So Jim Called...
Monday, November 17, 2008 at 6:56PM 
As the comment count approached 100 on my last post, Jim Jannard sent me a note asking if he could explain some of the less obvious aspects of the new RED product line. It turns out that Jim has a complex, nuanced vision for how people will use these cameras, and he was eager to explain it to someone who might grok it.
Without trying to speak for Jim, his basic message was that he is indeed making the camera I'm asking for—just not at the price I was hoping for.
The secret lies in the format chart that he posted late last week, the one that details the various windowing options for the various brains. Jim's vision is that these windowing options represent not just a way for us to compromise field of view for frames-per-second, but in fact a way for filmmakers to design their own formats, project by project.
For example, on the Epic S35, one could design a project around PL-mount lenses that might porthole noticeably at the full 5K width, which is wider than Super 35. But window in to 4K and you're basically shooting with the RED One, except you've got better dynamic range and up to 115 fps.
On the Epic FF35, you could define exactly the 4K window with your vast sensor that would capture an Academy-sized image and shoot with anamorphic lenses. This would be different than the 4K window you'd shoot to for a 2:35 center crop with spherical PL-mount lenses. And the next day you could be shooting 6K using every millimeter of your Canon L-series.
When you look at it this way, Jim's building blocks offer a dizzying number of options, more even than was clear at first glance. It's a ton of rope, well more than enough to hang oneself (the post options just got officially redonkulous), but also pretty freaking cool. Owner-operators will be able to dial in custom crops that make the most of their glass, and play with the relationship between over-the-top resolution, FOV, and slow-mo capability. Jim called it "elbow room"—the excessive resolutions and slightly-bigger-than-standard gates are designed not to become new standards as much as to encompass a perpetual choice of existing ones.
This diagram is mine, not RED's, and is just guesswork and supposition and therefore probably completely wrong. Stop looking at it already!
Jim's perspective was that he had indeed offered the camera I want. The Epic S35, windowed to 4K, is a cinema-sized sensor that takes cinema and DLSR lenses and offers frame rates up to 115 fps.
It's also $28,000 without ports, batteries, or recording media.
To my way of thinking, the camera features most important to a filmmaker are sensor size, dynamic range, and maximum frame rate. Images should look good projected in a theater and on HD—anything more gets into specialty territory.
To me, the sweet spot that addresses all of these, that matches the film cameras that owner-operators prize for flexibility, would be a 3K super-35 sensor that went up to 72fps. My feeling is that within the panoply of RED's offerings, this sweet spot, were it to exist, aught to be at the affordable end of things. I know big chips cost exponentially more than small, and I do expect to pay for that size, but my size/cost-ometer has been irrevocably jangled by the Canon 5D MarkII—legitimately or not. Canon sells me a VistaVision sensor with LCD and electronics for $2500. I dared to hope that RED could sell me a sensor half that size with 1/4 the pixels for maybe twice as much—without LCD, battery, or recording mechanism.
Did RED miss a filmmaking sweet spot? Only if you, like me, hoped that part of "killing" DSLRs meant getting anywhere near their price range.
Jim certainly opened my eyes to a jaw-droppingly deep array of options and flexibility within his new range of cameras. When you start shopping for your RED brain, be thinking not as much about which format you'd like to shoot, but which brain encompasses the array of formats you may want to shoot. It's a deep, complex spectrum of possibilities that's not easy to encapsulate in marketing materials. I can see why Jim was eager to describe it in words.
So let me say for the record that I'm impressed, and I can't wait to rent the living hell out of these cameras.
But I had really been hoping to buy one.
And I may still. Don't poo-poo the 2/3 format. The wee Scarlet body will fit a ton of fast, wide lens options and will make for a sweet little package. Anyone who thinks that real filmmaking can't happen on an imager that size hasn't heard of Super16.
And the fixed-lens Scarlet has the potential to be a solid Rebel cam. I can't wait to learn more about it.
But as these cameras march toward reality, the competition is not standing still. DSLRs are only three simple steps away from being a viable option, and Panasonic, who have shown that they can overcome engineering inertia by creating consumer 24p, have that micro-4/3 HD rig in the works—a large(ish)-sensor, interchangeable-lens HD camera that is likely to be in the $1,000 range.
Regardless of what happens, we the filmmakers win (as if we aren't in hog heaven already). It's going to be an interesting year.
Reader Comments (32)
Hmm. Maybe RED is offering solutions to high-end movie professionals, but for prosumers and enthusiast fillmakers I still don't see a solution that makes sense among the suggested offerings.
