<<<-- Originally posted by Ignacio Rodriguez : > Since MPEG-2 is a lossy compression scheme, editing of the
> footage is destructive. Any edit would require regenerating the
> necessary reference frames and re-encoding

>I strongly suspect this particular MPEG-2 implementation >would not have intermediate frames, only keyframes, so >editing is possible without losing info.

Since the HDV video format supports MPEG-2 compressed video (MP@H-14), at two 16:9 resolutions:
- 720p (1280 x 720, progressive), at approximately 19 Mbps data rate
- 1080i (1440 x 1080, interlaced), at approximately 25 Mbps data rate

I don't see how you can get the 25 Mbps data rate without using both I (intra picture) frames and p (predictive picture) frames. They may keep the GOPs smaller, but editing will still be somewhat destructive and transcoding to other resolutions will result in some loss of data. That said, 25 Mbps is a good data rate (NBC transmitted the Olymics at around 22 Mbps), but I still don't like the idea of a single pass encoder.

The test of this would be to shoot in moving fog or clouds. That usually wreaks havoc with the simple encoders.

<<<-- Take Digibeta. It has more dynamic range and a better color space, but those are fine points when you're comparing image quality this close...it's indisitinguishable to the vast majority of the people who view what we create.

What is definitely distinguishable to them is a smooth rich detailed image in wide screeen and 24p motion...the things they associate immediately with film making. -->>>

This is my feeling as well. The most obvious distinctive features of film are the dynamic range and 24fps sampling rate. Aspect ratio too to some extent (and resolution obviously). This is what we've come to like and look for. Call it conditioned, imperfect, poorer than 60i motion since the sampling is done less frequently and thus the motion is blured and requires more effort from the brain to connect the images between them and perceive motion, it's still the way film looks, and people who want a film look identify with those elements.

What HDV has right now that gets it closer to film quality (I guess we could arbitrarely say 35mm is the standard format by which every other is compared) is a higher resolution, which is still miles away from film resolution of course. But when looking at film, even on TV with a 3:2 pulldown, the first thing that separates between video and film is the sampling rate used to acquire the image and the level of motion blur associated, introduced by the 1/48th shutter speed. After this, dynamic range comes into play to create dramatic contrasts that video cannot handle.

If HDV cannot provide on those 2 fronts, I'm not interested. When it does, I'll listen. Obviously Sony is not targeting users like me with this new HDV camcorder, which is why I don't see the point of "getting worried". No matter the quality of this Sony cam, it doesn't produce what I'm looking for.

Well I completely agree - getting back to the title of the thread - why should anybody be "worried?" If the XL-2 has the feature set that you want, and if you are happy with the price then I don't see how the Sony camera can take anything away from that. The XL-2 sounds really nice to me. So does the Sony. It's good to have options. If nothing else it may put some downward pressure on the XL-2 price.

Of course there are always some people who aren't happy unless they have the latest and greatest, and they will probably worry. But life is really short to dwell on that, so make your choice intelligently and then concentrate on using whatever tools you own to do something great.

<<<-- Originally posted by Joel Guy : With the amount of information currently released about the camera, there is no way it will be out in a month or two. Surely, Sony will want months of anticipation leading up to the release of their first prosumer HD camera. And just think about all the delays that accompany any major electronic release. To me, it just seems like Sony is trying to take consumer attention away from the new XL2, and to make those same consumers doubt their decision to buy an XL2. "If Sony's will be out in just a few months...." -->>>
That makes it sound like Sony decided to release this camera on a whim because of the XL2. This Sony cam has been in development long before the XL2 was announced.Guys, I think we need to be realistic. The Sony camera looks awesome, and with a 'pro' version coming after (XLR inputs, etc), I think it's a strong contender against the XL2. Canon ecided not to go with HD and it might cost them. You'll be able to buy the Sony for less money than the XL2, have HD and SD, flip out viewscreen, etc. The Canon XL2 is great as well, don't get me wrong. But don't bust on the Sony without having even seen it or used it. I think it's great that I get to decide between a Canon XL2 and a Sony HD unit!

