What I meant is that for the same quality you can get by with about half the bit rate, hence "more compression" [with maintained quality].

"OK, I going to get pedantic again... AVCHD is not capable of "more compression" as compression is entirely a function of bitrate. AVCHD is capable of higher quality video at a given bitrate (up to a point.)"

Does this have anything to do with Sony refusing to go from main profile AVCHD to high profile AVCHD? They are staying at 16Mbps in their high end consumer cams and just working on lenses, processors, and imaging chips. Do you really think they can squeeze that much more out of main profiles 16Mbps ceiling and keep up with Canon and Panasonic?

And NCIS and many other major TV shows ($2M/episode) have used HDV cameras when that was the best tool.

Im sure they have never used a HDV camera in filming NCIS. The digital is by Panasonic HDX. No production crew would shoot to tape. The season 2 DVD special features shows how the show is filmed. It's 100% an all out effort.

Here is a location shoot video:

http://www.truveo.com/On-the-set-of-NCIS-Filming-Fifth-Season-Finale/id/1330702694

Reviews have said no to that. 16 mbps loses it with too much movement.

Coursedesign- Reviews have said no to that. 16 mbps loses it with too much movement."

I don't know if I buy into that too much. I understand that bitrate is vital to picture quality, but so is lens quality, chip quality/density, and the processor. If it were up to the bitrate alone, then Canon's jump from the HF10 to HF11 should have been considerable in picture quality but it wasn't. They hit the 24Mbps ceiling but improved the video quality so minimally that its hardly noticeable. Why? I believe its because they ONLY changed the bitrate and left everything else alone. (other than a flash storage increase)

So if the new Digic DVIII processor (along with a much larger CMOS sensor) is going to help maximize the potential of the high profile 24Mbps of the HF-S10/100 then why can't Sony greatly improve on the main profile 16Mbps of the XR500/520V's by working over the lens, CMOS chip, and sensor? I think Sony is squeezing everything they can out of main profile before they make the jump up.

Im sure they have never used a HDV camera in filming NCIS.

Jeff,

Apologies for my rusty memory. It was JAG, not NCIS.

I never took the time to watch any of these shows regularly, but in this post, Spot alerted us to JAG's use of a Z1 in the cockpit of a helicopter where there was no room for the F900 they normally used.

I projected the show from OTA HD ATSC (which is amazingly close to Blu-Ray when a channel is given full bandwidth) on my 110" screen and looked real close. There was absolutely no problem with the waves, or anything else.

The Z1 is very long in the tooth today, but the 25 Mbps codec it uses has held up very well, and I have seen live quite a bit of footage with high levels of movement across a lot of detail.

When Panasonic was only selling intra-frame codec cameras, they were spending a lot of money on getting the message out that "GOP formats were unable to depict motion."

Then they saw an opportunity to make money selling GOP-based AVCHD cameras, so they immediately switched to saying that GOP was OK, but only if it used MPEG-4 instead of MPEG-2.

Manufacturers make a lot of claims, and it's helpful to analyze where they're coming from, to see if they have any self-interest in a certain outcome.

This is a problem also in other fields: in medicine, it was scientific truth for 50 years that eating eggs raises cholesterol.

Then recently somebody found the sponsor of the study that this was based on. Who was it? The Cereal Board, an association of breakfast cereal manufacturers who were very interested in getting consumers to spend their money on cereal instead of on eggs.

The Cereal Board gave a researcher a lot of money to prove that eggs were not healthy.

How did he do it?

No problemo. Instead of fresh eggs, he used egg powder. With that it was easy to make the statement that "eating eggs increased cholesterol substantially in 97.3% of the research subjects."

So a lot of people switched to cereal after being told by their doctors (who at that time had 30 minutes of nutritional science as part of their 8-year MD education, later increased to 8 hours, and a few years ago increased to 3 weeks at some medical schools) that eggs were bad for them.

So what was the deal with the egg powder? The supposed idea was that this would help "standardize" the sample, to avoid egg size variation, etc.

The much touted end result actually measured only that "egg powder increases cholesterol," but they didn't say that.

Egg powder oxidizes very quickly. It was this oxidation that created the cholesterol-increasing effect that led to an immense profit increase for Kelloggs and other fine manufacturers of breakfast cereal, while consumers eating real eggs never saw increased cholesterol from this...

So always check what was actually measured, and who did the testing, whether you are buying your next camera or deciding on what to eat for breakfast.

If it were up to the bitrate alone, then Canon's jump from the HF10 to HF11 should have been considerable in picture quality but it wasn't. They hit the 24Mbps ceiling but improved the video quality so minimally that its hardly noticeable.

If you shoot a static subject, there may be no improvement at all even if you double the data rate.

GOP codecs need more bandwidth when movement increases.

Cameras like Sony's EX1/EX3, for example, economize by using a 35 Mbps variable bit rate for highest quality. They use ye olde MPEG-2, but boy does the footage look good.

Cameras like Sony's EX1/EX3, for example, economize by using a 35 Mbps variable bit rate for highest quality. They use ye olde MPEG-2, but boy does the footage look good.

