P 2 Work flow and unloading clips

[Lawrence]

I just received my 64 GB P2 E series and want to begin recording.

1. The 200HVX manual states that it is important to format the P2 cards with a P2 device. Do I need to purchase yet another piece of equipment? I know I can simply format in the camera. Confusing.

2. Is there some directions on transferring to FC Studio 2? The manual provide directions on transferring to hard disc, it further states that the hard disc drive must be formatted, which will erase all data. Does this mean that if I save my footage to my MacPro, or any other hard drive it has to be completely empty and formatted?

3. Half of my footage for this project was shot on the Fire-Store 100, shot in 720P24. But I noted that the footage was actually delivered to FCP as 59.97 in the timeline. ( This is due to the "over 60" frame rate from NOT using P2.) If I NOW begin to use the P2 will I have problems?

Some posts have indicated that I should not shoot in 24PN but instead use the 24 to avoid problems with different frame rates. And Barry Green suggested that I remove the pulldown on the Fire-store footage to match the new P2 footage but Shane Ross from the FCP users group says this can't be done. Any suggestions?

SHANE ROSS WRITES:
YOu have film like images. You have the 24p look....just in a 60 frame container.

Sorry, there is no way to "remove pulldown" from 60 frame material. That process is designed for removing pulldown from 29.97 footage...it doesn't work with 59.94 footage.

That camera does 23.98...if you shoot 720p 24pN. That option is only available on P2 cards. It also shoots 1080i 24pA...which is 24p in a 29.97 stream with Advance Pulldown..meaning it can be removed when you log and capture. But getting 23.98 while recording to a Firestore...not plausible.

 

[David Jimerson]

1) The HVX is a P2 device.

2) Transfer to HDD first, keeping the entire folder structure intact, including the LastClip.txt file. Then log & transfer in FCP. Any working HDD will be fine; you don't have to format it specially.

3) What kind of a "problem" did you have in mind?

On the FS-100, you could have used "PN" mode, which would have stripped out the extra frames on the fly and recorded/given you 23.976 files, so yes, it's not only "plausible," it's one of the features of the unit.

With the files you have now, if you want 23.976, you'll have to use Cinema Tools on it, if CT can do it.

If you want 24p and you're shooting 720 to P2, you should always shoot in 24pN mode. There is no reason to record 24p in over-60 unless your camera doesn't give you the choice. The HVX200 does.

All 24p pulldown is over-60, by the way. There's no such thing as 24p recorded in a 29.97 stream, or "over-30." This is mathematically unfeasible.

If you're shooting 720p, then over-60 is 24p recorded in a 60p stream. So, there are 60 full frames per second. (Yes, yes; it's 59.97, but let's use the round numbers.) Obviously, 60 is more than 24. So, what the camera does with 24p is repeat frames in order to fill out the rest of the "space." It takes one frame and repeats it, recording the same frame twice. Then it takes the next frame and repeats it, and then adds another copy of the frame, giving you that frame three times. So, frame one is recorded two times; frame two is recorded 3 times. This pattern is repeated over and over, so you get a frame 2 times, then 3 times, then 2 times, then 3 times, then 2 times, then 3 times, and so on. This is a "2:3" or "3:2" pattern (same thing), which is why it's called "2:3 pulldown."

In 720, ths is done with whole frames, complete pictures. The upshot is that because the frames repeat, your eyes see the pictures for longer, and it still looks like 24 frames per second. So yes, you still have the "film-like" look.

In 1080i (or 480i), the same thing happens, only with fields. There's really no such thing as a whole "frame" in interlaced footage. There is no whole picture ever recorded. There are only half-pictures. The first picture is of every other line; the second picture is of every line which would go between those lines. These are called "fields," because they're each only half of a frame. But the thing is, those fields are not imaged at the same time, so any moving objects are in different places every time a field is recorded. This is why you'll see the "comb" artifact in an interlaced "frame" -- it's because it's not a "frame" at all; it's two half-pictures taken at different times squashed together. It's no different than if you took two snapshots, then cut them both into strips, then discarded the even-numbered strips from the first snapshot, discarded the odd-numbered strips from the second snapshot, and put the remaining strips together and called it a "photo." You can imagine the problem, right? That's what happens with interlaced footage.

So, interlaced footage is NOT 30 (29.97) frames per second. It's 60 (59.94) fields per second. Calling it 30 (29.97) frames per second is archaic and needlessly confusing. What matters is that you have 60 distinct moments in time captured, not 30.

