[Bill Marcellus]

Jan and all- thanks for clarifying the need to keep the MXF version of the media. So far our tests are going well. We start shooting Monday on a project that will result in ten hours of finished program material. I’ll try to keep this forum updated on our experiences with the HVX-200 and P2 workflow.

Thanks again for your help!

Bill

[Bill Marcellus]

Thanks guys- I was hoping that you would be the ones to answer my question. We just happened to have a BVM-1910 sitting around, etc.

Thanks again.

Bill

[Bill Marcellus]

LOL! Yep, you gotta’ stay away from that XBox if you want to get anything done!

[Bill Marcellus]

Walter- thanks again! I figured the card was a waste of money.

Bill

[Bill Marcellus]

Thanks, Walter. That is what I was thinking we’d do…i.e. PVM for CC and a big client monitor for the “flash” required by the agency/network folks. I was just concerned about critical focus issues- believe it or not, we get dailies with plenty of focus issues…LOL!

Thanks for your advice!

Bill

[Bill Marcellus]

Thanks guys! We’re getting it installed today.

[Bill Marcellus]

Malcolm,

Like Jan said, the Canon records in the HDV format which has 4:2:0 colorspace. Period. The 4:2:2 output is ONLY through the SDI connection and is not recorded to tape.

Additionally, HDV is a “prosumer” format, DVCPROHD is a professional format. If your intent is to create programming for broadcast you may find it difficult to sell HDV material to the networks. You will also have much better results in compositing and VFX with the Panny.

It seems you really like the Canon based on your experience with the XL. I understand. I purchased one of the first XL-1s in my area and loved it in 1997. That is, until the DVX-100 came out…that was the end of the XL-1 for me.

Again, I would emphasize that you should shoot some test footage with both cameras, play with them in post and determine which one you think provides the best image.

Bill

[Bill Marcellus]

Malcolm,

I don’t have any specific links but a search here and on Google should return plenty of articles.

The main difficulty with the HDV codec is that it uses Long GOP MPEG-2 encoding. Long GOP MPEG-2 achieves smaller file sizes by creating GOPs that include different
types of frames that rely on each other to complete the image. For example, a Sony
HDV 1080i camera uses GOPs that are 15 frames long. The first frame of a GOP contains the only I-frame (complete picture). The rest of the GOP is made up of bidirectional (B) and predictive (P) frames. To see a B or P frame, the system must construct the image by reading multiple frames, starting with the first frame of the GOP. This means there is not a discrete, individual frame recorded but rather a group of frames with the codec predicting changes in the image from frame to frame. This becomes problematic when there is motion in the frame, i.e. a person running, a car speeding by, etc. The codec must predict the motion and build an intermediate frame within the codec.

This can lead to “mosquito noise” within the image…much like the noise you see in a poorly encoded MPEG image with motion in the frame. It can also show up as noise in the blacks of an image. Additionally, it can lead to problems in editing and compositing because there are no individual frames to begin with. “Frame accurate” editing has to be artificially created within the editing software.

Additionally, DVCPRO HD is a 4:2:2 signal, which means it retains half the color samples that the CCDs capture. HDV is a 4:2:0 signal, retaining only a quarter of the original color samples, as does DV. The lesser amount of color information leads to the all too familiar “jagglies” when compositing, especially when chromakeying.

The DVCPROHD codec utilized in the Panasonic HVX-200 records a discrete frame for each frame of the material being shot. This means that there are 24 individual frames recorded when shooting in 24p mode. The 4:2:2 color space afforded by DVCPROHD also provides a superior image.

Hope this helps.

Bill

[Bill Marcellus]

Get the Panasonic HVX-200 and forget about the Canon. HDV presents all kinds of problems when shooting and even worse when compositing- especially if you are doing VFX. The Panny has the DVCProHD codec- no GOP problems. You actually have discrete frames to composite and use in VFX work.

I have posted material from both cameras and the advantages that the Canon offers in lenses, etc. are greatly offset by the disadvantages of the HDV codec. As soon as you shoot anything with a large amount of motion in the frame and try to composite the footage you’ll see what I mean.

Bill

[Bill Marcellus]

Let me say that I am not a big fan of direct to laptop recording. We are just looking for the most efficient means of moving data from set to edit suite while also creating a backup file of all material. This project has a six month shooting schedule which will result in 13 hours of finished program material- so we’ll be dealing with a lot of footage.

We are very concerned about losing footage, securing a backup, etc. We are still exploring workflow options including utilizing the P2 store or using the Firestore.

Even if we transfer to a P2 store there is a possibility of data corruption on its internal drive- or in transferring that data to another drive via Firewire, etc. etc.

All of that being said I am excited about using this camera. The initial euphoria of eliminating the capture/logging stage has been tempered by the feeling of working “without a net” due to not having the comfort of a tape sitting on a shelf ready for recapture at any time. It does seem that no matter our choice of workflow we will be working without a net at some point in the process which is one of the vagaries of dealing exclusively in the digital domain.

Thanks for all of the input- and any additional advice is welcomed.

Bill