For example, take that "cheap" 2/3" Scarlet. According to this chart, it only does 3k, right? If that's true, then it's pretty much useless for a lot of people like me, as most of us "cheapskates" (that would go for the all-in-one Scarlet) are interested only in 1080p. We aren't gonna be printing back to film any time soon. And wasting CPU cycles, bandwidth, storage, time for something we don't need, doesn't look like a solution to me. Let alone that many video editors don't support editing above 1080p, and the $1000 Cineform Neo4k utility might need to be purchased separately so we can export the 3k image in 1080p without much loss of quality (I am on the PC side). And IF Scarlet gets a firmware update to support 1080p windowed, I am not sure I will fall for the windowed mode either.
To recap:
1. Most enthusiast filmmakers, like myself, only care about 1080p for the time being. It becomes too much pain in the butt otherwise.
2. Prosumers don't care much about dedicated movie lenses that are too expensive, and want to use their goold'ol Canon and Nikon lenses.
None of the Scarlet solutions actually give these two markets what they want IMHO. Only the EPIC does things well, but the EPIC is made for a completely different audience.
It might make sense to have a 4k sensor just to downsample it back to 2k (to get a better image quality), but the rest of the processing/recording doesn't need to happen at more than 2k. Also, it's probably not a deal-breaker if the frame rate can't go beyond 60fps at 2k (!).
I had plans to buy a Scarlet. But because of the current situation, I might just go for something lighter (e.g. a yet-unannounced Canon AVCHD camera that's supposed to be the next-generation HV20/30 camera kind of thing, at around $1500), or for a 5D-MII if Canon ever updates the firmware to add more manual controls and reasonable frame rates in its video mode.
thanks for the fantastic post Stu. It hit on a lot of the feelings I have had since the Red announcement. I personally have been a bit taken aback by some of the negative reactions towards Red, though I understand a bit of the disappointment. The camera I want, or is close to what I want is there, but is out of my price range. I am still interested in one of the 2/3" too, though again, I wait to see what total cost we are really talking about. I am coming from a Rebel standpoint, so cost is a huge consideration. If I was a full fledged working DP, and I could rent my kit out, these cameras would have me spinning.
I think I and a lot of people expected the Scarlet to come out, and there would be no question what camera we would get. It still looks pretty sweet, but there is a question now. If Canon, Nikon, or Panasonic can deliver then I may well be going one of those routes. But that's okay, my heart wasn't broken, my heart is just still on the dating block.
I still have questions about whether one of the dslr makers will make a camera that we want with a real RAW or at least serious and robust codec, and has the rest (24p, no jello, etc.) If they do though, I have a real tough decision coming.
Thanks Stu! I think for any photographer/videographer another thing that is extremely important (especially for photojournalism and uncontrolled action) is having a compact lightweight DSLR.
This to me is just as important as having a full frame sensor. Even the frame rate, price, and compressed video (h.264 like the 5D Mark II) are not that big of a deal if the final kit weighs around 10-15 pounds like an epic DSLR modified camera would after all the necessary modules are strapped on.
If Canon, Panasonic, or Nikon can nail a FF35 with at least 60-fps (including 24, 25, and 30 support), shoot 12 Megapixel stills at a minimum, 1080p (3K max but not necessary), include mini XLR inputs and stereo speakers built in to gather nat sound, have unlimited recording times (instead of the 10 minute cutoff on the 5D Mark II), cost under 5 grand for the body, and most importantly weigh under 3 pounds for the body then we are in business.
Too much to ask? I don't think so... the only thing the new 5D Mark II falls short on are minor and this is only their first stab at a dual camera.
Can we make the 5D Mark II work in a controlled environment where you can cut and re-stage the scene? Absolutely! However video mode on these DSLR's (5D Mark II and D90) need to be reliable at all times because in the real world there are no second chances to get the shot because a quick pan resulted in Jello or you were recording a historical speech and the video cut out on you every 10 minutes to miss key moments in history...
We are making progress but for me we are not quite there. However just like Stu said, “Regardless of what happens, we the filmmakers win (as if we aren't in hog heaven already). It's going to be an interesting year.”
Did you catch any intel on uses of 28k?
Great post regardless- I may settle for 2/3" -- I'm very interested in the high fps.
For now though I'll just keep shooting and getting better though.
Sadly after reading the boards and posts like this I realize that the market Red is going after with the Scarlet is not my own. I'm a hobbiest filmmaker who doesn't make films for money, but for the pure enjoyment of it. As such, even $3K is a stretch for me (yeah, that's right I have the rebel lighting kit - worklights). My best hope now is that these amazing cameras will help drive down the cost of the big boys competing products. If I could get something close to the EX-1 that shot on to affordable cards for somewhere around $2K I would be ecstatic. I have fingers & toes crossed for coming announcements at CES and NAB.
