Epic and Scarlet release in 2010

[Emanuel A Guedes]

Interesting

 

http://reduser.net/forum/showthread.php?t=47139

[Brian Drysdale]

Not actually noticing a date, I guess they'll out when they're ready.

[David Mullen ASC]

The delays may work to Red's advantage except perhaps for sales to the rental houses if they spend a lot on acquiring Alexa packages. But for the private owner crowd, there always seems to be new customers and Red's products are going to be newer, have more features, and be more affordable than the competition's. The wild card is whether Canon gets serious about a decent digital camera that uses a DSLR sensor as the base but provides a high-quality HD recording for a decent cost (i.e. under $10,000) -- that would directly compete against Scarlet. But I suspect Canon will do something half-satisfactory, like still only offer h.264 recording.

[Emanuel A Guedes]

Yes, Jim has been very cautelous with dates right now. Does this mean anything? ;)

 

You need to go down along the thread in order to find where it is explicit the release date. A few would call it viral marketing, the believers will say 2010.

[Saul Rodgar]

"When one of the most respected companies in the industry is just releasing their brightest new star that does 1/5 of the resolution, half the maximum frame rate, in a box three times the size and for 3 times the price… that should tell you how tough a trick EPIC is."

 

"1/5 of the resolution?" So I take it Jim is referring to the Alexa here? If so, Jim undoubtedly is now gone beyond the pale, even for a marketing-man, and officially lost his mind: That is right, Jim, thanks to your god-given right to design amazing camera circuitry, your cameras have the ONLY technology capable of high resolution data motion pictures the world will ever know, and everything else out there is just, well, crap . . . :rolleyes:

[Chris Burke]

The delays may work to Red's advantage except perhaps for sales to the rental houses if they spend a lot on acquiring Alexa packages. But for the private owner crowd, there always seems to be new customers and Red's products are going to be newer, have more features, and be more affordable than the competition's. The wild card is whether Canon gets serious about a decent digital camera that uses a DSLR sensor as the base but provides a high-quality HD recording for a decent cost (i.e. under $10,000) -- that would directly compete against Scarlet. But I suspect Canon will do something half-satisfactory, like still only offer h.264 recording.

 

 

i was saying this back in 2003 when I bought my first DSLR. I also noted that how slow Canon was to take advantage of market positions they have had over the years. They are always playing catch up to whomever or whatever the latest is. At the end of it all, Canon is a lens company that has always given the back seat to the cameras themselves. h.264 is good, but not that good. It is barely adequate for the prosumer market.

[Emanuel A Guedes]

The delays may work to Red's advantage except perhaps for sales to the rental houses if they spend a lot on acquiring Alexa packages. But for the private owner crowd, there always seems to be new customers and Red's products are going to be newer, have more features, and be more affordable than the competition's. The wild card is whether Canon gets serious about a decent digital camera that uses a DSLR sensor as the base but provides a high-quality HD recording for a decent cost (i.e. under $10,000) -- that would directly compete against Scarlet. But I suspect Canon will do something half-satisfactory, like still only offer h.264 recording.

To disregard Canon would actually be a full mistake, IMHO too. I think it has been a full surprise even for Canon. That's why they're still in the h.264 side, despite the power they have.

 

Apart the usual talk about the pros and cons as far as big screen concerns, without mention for online purposes (an important market to not neglect), I am in touch with people working for TV practically in a daily basis where the 5D Mark II has surpassed the most conservative expectations indeed.

[georg lamshöft]

"1/5 of the resolution"

 

What about spending a little bit less money and effort into marketing and invest into craftmanship, production, social/environmental/eductional standards instead? Yes, that would actually mean to risk something, making long-term investments that don't pay off within 3 or 4 years. Or simply accept the side-effects of 21th-century slave-labour and outsourcing policy.

But whining about being behind schedule (it's not just the mysterious "bug") while not willing to move production out of a sweatshop and at the same time claiming to offer similar (or superior) quality standards than companys that actually pay their (yes, they actually PRODUCE things - these 19th century basterds) production staff well, that care about complex processes and well-trained technicians and have to endure a lot in comparison to much cheaper competition but still stick to their standards? I seriously hope that for once, at least in this market, cheap, stylish sweatshop-stuff won't win over craftmanship, please!

