Ah, OK, found this - AVC-Intra - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/AVC-Intra) . AVC-Ultra has been talked about for a while, but I'd only heard it referred to an I-frame only version at 200Mbs - presumably more intended to rival HDCAM-SR. They now seem to have widened the meaning, so the version I was aware of is now "AVC-Intra Class 200".


Please remember that this is Wikipedia. AVC-Ultra was Vaporware for the past 3 years. I don't see citations or references, so this could all be hokum.

Everything after the F900 has been unncecessary :) ;).

Not always the case, but it's more that 8 bit formats can probably manage the productions that most people are actaully shooting.

There are trade offs going on in a camera design and you may find that you can have other 10bit codec, but because of compromises in keeping the cost down, other aspects of the camera's design don't in reality make it that worthwhile. It may be better having lower compression 8bit rather than higher compression 10bit.

Please remember that this is Wikipedia. AVC-Ultra was Vaporware for the past 3 years. I don't see citations or references, so this could all be hokum.
It was announced at IBC (video (http://www.youtube.com/watch?v=YnOc-5TQpRY)).

An overview of AVC-Ultra (in English, with Spanish subtitles):

Panasonic AVC Ultra on Vimeo

The problem with using the word ULTRA is.... then what? What do you call the codec once you implement 1080p 4:4:4? AVC-MAX? And then what? What do you call the AVC flavor that supports 4K in the future? AVC-SUPERDUPER?

Please remember that this is Wikipedia. AVC-Ultra was Vaporware for the past 3 years. I don't see citations or references, so this could all be hokum.
An overview of AVC-Ultra (in English, with Spanish subtitles):
Chris - that only refers to the I-frame only 200Mbs codec which has been talked about for quite a while - what Lawrence Bansbach referred to was a newer announcement: "its AVC-Ultra "family" of codecs. One is AVC Long-GOP which is indeed 10-bit 4:2:2 at up to 50 Mbps".

The wikipedia link talks of broadening the term to include three new codecs under the "Ultra" term - the original 200Mbs "AVC-Ultra" now becoming "AVC-Intra Class 200". (Confused? :-) )
Eight bits is enough when you...
1) Expose properly (that's the goal, but some extra headroom can save expensive, non-optimum footage.)
2) Get the S-curve and gamma just right, and
3) Don't grade.

But if you want to fix a non-ideally exposed image, mess with it's curve, and grade colors beyond reality while maintaining smooth gradients and natural texture, you need more bits.
But you don't just need more bits - you need to record something other than the fully processed video signal - S-log, RAW etc to get the most benefit out of any 10 bit system. Otherwise overexposure is overexposure and can't be clawed back no matter what the bitdepth. Yes, such will give far more control, but will require every shot to be graded - not a problem for some work, far too time consuming for other.

Unfortunately "10bit" just seems to have become a marketing silver bullet - "use a 10 bit system and it'll give you far more control, sir". The reality is a lot more complicated, and 10 bit is only one factor. Don't confuse 10 bit S-log with a 10 bit video codec.

The problem with using the word ULTRA is.... then what? What do you call the codec once you implement 1080p 4:4:4? AVC-MAX? And then what? What do you call the AVC flavor that supports 4K in the future? AVC-SUPERDUPER?

AVC- Flux Capacitor!

Still, even with that initial announcement, no products officially implement it yet. And furthermore, all references to the original announcement in official press form have long been down. That's what I call vaporware.

And in today's news:

Canon Hollywood Professional Technology and Support Center Premieres in Los Angeles at DVInfo.net (http://www.dvinfo.net/news/canon-hollywood-professional-technology-and-support-center-premieres-in-los-angeles.html)

Related to this Nov. 3rd announcement? You decide...

Kind of lends credance to those who postulated the big announcement was about a working arrangement with the Studio in question......which for most of us would mean nothing. Only a couple of weeks off now... Will this thread be sent to Area 51 ?? We will know soon...

The fact that that announcement was made today tends to mean that it isn't the one scheduled for Nov 3rd. :-) However, it does tend to lend weight the theory that the Nov 3rd announcement is more likely to be about something more relevant to the cinema industry than a photocopier......

And in today's news:

Canon Hollywood Professional Technology and Support Center Premieres in Los Angeles at DVInfo.net (http://www.dvinfo.net/news/canon-hollywood-professional-technology-and-support-center-premieres-in-los-angeles.html)

Related to this Nov. 3rd announcement? You decide...

Very very exciting.

Will this thread be sent to Area 51?I can assure you that it most definitely will *not* be sent to Area 51.

Nor will I be able to offer any more speculation in this thread after tonight...