Thanks Stu. Another great post.
Someone on Red Crack called DSMC 'Lego meets Hummer' Its certainly that and i look forward to getting my hands on ANY of the building blocks. Glad Jim is listening..nay communicating with the community, it's been their great strength and one that has made them many loyal fans.
Agree on the 2/3" options being very viable especially for the $$ the fixed lens scarlet could well go for, when details are finally released I think it will be a real star.
As i've twittered a couple of times this week I recently watched "The Beckoning Silence" http://tinyurl.com/6cl6bj Shot on Varicam half way up mountains and it looks stunning (yes 2/3"). The DP's of that show would have walked a mile on broken glass up a snowy mountainside for what Jim is essentially now offering...a Varicam the size of a couple of Varicam tapes for less than the price of a good HDV camera.
The DSMC line up has many cameras i NEED to rent (645, FF35, the 617 even) and ill be bugging my usual rental houses to get in the queue asap But the one I hoped to buy, the sub $28K 'Stu cam...The 'Starlet' is one that many others are still looking for. Many of whom are after a "Canon D5Mk that knows someone"
• Small
• S35
• Raw images
• 3K... (they dont need 4 or 5 or 9K - the S in DSMC just complicates things)
¶ Overcranking -but they're realistic, 50fps & expecting some cropping to get there
We know this is getting close to the Red One but that does much better frame rates, better resolution and ramps.
The Scarlet FF35 is still better than the 'StuCam' with better frame rates and resolution but still no ramping.
The camera below 'StuCam' it is still viable too .. the lensless 2/3" its an essential element to the 3D system and will sell like hot cakes to everyone with an existing investment in Nikon, B4 or Canon Glass and the Guys that shot The Beckoning Silence with the now bleeding feet.
I feel pathetic even typing these words as it seems like an ungrateful child at Christmas. especially when Ive already had SUCH nice presents already and SO many nice things from last xmas still to play with...but at least this year we know that Santa is real ...and he wants to listen.
Jas
Great post Stu.
When I take a step back, it actually makes sense that the camera you really want is $28,000 for the brain alone, I just wish it was cheaper.
It's very hard to go down the path of a 2/3" sensor camera and lenses when a Canon FF is so cheap, and so close... and you know some other company has got to be just around the corner with something closer to what we want.
Assuming you DO want to go with Red, you could start with a 2/3 body and a few used C-mount lenses. Then when you save up enough, you can throw down for a better sensor, and you haven't wasted TOO much on the lenses. It's a way to spread out the purchase of the other accessories you'll need for either rig, and it would make one helluva crash cam.
Of course once you start planning on that sort of long view, you know there will be some competing products that make you regret your plan.
We're definitely on the verge of a quantum leap in cameras, and those leaps are often over valleys filled with bleeding edges and the crumpled bodies of the early adopters. Hmm... morbid.
Given all the stuff you can do with 2/3", S35, and FF35, and given the overall RED user base, which brain do you think would be rented the most? FF35 Epic, FF35 Scarlet, S35 Epic, S35 Scarlet?
Another thing- wouldn't you think the 'sweet spot' would be targeted at the largest market, so they could make the most use of economies of scale?
I'd imagine Red's largest potential market are all the zillions of pro-sumers out there, rather than the more rarefied professionals.
Alas.
I'm sure this sweet spot is technically attainable but not sensible in business terms. The big professional manufacturers (Sony, Panasonic) have always crippled their prosumer products as not to infringe on their top of the line products.
I'm hoping a maverick company who's not beholden to their elite market will sweep in steal the thunder.
Fingers crossed.
I'm with you on this one. This is an absolutely amazing product lineup, but I still feel a wee bit disappointed that the Scarlet-configuration I was hoping for just isn't there. I'll probably be picking up the Epic for projects as soon as it becomes available at the local rentals, but I'm still unsure if I will invest in a Scarlet for my own personal use.
"Did RED miss a filmmaking sweet spot? Only if you, like me, hoped that part of "killing" DSLRs meant getting anywhere near their price range."
Spot on, Stu.
Just to quickly address a point which is being made by a few commenters: that 3k is too much resolution. Remember that RED (and dSLR manufacturers) report their sensor's resolution as the total number of photosites on the sensor. But these are http://en.wikipedia.org/wiki/Bayer_sensor" REL="nofollow">Bayer sensors and so a quarter of those photosites are filtered to detect only blue light, a quarter are for red and half are for green. Creating a 3k image where every pixel has a red, a green and a blue value from a 3k Bayer image (where each pixel is monochromatic*) requires some clever interpolation.