Yep, that EX-1 footage definately looks good!

I could care less about capture codecs, they all work okay IMO. Give me the 1/2" sensor block (or even 2/3") and a good lens anyday.

Sadly most of us are stuck with these 1/3" sensor toys (me included). Due to technology limitations, there just aren't any native 1920X1080 1/3" sensors available yet.

there just aren't any native 1920X1080 1/3" sensors available yet.
They'd be noisy, and very insensitive to light. Too much "stuff" packed into a small room, preventing much light from seeping through.
But you're right about the codec commentary. Most of em' today are 'pretty good' and all of them can be made to look great.
Last night, I saw some nearly finished pieces of an upcoming project we've got happening, on a 60' screen. I was dreading it.
Walked out with two guys next to me saying "what did you shoot that with." I didn't answer, not because I'm embarrassed to say I used a CX12, but because I didn't want to have the sort of whacked discussion of measurebation like this thread has spawned, and does so every time AVCHD is mentioned. At the end of the day, it's just another tool in a very large bag of tricks.
A professional has a lot of tools, and he knows which one to use for each task at hand for maximum efficiency.

They use ye olde MPEG-2, but boy does the footage look good.
No it doesn't. You don't know what you're not seeing. You don't know what you don't know. Didn't you see the back of Studio Monthly? MPEG2 is dead, dead end, over, done with.
Trust me.


man...my cheek hurts.

I repent.

Having seen this sworn testimony from P's marketing department on the back of Studio Monthly, I fully repent.

Lead me to my nearest Panny dealer so I may be absolved of sin.

"man...my cheek hurts. " Only??? ;-) This from a man who--- for fun, jumps from perfectly good airplanes with cameras strapped to various body parts without being chased by Ex's, pissed TSA agents, camera repair crews, Apple Fanboys.......

We all should realized long ago what the forum nickname "Spot" really meant! ;-) Makes conversions difficult when someone says " See that spot!", I quickly scan the Sat photograph for a man-make impact crater with camera parts scattered across it! ;-)

Now, this "Spot" wants us to believe "MPEG2 is dead, dead end, over, done with." Never, Never!!! ;-) Just remind me not to let him pack my chute, he'll probably add a few high-impact cameras for effect! ;-)

"man...my cheek hurts. "

Nothin like a little Panny advertising to bring out the laughter :D

Dave

Spot (and a lot of trees) will be saved by the fact that there won't be any more "back of Studio Monthly."

The printed publication is going the way of the trade show: to the internet only.

Wow, this was a fascinatingly long thread that took a long time to read. But educational.

So I wonder, when can we expect an nVidia CUDA assisted editing / rendering upgrade in Vegas?

"So I wonder, when can we expect an nVidia CUDA assisted editing / rendering upgrade in Vegas?"

Based on history I'd be very pleasantly surprised if we do.

If you're interested in understanding the complexities of AVC / H.264 there's a pretty full on article here:
http://www.vcodex.com/h264.html

You should not be too surprised to realise that mpeg-2 layer 10 has much in common with mpeg-2. Many of these "new' technologies are quite old, the theory has been around for decades, the silicon to implement it realtime hasn't.

Bob.

Food for thought:

Typed this up on another site in response to questions of the "next Genreation" avchd:

What makes you think there will be a "next generation" avchd?

(Sanyo has come out with the first consumer level 1080/60p cam... not avchd but rather avc/h.264)

blink3times- "(Sanyo has come out with the first consumer level 1080/60p cam... not avchd but rather avc/h.264)"

What's killing me is that it's not even a more "serious" consumer camcorder (i.e XR500V, HF S10, TM-300 etc.). It's something that soccer mom's and business men are probably gonna carry around. The manual controls are very lacking and still uses D.I.S., but I guarantee you though that it will produce awesome images.

No doubt!

This Sanyo thing is pretty cheap too. I think I heard a retail price of something like 500-600 bucks.

Now... if you took what Sanyo has done and stick it into a Sony or Canon quality cam.... can you imagine the picture quality!?!

personally I'm thrilled at being able to capture on an SR11 to small 4-8GB chip size memory cards I can then plug into my pc for ingestion/editing in Vegas, it's more than sufficient for seminar videotaping, greenscreen work, though maybe not for pro broadcasters and high end stuff, the video quality is much better than I ever got on dv-tape, and no more using a dvcapture software, so it's a huge timesaver...

-k

I agree with you, Ken. Maybe not for high end stuff, but sufficient for most of what I do as well, allthough mine is Canon ;)

/Ulf

From Shoot Expertise: First Look: Canon Vixia HF S10:

...what drew many pros, including cinematographers, to the HV20 and HV30 camcorders was not merely 24p, but their remarkable image quality. These petite HDV camcorders, which have collected a host of awards at tradeshows, demonstrated what a single 1/2.7in. CMOS sensor (larger than 1/3in.) combined with superb optics could achieve in a miniature camcorder.

I seriously suspect this camcorder will be on months-long backorder about 1 second after it arrives at B&H.