Edited July 6, 2010 by georg lamshöft

[David Mullen ASC]

If you compare 5K RAW recording on the Epic (probably 5120 x 2700) to 2.8K RAW on the Alexa (2880 x 1620), that's about a 3X difference, 13,824,000 pixels per frame versus 4,665,600. I'm not talking about measurable resolution, just recorded pixel dimensions.

 

If comparing a 2K RGB finish to a 4K RGB finish, that's a 4X difference in file size (if comparing a 2K finish to a 5K finish, I guess that would be a 5X difference but I think it makes more sense to compared recorded dimensions until one knows more about the resolving power of each camera).

 

Epic is being manufactured in the U.S. by the way.

[Saul Rodgar]

Unless they are re-writing their RAW code, let's not forget RED "RAW" is heavily compressed. So RED's got massive sensors whose output is put to through the wringer to keep the data in manageable file size. Last I heard, Alexa's RAW wasn't nowhere near as compressed as RED's, if at all. So while RED's camera may have a much bigger sensor and pixels on the frame, there is no telling what one is actually missing due to RED RAW's compression. The smaller Alexa frame without RED RAW compression could theoretically push Epic's bigger frame with far more compression out of the perceived resolution game.

Edited July 6, 2010 by Saul Rodgar

[David Mullen ASC]

Why don't we just say that a 2.8K RAW recording is going to make an excellent 2K RGB D.I. and a 5K RAW recording is going to make an excellent 4K RGB D.I. -- and that's a 4X difference. Sure, Red compresses RAW but the resulting resolution after conversion to RGB is well-documented by now. You can figure on getting approx. 75% of the recorded RAW Bayer resolution back as measurable line resolution.

[Thomas James]

Converting 5K to 4K introduces scaling artifacts which damage picture quality.

[David Mullen ASC]

Converting 5K to 4K introduces scaling artifacts which damage picture quality.

 

It's not exactly a simple rescaling since you are talking about a Bayer pattern being converted to RGB, there aren't equal red, green, and blue photosites in the original anyway so it may not be a simple scaling from 5K to 4K, it may be a conversion from 5K RAW to 4K RGB. Besides, there is a lot of resizing and rescaling all the time in D.I. work. I also think that as the resolutions get higher in general, you probably have more leeway for rescaling than you do with low pixel-count images.

 

Certainly ARRI believes in using a 2.8K sensor to get a 2K image rather than build a 2K sensor. If you only get 75% of your original resolution from a bayer sensor RAW image when converting to RGB, it makes sense to start out with more resolution.

[Phil Rhodes]

"When one of the most respected companies in the industry is just releasing their brightest new star that does 1/5 of the resolution, half the maximum frame rate, in a box three times the size and for 3 times the price… that should tell you how tough a trick EPIC is."

Yes Jim, it's amazing what you can do when you don't really give a poop.

[Keith Walters]

Interesting

No, not really.

Actually, we were hoping that if we left the lights off and didn't post in this folder, everyone at RED would think weren't home, but that blasted Brian Drysdale had to go and answer the door :lol:

 

"There's noting to see here. Move along..."

 

[David Williams]

1/5 the resolution is another of Reds efforts to redefine common terms to suit themselves. There's a whole thread at reduser with Jim justifying it. Resolution at reduser now means total pixels, not resolved lines as the rest of the world understands it.

 

5 times sounds far better than the 1.5 or so you might be able to resolve on a resolution chart.

[Thomas James]

Actually what Jim says is correct when you double the spatial resolution from 2K to 4K you actually get a second power exponential increase in resolution which is four fold. When you double the resolution in the third dimension you get a third power exponential increase which is an 8 fold increase in resolution. Doubling the temporal resolution gives a fourth power exponential increase in resolution resulting in a 16 fold increase. And doubling the color resolution of the system results in a fifth power exponential increase which is a whopping 32 fold increase in resolution.

[David Mullen ASC]

Actually Jim was comparing actual line resolution measured on a chart between a 1080P frame from the ARRI to a 5K frame from the EPIC when he came up with that 1/5th figure.