See, See, Canon is taking over Hollywood & buying Paramount & RED!!

Jim Martin
Filmtools.com

Or could it be the other way around? Hmm.

"" Canon’s customer service and support network includes Factory Service Centers in Irvine (Calif.), Jamesburg (N.J.) and Newport News (Va.) as well as five broadcast lens service centers located strategically throughout the United States.""

I guess that means the Irvine crew will not move into the new facility, what will Hollywood work on ??


another interesting paragraph;

""The facility will also support business opportunities by offering 1:1 meetings with major clients (studios, production houses, television networks, rental facilities, and others) to further build relationships and assess their needs. The new facility and local Canon staff will also offer hands-on education to current and potential clients on the latest Canon professional products. Courtesy visits by both existing and potential clients in the industry can be arranged to explore opportunities and assess Canon’s full capabilities"".

It's beginning to sound like I won't be able to afford an ND filter for this thing.

Or could it be the other way around? Hmm.
Well - I speculated much earlier in this thread that whilst I don't see either Red or Canon buying the other one out, there remains the possibility of a joint venture in a single area. I still haven't seen a single informed comment which categorically rules that out.

Many are taking some quotes by Jim Jannard as being such, taking comments about a forthcoming "battle" as Canon v Red. As said before ( http://www.dvinfo.net/forum/digital-video-industry-news/500720-something-new-canon-nov-3rd-9.html#post1683761 ) all the comments I've seen could be taken the other way - Canon AND Red versus "the rest".

Whether or not it WILL happen I simply don't know, I have no inside knowledge - but it would make a lot of business sense.......

Ah well. Chris at least will soon be in a position to know the truth. Roll on Nov 3rd for the rest of us....... :-)

For those with long memories, we went through this a year ago discussing the absence of a solid state exchangeable lens model in the XF lineup. As I recall, the leanings were toward a solid state large chip XL in the fall of 2010 that was wrong or was cancelled when the market shifted:
http://www.dvinfo.net/forum/canon-xf-series-hd-camcorders/483458-how-do-you-see-canons-line-up-down-line.html

Canon's historical behavior and Japanese management style would say they won't rock the boat in a huge way as that might cause embarassment. I read some analyst that claimed the big hit of the 5DM2 was unforeseen and unintentionally brash.

Personally, I think Canon's usual "last to enter the market" behavior and strategy to eat away at the broadcast market from below (they have no upper end market to erode or threaten) point to a high end model. Coupled with the announcement of a Hollywood service center for filmmakers, it seems to me that it makes sense for the Nov 3 product to be targeted squarely at hi-end digital filmmaking.

If there's a price driven aspect to this market like there was at the time of the XL-H1 (Camera Review: The Canon XL H1 By Dirck Halstead / The Digital Filmmaker / (http://digitalfilmmaker.net/dv/features/xlh1review/index.html)), then the "Historical" aspect of the announcement is probably the acquisition specs (color space and buss width) at a pricepoint ... aka a quality digital filmmaking camera with good enough specs to take away some Red and low-end Arri market all at a price where you can buy several for the price of an Arri.

I can assure you that it most definitely will *not* be sent to Area 51.

Nor will I be able to offer any more speculation in this thread after tonight...

Looks like Chris will know after tonight .. but there's a Non Disclosure Agreement pending ?

IMO it's a new big pro cine camera to rival Arri and Red. They couldn't let it dribble out on to the market.
Probably be followed closely by a cut down prosumer version .. maybe even announced on Nov 3.

Cheers.

A lot of landmark broadcast programmes have been shot using 8bit HDCAM, many of which have been graded in post. Exposing correctly is one of those skills DPs need to develop, although it's not as difficult with video as shooting reversal film.

While this is true, 8-bit HDCAM - or 8-bit DSLR - doesn't give the latitude and bit resolution needed to allow us to push things to their creative limits. And while the goal is to expose properly, in high contrast environments, this is subjective. Do you expose for highlights or shadows? You can't have both. And, if the contrast is low enough to capture the extremes, you lose nuance in the mid- skin tones.

My son recently did the post work on a music video with content from two different DPs. It was daylight DSLR content shot before CineStyle was available. Sure enough, one DP exposed high, the other low. Neither was "wrong", but it made things tough in post, even for a fairly conservative grade. In essence, 8-bits is "brittle".

But you don't just need more bits - you need to record something other than the fully processed video signal - S-log, RAW etc to get the most benefit out of any 10 bit system.
True. By requesting "more bits", I'm really asking for a more gradable, less brittle output. Whether this is 14-bits linear, 12-bits S-log, or whatever is up to the manufacturer.