THIS MATTERS, because you CAN shoot footage which DOES record 30 (29.97) actual pictures per second, and that's 30p. It's not the same thing, not at all. The motion -- which is what matters with a frame rate -- is entirely different. Native 60i looks "live." 30p doesn't.

This is why 1080i footage is called 60i. Some older systems refer to it as "30i," and they shouldn't, because it's inaccurate and confusing. But if you see "30i," know that it really should be "60i." It's the number of pictures per second which matter, even if they're half-pictures.

How this relates to 24p recording is that if you record 24p footage in a 60i stream, then you're also repeating 24p frames, but instead of copies being made of whole frames, you're actually splitting up the frames into fields and recording them that way. Instead of the first frame being copied and shown twice, it's split into two fields, one with the odd-numbered lines, the next with the even-numbered lines. This has the same effect as showing the same frame twice, because nothing moves between the fields. (If you put them back together, you have no comb artifact, because it's the same picture.)

And you can do it in the same pattern -- Frame 1, 2 fields; frame 2, 3 fields (making an extra copy of the first field), frame 3, 2 fields, frame 4, 3 fields . . . so, 2:3:2:3 again.

But recording fields instead of frames also gives you a flexibility you don't have if you record whole frames -- you can spit it up differently. Frame 1, 2 fields, frame 2, 3 fields, frame 3, 3 fields, frame 4, 2 fields, giving you a 2:3:3:2 pattern instead of 2:3. THIS IS ADVANCED (24pA) pulldown. I won't get too far into the weeds (too late) to explain why, but this pattern makes it easier to remove the pulldown pattern and restore the original 24p frames.

This is why, if you shoot 1080 or 480, you want 24pA. It's especially important when using FCP, because FCP can only remove ADVANCED pulldown automatically; it can't remove 2:3 pulldown automatically.

So, to sum up: If you're shooting 720 to P2, shoot 24pN; if you're shooting to the Firestore, use PN mode. Either way, the files recorded are 23.976 and there's no pulldown to remove; you're golden. If you're shooting 1080 or 480, shoot 24pA. This will record in a 60i (not 29.97) stream, but it will be easy to remove pulldown and get your 23.976 back.

If you want 24p, never shoot 30p. 30p can't be converted to 24p very well. (Unless you're using it for slow motion, but that's a different topic.)

This will take you few reads to grasp, I'm sure, but it's not THAT complicated once you see it.

 

[Lawrence]

David,

Thanks for that in-depth instruction on frame rates and pull down.

The problem really is that I have already shot 24P with the Firestore, not using PN mode. Therefore I now have a timeline with footage at 59.94 frames/sec.

When I first introduce a clip into the timeline and I get the pop up asking if I want to match the timeline setting to the clip I always click YES.

Therefore I have 3 different bases.

1. My Canon Vixia HDV footage, so it's a 29.97 timebase. ( interlaced)
2. Footage from my HVX200 captured from the Fire-store ( Non PN mode) so, it's 59.94 editing timebase.
3. Now my P2 card which I will begin shooting and will be 23.98

I am now mixing and matching all of the 3 different footage into the multiple timelines I am using for this project. Example: After I already set the editing timebase to 29.97 for the Canon 29.97 footage, as the pop up window suggests, I then add footage from the HVX200, "Over 60" to that same timeline.

This leads me to the QUESTION: If the pop up window you see when you first introduce a clip into the time line is correct: ( "For best performance your sequence and External Video should be set to the format of the clips you are editing.) Then I would suppose putting different clips from the 3 above mentioned scenarios onto one timeline would not give me "best performance"?

I was thinking of just finishing this project with the Firestore-100, Over 60, keeping everything consistent. But I LOVE the P2, and hate the noise the mic picks up from the Firestore.

 

[David Jimerson]

If it were me using this footage, and I couldn't use Cinema Tools to remove the 2:3 pulldown from the Firestore footage, I'd edit it all together as 720/60p. The HDV footage timebase is actually 59.94, as I explained in my post above; what will happen is that each interlaced field will be interpolated into its own frame. I assume, being interlaced HDV, this is 1080i footage, so the interpolation to 720 should be fairly good.

If you want to use P2, you can still shoot 720/24P, not 24pN. Then, it will be recorded as over-60 to the P2 card. You don't have to, but if you want to keep the footage timebases consistent, you can.

Your recording time will be much shorter compared to 24pN, though, so it's not your better choice.