 

Now you can say that it is a bit unfair since the ARRI has the potential of recording 2.8K RAW once that feature is activated, but on the other hand, a lot of people are going to shoot in 1080P on the ARRI, though I hope once ARRIRAW is available, it will be 1080P for TV work and 2.8K RAW for feature work, probably finished to 2K. But then, how likely is it that all 5K RAW projects will be finished to 5K RGB? You could say that 5K RAW is more likely to be finished to 4K RGB and 2.8K RAW is likely to be finished to 2K RGB.

 

I think if numbers are going to be thrown around, there has to be more background given to the nature of the tests, that's all.

 

But while the 1/5th figure seems high, remember that 4K is 4X the information of 2K, so the 5X figure is actually not hard to believe when comparing a 1080P recording to a debayered 5K RAW recording. But comparing HD out of the Alexa to 5K out of the Epic is a bit like comparing the worst option out of one camera to the best option out of the second, even if that's a likely scenario.

 

But if you are comparing 1080P to 5K RAW, and you figure that 1080P (1.9K) measures out to below 1.9K, more like 1.6K, and 5K RAW would measure out to something close to 4K, then that is a 5X difference in information.

[David Williams]

Actually what Jim says is correct when you double the spatial resolution from 2K to 4K you actually get a second power exponential increase in resolution which is four fold. When you double the resolution in the third dimension you get a third power exponential increase which is an 8 fold increase in resolution. Doubling the temporal resolution gives a fourth power exponential increase in resolution resulting in a 16 fold increase. And doubling the color resolution of the system results in a fifth power exponential increase which is a whopping 32 fold increase in resolution.

 

In the context of film and video cameras, resolution has always meant resolved lines. With video, removing variables like lens resolution, debayering, etc, you need four times the pixels to gain two times the resolution.

 

You can't redefine common terms just to suit a better marketing line. This in my mind is one of the major causes of antipathy with Red amongst many cinematographers I know. Changing the rules to suit yourself doesn't come across very well in any context.

[Adrian Sierkowski]

David (W) I do believe changing rules as such is called cheating ;)

[Brian Drysdale]

 

But if you are comparing 1080P to 5K RAW, and you figure that 1080P (1.9K) measures out to below 1.9K, more like 1.6K, and 5K RAW would measure out to something close to 4K, then that is a 5X difference in information.

 

 

I'm not sure how the 1.6K figure comes about. Assuming that Arri are going to debayer for HD, the camera should be capable of a 1.9k with I calculate as 2700 (Bayer) when shooting HD. The Arri table uses the same figures for the RAW and HD, even through the latter cover a very slightly smaller area on the sensor. http://www.arridigital.com/technical/cameraspecs

[Chris D Walker]

The reality of the situation: The Alexa is shipping units right now as per their original release date, no delays. The Mysterium-X upgrade is now available for the RED. The Epic and the Scarlet have not been released and have suffered several extensive delays. Will they be released by the end of the year? Maybe.

 

Why is resolution such a big deal? 80%+ of the viewing public don't have 20/20 vision. People can't resolve 4K+ of resolution. Sharpness, contrast, latitude and refresh rate are more important for great images than sheer number of pixels. We know this yet ignore it. For the majority of applications HD/2K is ideal in terms of image quality, data management etc. for presentation.

 

Another thing, how many RED shoots master their projects in 4K? I imagine very few. In Hollywood, sure, but for independent features they mostly finish in HD/2K, right?

 

This isn't an attack (despite having done it before with my 'I hate RED' post), but a real question of application in production and post.

[Keith Walters]

In the context of film and video cameras, resolution has always meant resolved lines. With video, removing variables like lens resolution, debayering, etc, you need four times the pixels to gain two times the resolution.

They didn't talk about pixels then, but all of this is pretty well covered in my 70 year old Television Engineering Texbook!

 

You can't redefine common terms just to suit a better marketing line. This in my mind is one of the major causes of antipathy with Red amongst many cinematographers I know. Changing the rules to suit yourself doesn't come across very well in any context.

It's had pretty much the result you'd expect.

There are 7,000+ REDs, and something less than 500 Genesis's and F35s out there.

As far as I'm aware, no true Prime-Time has ever been shot on the RED, (although the goalposts that define Prime Time are of course freely available for interpretation). There's not much film being used for TV any more, it's pretty much carved up between the Genesis, the F35 and a fair bit with 2/3" cameras (mostly Sony).