It would be nice if there was metadata included in video files that told NLEs what default lookup table to use. Drop it on a timeline or in a video player and the image would look nice right off the card. But you could still grade it to extremes. Best of all worlds - aside from the need to buy more memory cards and hard drives, of course. :)

Dear Jon,

We will implement, in our Gemini 4:4:4, the ability to record full uncompressed 10-Bit, S-Log or other Log, to one SSD while simultaneously recording to a second SSD with a Grading LUT applied.

The goal is to allow post to have the native S-Log, plus a copy with the grading LUT applied so that they can have an idea what was intended.

Of couse, we will allow, in the future for user loadable and possibly, user programmable, LUTs to be used for the grading LUT.

And the Gemini 4:4:4 will allow one to select a viewing LUT or not. We will have the ability, in future firmware releases, to select or not a viewing LUT for the LCD and each of the HD-SDI and HDMI outputs.

Our intial firmware release will not have all of these features, but we will have these as soon as possible.

We will implement, in our Gemini 4:4:4, the ability to record full uncompressed 10-Bit, S-Log or other Log, to one SSD while simultaneously recording to a second SSD with a Grading LUT applied.
It may be worth emphasing that this is only valid if the camera supports it? In other words, you couldn't make proper use of it with a camera that "only" had a straightforward HD-SDI output. (Even if that was 10 bit.)

While this is true, 8-bit HDCAM - or 8-bit DSLR - doesn't give the latitude and bit resolution needed to allow us to push things to their creative limits. And while the goal is to expose properly, in high contrast environments, this is subjective. Do you expose for highlights or shadows? You can't have both. And, if the contrast is low enough to capture the extremes, you lose nuance in the mid- skin tones.


The whole process is subjective, you have to embrace the limitations. The final display may be 6 bits or if you're lucky 8 bits and the final image will be have a lower dynamic range than the camera. On video you usually expose for the highlights, the creative part can be deciding how much highlight burn out you use. The curve used the camera will influence the decision, also if you're going to do much grading in post. Nailing it pretty much in the camera is the way to go with 8 bit for the best results and a lot of the fun.

I wouldn't compare a highly compressed 8 bit recording with a lightly compressed 8 bit recording. DSLRs are limited in what you can do with the exposure adjustments in post anyway, you don't want to be bringing up the noise in the shadows.

Dear David,

The Gemini 4:4:4 can record S-Log only if the camera can output S-Log.

If the camera outputs Rec. 709, we can record that also, or any other normal flavor of HD-SDI.

Single Link HD-SDI, Dual Link HD-SDI and in a future firmware release HD-SDI 3G.

The Gemini 4:4:4 supports 4:2:2 as well as 4:4:4 cameras.

It is proper to note that S-Log or other Log footage has to be generated in the camera, not in the Gemini 4:4:4.

Also, we plan to fully support ARRIRAW next year.

See, See, Canon is taking over Hollywood & buying Paramount & RED!!

Jim Martin
Filmtools.com

or the other way......

I like the way Red Canon rolls off the tongue.

My son recently did the post work on a music video with content from two different DPs. It was daylight DSLR content shot before CineStyle was available. Sure enough, one DP exposed high, the other low. Neither was "wrong", but it made things tough in post, even for a fairly conservative grade. In essence, 8-bits is "brittle".
.............

True. By requesting "more bits", I'm really asking for a more gradable, less brittle output. Whether this is 14-bits linear, 12-bits S-log, or whatever is up to the manufacturer.
Jon - fundamentally, I agree with pretty much everything you say. I just think you're putting the emphasis in the wrong place. It's not the bitdepth that's primarily important, rather what lies behind it. (S-log, RAW, or whatever.) Yes, they need a bitdepth greater than 8 to be realistic - but that's what follows on - not what's most important.

I suspect you are well aware of that in principle, but other people may not be, and get the wrong message from the talk of bitdepth. What really needs to be got across is that simply recording a 10 bit video signal isn't necessarily much advantage over 8 bit. It's when used in conjunction with something like S-log it really gives the benefit.

Of course, even with something like you describe (default look up table) the downside may still be far greater rendering. Hence why I'd like to see the option in any future camera of an either/or approach. Either processed video (to something like XDCAM422) OR something like RAW. Or maybe even both recorded at the same time......

Jon - fundamentally, I agree with pretty much everything you say. I just think you're putting the emphasis in the wrong place. It's not the bitdepth that's primarily important, rather what lies behind it. (S-log, RAW, or whatever.) Yes, they need a bitdepth greater than 8 to be realistic - but that's what follows on - not what's most important.

