View Full Version : Canon HF100 - 60i/30p Questions!


Larry Silverberg
October 19th, 2008, 01:01 PM
I have been looking at every link I can but I am still not clear how the camera shoots 30p. (And how this will then work in my final cut pro)

Apparently, the camera shoots in 30p but it is still actually "wrapped" in 60i? Is this correct? I do not understand this. So, if I do shoot in 30p, what happens when I do the "log & Transfer" into Final Cut Pro and start to edit. Am I getting 30p?

Can anyone EXPLAIN this to me?

Thanks!

Larry

Marcelo Lima
October 19th, 2008, 05:16 PM
Yes.. This cameras records the 24f and 30f shots of 60i... When we Log and Transfer on FCP, we have to convert to Apple Pro-Res. (this setting is automatic.)

After converted, the file is shown on FCP like this:
1980x1080 29,97 fps, pixel aspect: square, Apple Prores 422 17,7Mbits/ sec...

Larry Silverberg
October 19th, 2008, 05:33 PM
thanks marcelo,

sorry but i still do not understand. are you saying that final cut pro brings in the 30p as 60i and transfers into 30p again?

can you explain?

larry

Marcelo Lima
October 20th, 2008, 07:02 AM
As far as i know, the file type (original of canon) will be always 29.97 intrelaced (60i)m the 20f or 24f is simulation only...

Mike Gunter
October 20th, 2008, 08:18 AM
Hi,

I don't have a Mac (nor FCP, of course), but you can remove pull down of the HF100's (and HF10's) Frame Rate (24F) using TMPGEnc - Products: TMPGEnc 4.0 XPress Product Information (http://tmpgenc.pegasys-inc.com/en/product/te4xp.html). My guess is that most folks who have Macs have the the PC software, and TMPGEnc Xpress is great conversion software. I realize that you are using Macs, but there is software that allow you to use PC programs and this is a good one to get.

I posted the settings for converting AVCHD to HDV in this thread http://www.dvinfo.net/conf/hdv2-canon-vixia-hv30-hv20-hv10/136175-tmpgenc-4-0-xpresss-24p-hv20-best-settings.html. I haven't seen any significant loss from AVCHD to HDV, certainly not enough to warrant not using the conversion. The 24P is silky smooth, far more film-like than most I see in typical applications that drop frames.

You'll end up with a actual 24P product, too, assuming that FCP can edit HDV in a 24P MPEG2 Transport Stream (my guess is that it can).

Steve Mullen
October 20th, 2008, 09:14 PM
... the 20f or 24f is simulation only...

You should do some reading about pulldown. The 24p is real. However like ALL film, 2-3 pulldown is added so it can become 60i video. After conversion you need to use Cinema Tools to remove the pulldown. So read your Cinema Tools manual.

Les Wilson
October 21st, 2008, 05:40 AM
30P is stored on the tape in 60i because that is the only way data can be stored on tape. When you understand what 30p and 60i are, you will understand that.

It matters nothing that 30p is stored in 60i format on tape. What matters is that the camera captures all the lines of resolution every 30th of a second in 30p. When you capture it into Final Cut or whatever, the image data is the same. Just setup your Final Cut for the right format and it will take care of it.

Marcelo Lima
October 21st, 2008, 05:47 AM
You should do some reading about pulldown. The 24p is real. However like ALL film, 2-3 pulldown is added so it can become 60i video. After conversion you need to use Cinema Tools to remove the pulldown. So read your Cinema Tools manual.

For sure! I know very little about pulldown and this stuffs...
One question: If my final products will be video the best way to use 24f is leave the canon video files (60i) at the same way that is recorded...and edit normally??? Thanks..

Chris Hurd
October 21st, 2008, 07:51 AM
This cameras records the 24f and 30f...

...the best way to use 24f is leave the canon video files...Actually 24P, not 24F. There is no visual difference between the two, but it's important to understand the terminology. Because of the way it works, 24F requires a three-chip camcorder. The Canon VIXIA product line, including the HF10, HF11 and HF100, are single-chip camcorders which record 24P.

Marcelo Lima
October 21st, 2008, 12:40 PM
Actually 24P, not 24F. There is no visual difference between the two, but it's important to understand the terminology. Because of the way it works, 24F requires a three-chip camcorder. The Canon VIXIA product line, including the HF10, HF11 and HF100, are single-chip camcorders which record 24P.

