Are You Guys Kidding??

[Max M.]

Let not go to much off topic here. This thread is to let red know that we really need solutions for today, and not only for tomorrow or future proof. We're asking to bring us products we can use now and not what we should use tomorrow. 1080p is not an error like Jim says. I agree 4K will look much better on a big screen. But reality is now that over 90% of everything shot is finished in 1080p. So if red would just accept that and give us what we are asking for, that would be really nice instead of preaching it is wrong what we are wanting. I also want 4K on youtube and tv everywhere but that takes time and isn't worth the hassle yet. Content is king and 1080p will be fine for 100% of my work upcoming years. So please give us the tools we need now aka the prores module

 

[Peter Moretti]

As far as I know, since the latest firmware update, you can use RED naming conventions on the Atomos Samurai.

But actually there is no need for a proxy workflow.
Premiere on a decent PC does everything in realtime from the original R3D files.
Nobody needs to edit in 5k, so you set it to 1/8 debayer and your golden.

Frank

How much metadata is in those Samurai files?

 

[Peter Proniewicz-Brooks]

Let not go to much off topic here. This thread is to let red know that we really need solutions for today, and not only for tomorrow or future proof. We're asking to bring us products we can use now and not what we should use tomorrow. 1080p is not an error like Jim says. I agree 4K will look much better on a big screen. But reality is now that over 90% of everything shot is finished in 1080p. So if red would just accept that and give us what we are asking for, that would be really nice instead of preaching it is wrong what we are wanting. I also want 4K on youtube and tv everywhere but that takes time and isn't worth the hassle yet. Content is king and 1080p will be fine for 100% of my work upcoming years. So please give us the tools we need now aka the prores module

And it's coming, it has been confirmed in this thread, by RED. There's no need to persuade RED to develop one, its been confirmed as on its way. There are also 3rd party recorders that can give it to you now.

If the point of this thread is mearly to get a pro res proxy solution from RED then two are coming, thread is largely pointless.

 

[Bob Minervini]

yep, but the point of the thread is that we needed a solution yesterday. that didn't happen. thats why people are jumping ship.

 

[Peter Proniewicz-Brooks]

yep, but the point of the thread is that we needed a solution yesterday. that didn't happen. thats why people are jumping ship.

There is a solution, 3rd party recorders, sufficient for now, likely cheaper than even the as yet unpriced and in spec cheaper of REDs two modules.

 

[Mark Toia]

May 30th

arty:

 

[Scott Brown]

Lets also keep focused on the other camera issues that need to be addressed if Epic/Scarlet are to survive in the 4K battlefield!

Sort the noise issues, gives us decent audio with XLR inputs, built in ND's...only way any of these are going to happen us with a camera body redesign...failure to do this will just further compromise RED's cameras & cost them sales.

Scott

 

[Ketch Rossi]

This thread is all over the place. Not sure I even want to chime in here - but ... what the hell ...

There are many Redusers who create their own content from start to finish. They can pick what workflow works best for them and control the entire process. This is awesome. But it doesn't work like that larger films and TV shows.

One thing to remember, and it is VERY important, is that there us a COMMON GOAL between Redusers and RED. That common goal is to get MORE people to shoot on RED's cameras. PERIOD. If more people shoot on their cameras, there is more demand for the cameras and those with experience with it. RED sells more cameras and makes more money to grow and build more stuff ... so that you have more things to bitch about not shipping fast enough.

Things change.

RED was insanely disruptive and basically had zero competition as a 4K digital cinema camera. They could afford to be forward thinking and place some bets on how workflows for their system would be adopted and evolve with 3rd parties like APPLE. The problem with 3rd parties is that you don't own them. ARRI answered RED with a clear message - and that was "EASY". The camera was "EASY" to learn and use. Big buttons like a film camera. ProRes = "EASY". Their RAW solution was NOT easy ... or cheap. And because ARRI RAW is only 2.8K - when you compare LOG-C 444 ProRes in HD (now 2K) to ARRI RAW 2.8K - you CAN see the difference - but the difference doesn't justify the (current) expense and hassle. Recently, there have been some major studio films (over $30M) actually shot on 2K ProRes instead of ARRI RAW.

Things change.

RED Vs. SONY - Sony F55 vs. ALEXA XT, ALEXA XT Plus, ALEXA XT Studio, ALEXA XT M, ALEXA XT Deluxe Home Edition Extreme - Canon C300 … C500 … C600 … C800 … a GoPro that shoots 4K bursts … a "digital cinema camera" from BlackMagic … a "digital BOLEX" on Kickstarter … Panavision STILL trying to build their own camera … dogs and cats living together …

What does it all mean?

It means other companies got the RED memo. They all got the 4K memo too - it's just hard to read that memo in German. There's real competition now.