It's really both (bit depth and curve) isn't it? A relatively flat curve with subtle knees at the ends is ideal for capturing a high contrast image, but the bits that you give to the extremes are donated by the mid-tones. On the other hand, more bits can help smooth out a shallow gradient like on a mid-gray balloon, but will do little at the extremes is the s-curve is too strong. For instance, with the 5D2, I really like the CineStyle curve for high-contrast scenes, but I go back to Natural for lower contrast stuff where skin tones are critical.

We really want both, don't we? A nice curve and enough bits lets us capture a wide range as well as subtle textures and smooth gradients. And add really high compression quality to the ask. Lots of bits and the perfect curve do no good if the image is reduced to a bunch of large monotone boxes. :)

@Dan Keaton... The Gemini 4:4:4 solution sounds really nice (assuming that the camera can feed it.) It solves the problem David mentioned about sometimes wanting gradable material and sometimes wanting the fastest possible render and delivery. Having a record of the on-set grade answers some comments I heard from DPs at NAB: Digital makes it easier to fix things in post, but it can also undo the creative intent of the DP. By locking the on-set grade (and providing a safety recording), the DP can re-assert some authority.

Well - I speculated much earlier in this thread that whilst I don't see either Red or Canon buying the other one out, there remains the possibility of a joint venture in a single area. I still haven't seen a single informed comment which categorically rules that out.
...

I ceratainly understand the line of reasoning to your speculation, but I think as of 10:06 yesterday, Jim Jannard just categorically ruled it out.

"...
Canon on the red carpet in Hollywood and RED from the garage. Head to head.

Jim"

Yep I'd hate to be trying to sell high end video cameras right now. The area of the industry has likely gone stone cold.

Cheers.

Neither was "wrong", but it made things tough in post, even for a fairly conservative grade. In essence, 8-bits is "brittle".

You can expose RED improperly too and get bad results even with that. Nothing can really help if one dp exposes wrongly. We finished a big commercial a month ago where the first shot was shot way too dark with the RED. End result was huge amounts of noise.

It is getting kind of thick!... Really, what is going to shatter the earth? We are going to get news of products that might give more for less money than in the past. Canon better have some good stuff or they risk look like fools imho. At least RED can release their 4-years delayed Scarlet. Oh how pessimistic I have become!

Everything after the F900 has been unncecessary :) ;).

Best post in this thread. Thanks for the laughs Peter!

I think we now know its not likely to be a dSLR camera right?

It's pretty clear that

1) yesterday we got the news about the 1D X (shipping in March) to set initial expectations,
http://www.dvinfo.net/forum/digital-video-industry-news/501724-canon-usa-introduces-eos-1d-x-digital-slr-camera.html

2) on the 26th, we will learn more about consumer products, including EOS cameras, and

3) on the 3rd, we will learn what's up in digital cinema.

Something for everybody! :)

Less than 2 weeks to go!

Jim Martin
Filmtools.com

Ummmmmm.........anyone here that the Red Scarlet is going on sale Nov 3? I just read about it over on fxguide.com.

Also, in the same article, they say how they will be covering this Red Scarlet event as well as "the major Canon announcement earlier in the day". Nov 3rd is setting up to be a HUGE day for our industry. However, my patience will be put to the test with all of the Red and Canon fanboys going crazy over cameras they have never used and camera footage they have never seen.

anyone here that the Red Scarlet is going on sale Nov 3?

Mentioned in this very thread about a month ago: "RED will be announcing on... you guessed it... November 3rd! " (http://www.dvinfo.net/forum/digital-video-industry-news/500720-something-new-canon-nov-3rd-8.html#post1683618)

my patience will be put to the test with all of the Red and Canon fanboys going crazy over cameras they have never used and camera footage they have never seen.

Just stick with DV Info Net and you won't have that problem.

;-)

Whatever we have here, it's good for us.
Scarlet will likely be great, but it's difficult to read the Red forums gushing about prospective unavailable products. How do people do that year after year?
Will Canon give us video raw? Will Canon do more than 1080p? Interesting question here if Canon really "gets it".
Scarlet seems more straightforward as a 2/3 sensor with all the good red raw software. Canon's never made a big video specific sensor, so that's very interesting technically. Likely very distinct approaches by these companies, with many of us wanting aspects of both.

I expected the same (a video camera with 5D2 technology.)



I'm very confident it will be considerably better than that. Considerably. But I will be shocked if it's more than 1080p.