We will have to wait for the real cost of the "pro" version of this. I've heard around the $7K mark. Now that's a lot more than the Xl2. But yes, If Canon had put an LCD on the Xl2, along with a good wide angle and barrel markings, it would've pretty much been a gimme for me. But unfortunately it wasn't to be.



Aaron

Wow
I didnt realise the debate that this 'simple' little topic would start

After reading through all the posts a few points have been identified that has made me feel a little colder towards the sony

1 - Higher resolution does not necessarily mean a better image. I think megapixels is a bit of a marketing ploy, being that it is assumed more is better, well let me tell you I downscaled a good photo to 1 megapixel and took it to a professional printers to test, and it looked much better than a 4 megapixel photo taken on a cheapo camera. Also my very good 4 megapixel camera outshines some higher megapixel cameras too.

It didn't really occur to me that this infact could happen to the digital video world, but it is now obvious that it can, and more importantly is. If sony can't get around the problems associated with this new mpeg2 technology 100% (which is still in it's relative infancy), then there is no point in having a high res, poorly coloured picture. It seems for now that the composition of the picture and colour reproduction on the XL2 should still shine through - for the time being anyway

2 - This does however lead me to a second point. I have always thought that if MiniDV is a digital tape, then it doesn't necessarily have to store info encoded in DV, why not be able to back up you hard drive on it (good idea hey? - 15GB per tape - each tape £2.00!) Anyway, with the continued and fast development of good video codecs is seems inevitable that soon someone will perfect a new codec that will squeeze as much as humanly possible out of the 25Mbit (i think thats right) bandwidth of the tape. How about a modified WMV9 codec or even a DiVX Codec (i know this doesn't look too great, but the data rate is tiny, so you could upscale the data rate quite alot and still be able to fit into the available bandwidth, and hence give a better image)

I think I will stick with the XL2 for the time being, and be happy that I can produce some of the stunning images I have seen from the various sample footages!

Lawrence

Aaron: just for my own curiosity, is the 16x manual (who has
markings) not good enough for some reason? And I assume the
flipover viewfinder has a too small an LCD screen for you to
consider it as an "LCD" ?

From a news article on MSNBC's web site this morning:

"Recording precious images at the best visual quality is important and the demand for that is sure to spread around the world."

How would you define "precious images"? Another quote from the same article may hold the key.

"The high-end offering, which Sony said was a world first as a high-definition camcorder for consumers, is instrumental for profitability at a time when camcorder prices are rapidly coming down, according to Sony."

The key words in the above are "camcorder for consumers." So I interpret "precious images" to be birthday parties, graduations, piano recitals, family picnics, and so forth. From everything I've read about this camera (and I haven't read it all), it is not aimed at professionals, as is the XL2, period.

Bill and Jim both have made excellent points and, based on my experience, I agree with them whole-heartedly. My XL2 is on the way and I can't wait to get my hands on it. I know I won't be disappointed.

Jay

A local news paper also ran such a story that this was "the first"
HD consumer camera according to Sony. Well I guess they haven't
seen the JVC. Or does that somehow not count as consumer and
theirs does?

i agree completely that single pass encoder is just not enough for particles (like water, fire, smoke..) or fast moving blur the lines action...

but then again dv does handle this that well either, without actually seeing the output from the sony cam i'm going to say i like what i see from dvcpro50...

as far as editing mpeg, i have been doing so since 1997 with my dc2000 and dc1000 cards in realtime, sd of course...

it can be done, and done well as in non destructive...

It's my understanding that HDV uses a GOP of 6.

Okay, I see what you mean, but still... you do know that HDCAM compresses the image, and in fact it also decimates the video from 1920 x 1080 down to 1440 x 1080 before compressing it and recording it to tape, right? And DVCPROHD employs DV-style compression as well. I don't think there's a system out there that stores uncompressed HD... maybe the Viper, but that's about it. All other HD systems are compressed.