I have a CANON HG-21 and it shows on display 30F or 24 F,....PF30, PF24...

Lorenzo Asso
October 27th, 2008, 03:11 PM
Mike Gunter,
even if your eye doesn't note any significant quality difference, is never a good idea 'transcode' a lossy long gop format to another lossy long gop format (avchd to hdv).
There are others alternative better and faster (only spatial codec that is intra-codec or loseless codec..). You only will need more space on disk but today a large hdd is not a problem since it is cheap.

ciao!

Steve Mullen
October 27th, 2008, 07:07 PM
Actually 24P, not 24F. There is no visual difference between the two, but it's important to understand the terminology. Because of the way it works, 24F requires a three-chip camcorder. The Canon VIXIA product line, including the HF10, HF11 and HF100, are single-chip camcorders which record 24P.

I think the difference really relate to CCD (which for Canon has been used for 3 chip camcorders) and CMOS (which for Canon has been used for 1 chip camcorders). The CMOS chips are running at 50P (for 50i), 60P (for 60i), and 24P, 25P, and 30P.

However, it seems Canon is still using "F" for 24P, 25P, and 30P.

I suspect they do this because they have a history of two types of progressive RECORDING: as Progressive and as Interlace. By saying 24F they are not saying HOW the chip is readout, but HOW it is recorded. It is a tip that 2-3 pulldown must be removed to get to 24p.

PS: Of course, there is still the question on HOW the 3 CCD chips running in "interlace mode" obtains video that has equal vertical to 3 CCDs running in "progressive mode." This is where Canon first introduced "F" to stand for sensor technology that was not "progressive."

Recent tests of deinterlacing in HDTVs show that a REALLY good deinterlacer can obtain 800-lines (from 1080) even under motion. Which matches my memory of what Canon delivers in F-mode. This is a great deal more than the 540-lines from discarding alternate fields. (Chris, is this correct?) By the way some HDTVs are sooo bad they deliver only 300-lines of V rez. when there is motion. Amazingly bad electronics.

Steve Nunez
October 28th, 2008, 09:48 AM
But has Larry's question been answered? What sequence preset does he use (FCP) to edit the 30p footage and retain it's 30p (non-interlaced) look?

My question: if I use Toast 9 to transcode the mts files into ProRes, what settings should I chose for video shot in 30p? (I want no signs of interlacing)
Thanks guys.

Larry Silverberg
October 28th, 2008, 02:08 PM
Hi Steve,

Well, I talked with a guy at B&H photo where I just bought my HF100 (will arrive in two days) and all he said was that was whether you shoot in 30p or 60i, Final Cut Pro's "log and transfer" will recognize the footage and bring it into the timeline in the correct manner. I guess that's all i really need to know. i look forward to trying it out.

On another note, I looked at your airbrush website and man you are amazing! I saw a few of the videos and seeing you paint freehand with an airbrush was incredible, really!

Thanks,

Larry

Steve Nunez
October 28th, 2008, 05:35 PM
Thanks Larry, much appreciated.
I was interested in your thread as I too was curious as to the ProRes Transcode settings for 30p footage.....and wanted to use Toast 9 to convert directly, instead of FCP's L/T utility....but I'll have to do it via FCP til I get the correct settings for 30p with the HF10/100's AVCHD footage.

Mike Gunter
October 29th, 2008, 09:41 AM
Hi Lorenzo Asso,

Transcoding to HDV wasn't getting there for the "goodness" of the CODEC. Some folks didn't have access to AVCHD editing software, but did have access to HDV editing software. This was one way in which they could both remove pull down and encode the AVCHD to an "easy-to-edit" HDV content.

As for the conversion, while you're right in the lossy CODEC in the HDV long GOP, but I also think that's an old chestnut in that the conversion is going to lose sufficient quality to warrant not doing it as to avoid it rather than having a bad editing experience. If you can have a smoother editing experience with your NLE with HDV, I would recommend converting.

If AVCHD is smooth already, that's great.

Most systems will be AVCHD capable soon, of course. I'm looking forward to it.

My best,

Mike