched the FS the whole day. Turned it on, shot, left it on, unless I was grabbing a drink and flirting with a bar maid, then I shut the whole rig down. So the only problem with the FS was the "HVX's FW Port", not the FS-100. Who's fault is that? Panasonic, not Focus. Sorry, have to throw that in. Every time I mention FS-100 folks say "the FW is unreliable on it." Wrong, it's unreliable on the HVX, not on the FS.

It was fun, had a great time, and as you see, I spent a little time in the four blocks of Bourbon Street that we call the "Girls Gone Wild section of Mardi Gras." This is the only four blocks of New Orleans Mardi Gras anyone outside the city ever gets to see. I also have tons of footage from the rest of the Quarter and the city, where it's family oriented. To bad none of the national media cover that part. All they want to show is this one four block stretch that is not what the rest of the city is like. Go figure. I guess news shows have to sell soap, too...
 
Last edited:
Back
Top