Like it or not, cameras will get better and cheaper. Every year. Every single year. And now that there is real competition, the major players are going to leverage on ANY weakness, real or perceived in the competition.

ARRI clearly does not have a new sensor ready. Especially if they are actually upgrading other camera components as part of their upgrade program like the fan and internal NDs, etc. - if there were EVEN CLOSE on a new sensor ... they would make that part of the upgrade even if they had to delay it.

So ... ARRI's "play" here is "onboard RAW recording simultaneous with edit proxy". SAME "EASY" workflow as before, but now you get RAW without spending a fortune and strapping a big brick on top of the camera. This would have been a GREAT move a year ago. Problem is .. as great as the Alexa sensor is ... it's almost as old as the RED's MX. ARRI now has to compete with SONY's new sensors on the F5/F55 AND RED's new forthcoming DRAGON sensor - both of which capture MUCH more resolution, extended light sensitivity and dynamic range. Alexa XT Plus is $125,000.

SONY's "play" is "the Alexa and Epic in a blender". They are going for the "EASY" perception as well. onboard RAW recording simultaneous with edit proxy - 4K and a new sensor. And a price point on par with RED's EPIC.

Hopefully, without stirring up a total shit-storm - I want to make a few comments on the RED ROCKET and GPU debayer:

I provided cameras & post for MANY feature films on RED before their was a RED ROCKET card. At the time - we were only concerned with processing dailies FASTER THAN 35MM FILM PROCESSING IN A LAB - which basically meant overnight. We would render HD edit media using 1/2 debayer - (which is good enough for editorial and viewing rushes) - and spread the RAW files over a few computers. It really wasn't hard ... but when we got the RED ROCKETS it all became faster and easier.

Things change.

Realtime is the old standard. Faster than realtime is the norm for transcodes now. On-Set and Near-Set transcoding is more and more popular. The problem with the RED ROCKET is not the price. It's THE PERCEPTION. On the street - RED ROCKET = "HARD". It's that simple. What is "EASY"? GPU debayer. Canon does it. ARRI does it. SONY does it. So ... RED's "workflow = harder". Now ... I can only assume that the debayer with REDCODE compression can not currently decode with same image fidelity as CPU or RED ROCKET. Especially since Jarred has indicated a new version of the RED ROCKET card in the future.

My opinion:

#1: If RED can let a few key 3rd parties do GPU debayer for dailies - EVEN IF THE QUALITY WAS NOT 100% THE SAME AS CPU/ROCKET DEBAYER - that would be VERY, VERY smart. Sure, I get the desire to control the quality of the image - I get that - but if you don't put it in the SDK - so all the DI/Finishing systems have to use CPU or ROCKET. Just implement a GPU debayer in a few key 3rd party dailies tools used by the majority - these tools don't have secondary color correction and are not used to "finish" projects.

#2: If RED can't or won't do #1 - they better evangelize the holy hell of of the Meizler Module. (I know I'm going too ) And to me - the MOST EXCITING thing about the Meizler Module is what nobody seems to even mention:

THE BEST COLOR WORKFLOW FOR DPs EVER

The fact that a DP can now just move some sliders on an iPad or iPad mini while standing next to the camera and looking at a monitor - seeing what the editorial is going to see - and he or she can quickly "crush the blacks more", "warm it up", etc. etc. - without looking over the shoulder of someone who is changing the color off of an SDI feed and saving CDLs - it's going to be a really big deal. It's faster, easier - and I think DPs will love it once they have experienced it. It's the future. DPs set the looks at camera. The "looks" are baked into rushes for edit/viewing - the same look is saved in the meta-data of the RAW file.

You can't do anything remotely close to that on an F55.

Anyway ... just thought I would chime into a messy thread with some thoughts.

That is some nice tough out post Mark, and that taking in consideration that you didn't even wanted to chime in ha ha.

I agree hearly with all the above, and then there is a Load more at our end that some people are simply too lazy to show or learn themselves ot then show their prospected clients, all I know is that, I as well as literally every single company and professional individual ares stunned by what can be achieved with the RED EPIC now, wait till DRAGON enters the skies and we'll talk again...

 

[Rob Macey]

From my experience, even still, a RED workflow is an expensive mystery to deal with according to some TV producers I work with. Not all. I'm not generalizing. But, some. Not their fault. That's their perception. My work is network TV, TVC's and cable TV shows. ENG based cameras are the norm for my shows which makes perfect sense as to why RED is a mystery. I've shot ENG professionally since '94. Now, I keep pushing for RED on every project possible. Some, more than others. We're starting a 3 year production on the new LAX airport being built now. TVC's, Documentary, Time-Lapses. Shooting on Epic mainly. Could've gone Alexa, F55... Chose Epic. Bought Epic. It's the best.