And HD broadcasts are all compressed to MPEG-2. If you're watching HD over the airwaves, you're watching MPEG-2-compressed HD, at a 19mbps data rate, the same data rate that HDV records 720P at... at least HDV uses 25mbps for recording 1080i. However, I'm sure that a dedicated post compressor can do a much better job with those 19 megabits than an on-the-fly realtime in-camera conversion can!

You can get two streams of realtime HDV with effects and transitions right now, today. CineForm/AspectHD etc. can give you editing solutions that will run at full speed. But yes, it will be 4:2:0, not 4:2:2 (or the 4:4:4 of HDCAM SR!)

Oh, totally agreed. I don't think HDV qualifies as a holy thimble, or a holy anything. I'm actually one of the doubters who wonders if HD will have any substantial success in the marketplace, since I believe the consumer couldn't give a flying fig about HD. I'm just trying to keep the discussion on track, by pointing out that the HDV cameras do record full resolution fully compliant with the ATSC HD standards, so it IS high-def TV, and all HD gets compressed to MPEG-2 somewhere along the way, whether it's in-camera, during broadcast, or on a forthcoming HD-DVD.


Totally agreed. I did some side-by-side comparisons between the HD1 and the DVX100, and I thought the DVX was heads and shoulders above the HD1 in all measurable ways except resolution. I then took the DVX footage into S-Spline PRO (now known as PhotoZoom PRO) and up-rezzed it to 1280x720... and the resolution was *surprisingly* comparable to the HD1. At that point I forever ruled out the HD1 as having any merit, when I could get much superior *overall* results from my so-called HDVX100! The native 16x9 and newer tech in the XL2 may make it even more directly comparable after up-rezzing.

With that said, I'm *very* interested in exploring the new Sony, but absolutely floored/stunned/disappointed/in disbelief that they didn't include XLR audio. That's not an oversight, that's a direct crippling of the camera, and quite infuriating.

Go to the HD1/HD10 board here on dvinfo.net, and you'll find all sorts of clips and discussions about color correction and chroma keying.

The reason people are looking at the Sony as an alternative is because it offers many of the same options: it has true 16:9 CCD's, a 24fps simulation, a 30fps simulation, and a CinemaGamma mode.

Superb summary. The DVX/XL2 look filmlike, and while they do not look exactly like film, but they absolutely look like not-video, which is a valuable option to offer. I've been shooting the DVX for a year and a half now, and I think I've shot 60i video maybe for four jobs total, all the rest has been 24P. It's what people like and what they want.

> This does however lead me to a second point. I have always thought
> that if MiniDV is a digital tape, then it doesn't necessarily have to store
> info encoded in DV, why not be able to back up you hard drive on it
> (good idea hey? - 15GB per tape - each tape £2.00!)

Somewhat OT, but yes, there is a program that does this on the Mac, I think it even has a free version. I have never tried it personally, since it's so easy now to backup up to CD or DVD.

> I'm *very* interested in exploring the new Sony, but absolutely
> floored/stunned/disappointed/in disbelief that they didn't include XLR audio.

Trust me on this one Barry, it's really not much of an issue. If the miniplug can accept line-level audio or even mic level with decent electronics and a good AD (like the original XL1) you can connect a Beachtek or similar adapter if you need near-camera XLR. In a film-style shoot, you will usually have a better preamp close to the mic with line-out. Sure, it would be great to have an XLR mic input option for ENG, or line level XLR from when you run really very long cables, that is the neat thing about the Panasonic '30 and the PDX10, you can piggyback the audio modules if you need them, but it is not a serious shortcoming.

Besides, I would guess (wild guess of course, as usual) that the $7k forecast for the pro model is just to keep sales from going total zero in the next few months. I already spoke to a friend who owns BetaSX equipment, he was about to go DVCAM and he is now holding off buying ANYTHING until the HDV cameras get to this market. In the meantime, he has his SX stuff on sale.

Most likely the price difference for a pro version of the FX1 will be similar to the difference between the VX2100 and the PD170, perhaps a little more if there is real proscan.

Now if for $7K it will have an interchangeable lens, a shoulder mount and accept large format tapes, that would make sense ;-)