I'm sure there is an enormous amount of frames of REDcode being shot, but far from being a paradigm shift, it's just low-level noise spread over a vast geographical area. As far as so-called "4K archival value" goes, for the vast majority of it, low-band U-Matic archival would be overkill...

 

Meanwhile, the vastly less capable 5D is getting lots of eager attention from top-dollar operators.

 

What does this all say?

It says that people

A. Don't like delivery systems that make you wait

B. Don't think much of the idea that they MUST either:

(1)Convince their regular Post outfit that they need to equip for Redcode or

(2)Put their precious footage in the hands of someone they don't know, just because they have REDcode capability (and judging by some of the posts made by certain people on REDuser, they wouldn't want to know them).

C. Don't like being lectured with false dichotomies as in (1) and (2) above, when the simple answer is "no thanks, we'll stick to what we know".

 

If people don't like you, they won't hire you. Nobody likes a Zealot. Get over it.

 

If Jim Jannard could only have only have prised himself loose from this millstone/albatross called REDCode and offer a proper realtime delivery system,(And NO! I don't mean REDCode RAW!)the RED would have slaughtered the broadcast market.

The RED should be first and foremost a 1920 x 1080 realtime camera with option of non-realtime 4K (WHATEVER that means).

As it is, they have the whole concept completely arse-about, with exactly the results a lot of people predicted.

[David Mullen ASC]

I have completely the opposite opinion -- to design a digital cinema camera around 1920 x 1080 systems and make it beholden to 10-year-old HD tape workflow technology is completely assbackwards. 10 years ago, I has a conversation with Bill Feightner at Efilm about this, and he told me that the future was file-based data, not HD video, and that video would only become a deliverable option, not an origination format.

 

But to be practical, when the Genesis/F35 was designed more than five years ago, it made sense to build it around an HDCAM-SR recording and workflow. But that doesn't mean it was a good idea in the long run to make a digital cinema camera restricted to HD broadcast technology, and today, Panavision and Sony are looking into ways of turning the Genesis and F35 into a file-based camera by replacing the SR tape deck with data recorders. But the problem is that half the bulk of the cameras is due to the SR deck control unit being built internally into the body, so data recorders are only going to shave a few pounds off of these heavy cameras.

 

Post houses around Los Angeles are converting to file-based workflows and are not having much problem handling cameras like the ARRI D21 and Red that are recording data instead of HD video. Yes, it's an ongoing process and for television work, cameras that record broadcast HD do allow faster turnarounds. However, I would argue that if you shoot in PanaLog or S-Log on these cameras, you are complicating your post workflow for TV broadcast because of the additional Log-to-Rec.709 step needed. So RAW-to-Rec.709 is not necessarily a much bigger deal.

 

But this is all skirting a much bigger issue, whether these cameras are optimal for theatrical exhibition rather than HD broadcast.

 

I think Red has completely the right approach if digital cinema and independent filmmaking are the primary goals, which is using higher-than-1080P resolution sensors whose output is recorded as data, not in an HD video codec. Sure, a decent 1080P RGB recording option would make TV people happier and I believe that feature is planned for Red's upcoming Epic camera, but to have added that in the original Red One three years ago probably would have increased the price and perhaps even the size of the camera at the time for the added processing power needed. I believe the fundamental concept was a sound one and more and more digital cinema cameras are going to follow their lead, i.e. RAW sensor recording at 2K and higher.

 

There are many people who have not been happy about ARRI's decision to implement 1080P first in the D21 and Alexa when they know that ARRIRAW would deliver better results for people who don't need the fast turnarounds that 1080P recording allows.

[Frank Glencairn]

 

 

If Jim Jannard could only have only have prised himself loose from this millstone/albatross called RED Code and offer a proper realtime delivery system,(And NO! I don't mean RED Code RAW!)the RED would have slaughtered the broadcast market.

The RED should be first and foremost a 1920 x 1080 realtime camera with option of non-realtime 4K (WHATEVER that means).

As it is, they have the whole concept completely arse-about, with exactly the results a lot of people predicted.

 

I don´t get it Keith.

I can throw any 4k RED Material onto Premiere CS5 and - without any other special hardware than a 350 Euro Nvida card - I can edit it like DV or DVCHDpro. Easy, smooth, realtime.

 

What seems to be the problem (when they not force you to work on FCP or AVID)?

 

Frank