Here's what I'm doing to de-mystify some of my clients' perception.

I rarely ask clients "what camera do you need?" I usually ask "what codec?" Depending on the project type, I'll mention I'll use my Epic. Cue immediate reaction "Oh, we haven't worked with RED files before. How does that work?" So, I mention not to worry, I'll record to Pro Res, DnxHD, .MXF, or .MOV. to break the ice, a nice security blanket... When my Pix 240 is burning them a ProRes or DnxHD file, baked in of course, they "feel" better because that's the easy, comfortable and "confident" way to post for them. It's what they're used to. It gets RED in the door where another one of my cameras would have been.

Good... But, even though it's nowhere near it's raw potential, it's a start. I look at it as a gateway drug. Once I get them hooked on RED and ProRes/DnxHD, which is great but limited, then I hold their hand and show them what Red Raw can do, work with them about file storage, rcx and grading.

That's it. So far, it has worked. I keep it simple. You want ProRes? Fine. You want R3D's? Great! A few stay ProRes. Totally fine for their needs. They love it. Looks beautiful. Even though it's like driving a Ferrari through a school zone, it get's RED in the door for me.

 

[Nikhil Kamkolkar]

Lots of varying thoughts on this thread, yes.

Some want a proxy module. (I feel the time has passed. Meizler is the proxy module now. EPIC is a DIGITAL CINEMA CAMERA, dammit! Not a DSLR).
Some want an upgrade to RED Rocket (I feel we need to make it unnecessary, or more of a NVIDIA product than a RED product)
Some want...

...on and on.

Bottomline? IMHO, RED is trying to compete with Sony et al. Slashing prices is reactive. Everything they seem to be doing is reactive. Instead of leading the way, and keeping their eye on their customers, RED has refocused on their competition. The competition has caught up/leapfrogged/whatever... but, I hope RED moves their attention back to the current customers. And, takes some of the input into consideration when they allocate their resources/energies.

Post is huge.

Pro-Meizler and a mini-Meizler would be the solution. A less than $500 dollar RED Rocket card (subsidized if need be) is a short-term band aid, but ideally, a collaboration with NVIDIA would be better. Nothing less will be a strategic move. This is so clear to me, I'm sitting here tearing my hair out wondering why RED doesn't see this.

 

[jake blackstone]

Something, that pretty much everyone is missing. Red Rocket is NOT Red technology. It is designed and built by DVS http://www.dvs.de/products/video-boards/atomix-lt.html.
That's a fact. That's why Red doesn't offer RR repairs. They just re-sell Atomix card re-badged for Red.
And now comes the conjecture
At the time RR was originally offered, Red was desperate to offer something, that would process R3D in real time at full resolution. The only company at the time, that had the technology was DVS. DVS's Clipster was the first commercial product capable of real time performance without resorting to building massive render farms. So, I suspect Red was forced to sign a long term contract to distribute RR in exchange for exclusive relationship and prohibition of any other methods of decoding, except CPU. And now Red is stuck, until the contract runs out. As I said, I have ZERO information on the contract between Red and DVS, or even it's existence, so take this post for all it's worth

 

[Peter Proniewicz-Brooks]

Lots of varying thoughts on this thread, yes.

Some want a proxy module. (I feel the time has passed. Meizler is the proxy module now. EPIC is a DIGITAL CINEMA CAMERA, dammit! Not a DSLR).
Some want an upgrade to RED Rocket (I feel we need to make it unnecessary, or more of a NVIDIA product than a RED product)
Some want...

...on and on.

Bottomline? IMHO, RED is trying to compete with Sony et al. Slashing prices is reactive. Everything they seem to be doing is reactive. Instead of leading the way, and keeping their eye on their customers, RED has refocused on their competition. The competition has caught up/leapfrogged/whatever... but, I hope RED moves their attention back to the current customers. And, takes some of the input into consideration when they allocate their resources/energies.

Post is huge.

Pro-Meizler and a mini-Meizler would be the solution. A less than $500 dollar RED Rocket card (subsidized if need be) is a short-term band aid, but ideally, a collaboration with NVIDIA would be better. Nothing less will be a strategic move. This is so clear to me, I'm sitting here tearing my hair out wondering why RED doesn't see this.

We have been told that in addition to the the Meizler ther is some sort of mini version coming out later, proxy encodes is all we know this less capable version is doing.

Price cuts likely were reactive but your post implies that everything is purely reactive. I'm not sure this is the case.

Are Dragon, REDray, the REDray projector and the Crimson projector all purely reactive?

I should note that on the need/lack of Need for rockets, at least for Dragon they look to ahve improved the REDcode codec to take less space.

 

[Jeff Kilgroe]

You may be on to something there, Jake.