Scarlet will likely be great, but it's difficult to read the Red forums gushing about prospective unavailable products. How do people do that year after year?

for the record, RED has said that the Nov. 3rd announcement is for a SCARLET that is now done and ready to ship. final specs will be announced on that day.

looking forward to seeing what both companies announce, even though I'm not buying any new cameras any time soon.

I know the Scarlet that has been bandied about has a 2/3" sensor, but I wonder if RED decided to make a completely new model, with a larger sensor and interchangable lenses? It would compete more with the FS100, AF100 and F3, but offer RAW image capture.
Still, a 2/3" sensor, fixed lens camera with 3K RAW capture sounds pretty good to me.

I'm certain Canon's gonna give us a new cam from latest updates. Nothing else heard out, the folks on NDA r pretty tight about it.

^^^I don't see how Red could make a large sensor Scarlet considering the Epic-X costing $55K. Somethings gotta give to be under $10K. I think if you're Red you don't want to change the software side - I think Scarlet has to make redcode. So all I can think of is sensor size.
But really don't know.

I don't get Red's confidence if Scarlet is a fixed lens small sensor camera. Three years ago that was exciting, but not so much with what's available today. 3K, 2/3, and interchangeable is interesting to me. Not that they can't sell a bunch of these at the right price. But the whole 5DII phenomenon is about creative control and cinematic look. So was the Red One phenomenon.

I'm actually more curious about the Canon announcement. Their path to this point has been remarkably odd with the unexpected success of 5DII video. How does Canon exploit a major unexpected opportunity?

I know the Scarlet that has been bandied about has a 2/3" sensor, but I wonder if RED decided to make a completely new model, with a larger sensor and interchangable lenses? It would compete more with the FS100, AF100 and F3, but offer RAW image capture.

I thought the Epic S was supposed to be in the F3 price bracket. There's no reason why progress on that camera can't be announced on Nov 3rd together with the 2/3" Scarlet, there are markets for both cameras.

The 2/3" sensor size is great for documentaries. You don't need FF35 for a cinematic look, "The Hurt Locker" was Super 16

With info that RED has announced, it will have two models...

2/3 sensor, 3k @ 120fps

F-SCARLET (Fixed Scarlet): 8x Zoom lens, full fixed. 35mm zoom range of 28mm-224mm. T2.6/F/2.4 throughout the full zoom.

C-SCARLET (Cinema Scarlet): Smaller package due to lack of zoom lens. Utilizes RED Mini Primes which have a "projected" cost of $900-$950 for each prime. Option for a C-Mount and a B4 Mount.

Mini Primes:
6.5mm T1.9
8mm T1.5
16mm T1.5
25mm T1.8
50mm T2.9
75mm T4

This is what has been... confirmed by red as much as you really can confirm right now, but it sounds practical. The price point for the F-Scarlet has been mentioned by Jannard as being around 6k (my guess, 6500-7000).

I too would be surprised if Canon offered something beyond 1080p, and from the pace of things, I dont even know if 1080/60p is something we can expect.... If they offered 2k @ 60p, that would be pretty good and bolster their position in the market a whole lot better... If they want to compete with F3 they gotta do 1080/60 or better...

2/3 is nice for reach too. Nothing wrong with 2/3, it's just different than the big CMOS sensors that has received all the attention. Starting with Red One.

3K 120 fps requires a substantial processor package, which makes me wonder a bit about that spec.

I can't get excited at all about yet another lens type.

New Canon at closer to S35 size should have considerably better low light sensitivity than 2/3.

I thought the Epic S was supposed to be in the F3 price bracket. There's no reason why progress on that camera can't be announced on Nov 3rd together with the 2/3" Scarlet, there are markets for both cameras.

The 2/3" sensor size is great for documentaries. You don't need FF35 for a cinematic look, "The Hurt Locker" was Super 16
Anything is possible, but November 3 is Scarlet day. Epic-S was conspicuously absent from Red's latest roadmap. When pressed about it, Jim Jannard replied, "The camera you want is at the end of the year." That tells me that the Epic-S as previously specced no longer exists, at least by that name.

RED could come up with any number of permutations, so no reason why they couldn't change the Epic S name back to Scarlet and complete on the price with the F3 and possibly the new Canon. Their new factory would seem to be capable of producing a larger volume than I suspect the current Epic camera models need.

What turns up on the day remains to be seen, but the demand seems to be for a new camera(s) to be hitting the street, rather than announcing future cameras.

It's funny I saw a post from Jim Jannard on reduser stating that people should be realistic and this is a Scarlet announcement, I think the rumour mill is going nuts with people expecting a camera as good as Epic for under 10K

Indeed, there's a good chance that quite a few people will be disappointed by what both Canon and RED offer because it doesn't match their expectations.