I'm also curious about Jarred's recent comments about a Turbo Rocket and that it would be more powerful, but also more expensive, coming back to FPGA considerations. I'm still scratching my head over that... OK, I believe all of it when considering FPGAs and the state of the industry, availability, production, etc.. Fine. But tell me this.. RED has developed their own ASICs for dealing with REDCODE. They're in the cameras, they're in the REDRAY player, they're in various modules that attach to the camera. Albeit most of the items that incorporate these ASICs are also rather expensive. But wouldn't it seem more appropriate for RED to be building hardware accelerator products built around their own ASICs, rather than pursuing third-party FPGAs???

I'm also inclined to believe that the Xeon Phi (Intel MIC) could have something very significant to offer here. The new GK110 NVIDIA chip found in the Tesla K20X and GTX Titan is also intriguing. It's still hampered in wavelet constructs as past GPUs have been somewhat, but also has some newer abilities and vastly improved FP32 and FP64 compute performance. I wouldn't rule it out so quickly as has been the case with prior Fermi and Kepler chips.

 

[jake blackstone]

I should note that on the need/lack of Need for rockets, at least for Dragon they look to ahve improved the REDcode codec to take less space.

As far as I understand, the codec is the same. Dragon sensor is demonstrably less noisy, allowing for higher compression levels, resulting in smaller file sizes for the same image quality.

 

[Scott Crawley]

You may be on to something there, Jake.
...
RED has developed their own ASICs for dealing with REDCODE. They're in the cameras, they're in the REDRAY player, they're in various modules that attach to the camera. Albeit most of the items that incorporate these ASICs are also rather expensive. But wouldn't it seem more appropriate for RED to be building hardware accelerator products built around their own ASICs, rather than pursuing third-party FPGAs???

Red only developed capable ASICs as part of the Epic program though, right? Did the R1 have playback @ launch? It supports Jake's theory if true.

 

[Nikhil Kamkolkar]

Something, that pretty much everyone is missing. Red Rocket is NOT Red technology. It is designed and built by DVS http://www.dvs.de/products/video-boards/atomix-lt.html.
That's a fact. That's why Red doesn't offer RR repairs. They just re-sell Atomix card re-badged for Red.
And now comes the conjecture
At the time RR was originally offered, Red was desperate to offer something, that would process R3D in real time at full resolution. The only company at the time, that had the technology was DVS. DVS's Clipster was the first commercial product capable of real time performance without resorting to building massive render farms. So, I suspect Red was forced to sign a long term contract to distribute RR in exchange for exclusive relationship and prohibition of any other methods of decoding, except CPU. And now Red is stuck, until the contract runs out. As I said, I have ZERO information on the contract between Red and DVS, or even it's existence, so take this post for all it's worth

If true, RED should buy themselves out of the contract. Would have been a better use of the cash flow than to drop EPIC prices.

Agreed, the new projector et al is not reactive, but people like me will think twice about buying in. I need my investment to not be a wash. This is not a short term relationship. So I need to see RED thrive and be the toast of the town. And PRIORITIZING post over a lot of the other work (yes, even the projector etc) would help them, IMHO, get there.

 

[David Battistella]

As far as I understand, the codec is the same. Dragon sensor is demonstrably less noisy, allowing for higher compression levels, resulting in smaller file sizes for the same image quality.

This has come up in a thread that was deleted. I presume it will be a major selling point at the NAB launch.
You might be on to something. Remember Assimilate had their hand in the windows version of redcine for a long time, then we got REdCine X

david

 

[Mike Kushner]

If a proxy module is released by Red, hopefully it won't require a different set of SD cards and SD reader. If it does require that i carry a different card reader and a different set of cards, i don't think i'll be interested in purchasing one. Already too many little parts and pieces to pack and prep for every shoot that are crucial to the shoot. Please listen, Red, and make it work with the current cards and reader. I'm sure i'm not the only one who feels this way.

 

[David Battistella]

If a proxy module is released by Red, hopefully it won't require a different set of SD cards and SD reader. If it does require that i carry a different card reader and a different set of cards, i don't think i'll be interested in purchasing one. Already too many little parts and pieces to pack and prep for every shoot that are crucial to the shoot. Please listen, Red, and make it work with the current cards and reader. I'm sure i'm not the only one who feels this way.

Did you watch the video on the red website? The meizeler module uses RED mini mags, a new type of recording media. We can not assume the proxy module uses the same media.

David

 

[Bob Gundu]

If a proxy module is released by Red, hopefully it won't require a different set of SD cards and SD reader. If it does require that i carry a different card reader and a different set of cards, i don't think i'll be interested in purchasing one. Already too many little parts and pieces to pack and prep for every shoot that are crucial to the shoot. Please listen, Red, and make it work with the current cards and reader. I'm sure i'm not the only one who feels this way.

It's going to have an adapter so you can use your current reader.