The bottom line is that (as Stu has pointed out) you probably want at least a 3k Bayer sensor in order to produce a really nice 1920x1080 RGB image.
Also, http://www.dvinfo.net/conf/photo-hd-video-d-slr-others/137810-open-letter-nikon-begging-pro-video-mode.html" REL="nofollow">I'm preparing a letter to send to Nikon begging them to improve their "movie" mode. Comments welcome!
* Apparently it's not strictly true to say that each photosite is monochromatic in a Bayer sensor. Apparently there is a small amount of colour bleed across neighbouring photosites and that this bleed can actually be used to good effect if the DeBayer algorithm is sufficiently rigorous. Many thanks to http://www.dvinfo.net/conf/cineform-software-showcase/137817-possible-downsample-raw-camera-without-demosaicing.html" REL="nofollow">David Taylor from Cineform for patiently explaining this.
I kinda feel the pain of folks who feel outpriced by the Red lineup, but I kinda don't. I, too, was hoping for a 35mm sensor 3K for $3K. Yet the first thing I thought when I saw that the camera I wanted was beyond my current price point was, "I will find a way to afford it." The masterminds at Red have turned the market on its head and made history with this new product line. They have made the impossible possible, bent over backwards and then some to serve their customers. They have performed something close to miraculous. I've decided to do my best to return the favor, to pull all my resources and ingenuity and find a way to afford the camera I want. Isn't that what filmmakers do? How many films would not have been made if the filmmaker thought "Well, I really don't see how..." Come on folks, get out there and hustle. That's what the guys at Red did. BTW - THANK YOU Red guys. Hopefully I'll be shooting on a 35mm Scarlet while all the naysayers are still sobbing in their cornflakes.
perhaps the advantage of taking a FF35 sensor and making it 3K instead of 6K is that you might be able to build in higher DR, ISO and FPS without having to crop the sensor (prioritized in that order)
an inexpensive 3K FF35 sensor could simply be just another brain option while keeping the ones already planned
Stu,
Great post.
Why wouldn't the Ikonoskop hit that sweet spot for you?
1-60fps.
Decent price (7,000 Euro or approx $10,000)
Wide variety of lenses.
The Ikonoskop seems pretty good to me. And it ships in a couple of months.
Thoughts?
For me the most disappointing thing about the Scarlet end of the equation is the almost-worthlessness of "windowing." If you *are* willing to trade DOF for slow-mo, 60fps just isn't quite enough overcranking for my taste.
I don't mind my slow-mo being sharper. I'd even be okay with the relatively logarithmic pricing curve, if the Scarlets were a compromise of DOF for windowed slow-mo (but we had legitimate slow-mo of at least 72 but preferably >90fps) and then the Epics gave us a no-compromise slow-mo *without* windowing. (Which they do -- it seems the Epics can do 5K and more >100fps no problem.)
I think 60fps is just too low for the Scarlet overcrank. Otherwise I think those S35 Scarlets might be legitimate, even in the >=$7K range. But as it is, the DOF alone is a wee bit hard to step up to for that price. If you didn't feel you were giving up a feature (120fps on the 2/3 sensor), you might actually pay for the DOF on that bigger sensor.
Of course the Epic is the camera Stu (and everyone) wants. No compromises. I'm personally thrilled (compared to the previous design) that there is a Scarlet with a 35-sized sensor I can put my own glass on. But that relative lack of overcrank is a bit hard to swallow. (Sorry, 60fps just doesn't quite qualify for me.)
I do look forward to seeing where the continuing Canon/Nikon evolution will take us. Especially by the time Scarlets are actually in the field.
The scarlet is very exciting stuff and I would love to have one...it is not very expensive so I could buy it no problem. But the thing for me is that at best I am a mediocre camera operator and it would be on par to going into the sport shop and buying a £500 golf club when it will spend most of its time firing clumps of dirt down the range.
I don't type using a sledge hammer, and probably shouldnt start filmmaking using one either.
Christopher, see Jack's post. My concern about the Ikonoskop is that its sensor is 1:1 HD. But that is a nit-pick considering its other features—I can't wait to see some footage. I think people will be bowled over by the uncompressed data rate though. Part of the genius of Recode Raw is that it has a low enough data rate to be usable on a laptop.
Ikonoskop footage is now on line at
http://www.ikonoskop.com/dii/footage
I'm busy downloading as fast as I can..........
Thanks for the link! The Watch movie is worth a look—there's some sparkle on the sharp edges, as one might expect from 1:1 debayering. Still, damn good looking stuff, and just early tests.
I mean, I see the benefits Jim is touting, but you can make the greatest camera in the world and if it's not affordable, people aren't gonna buy it. What these announcements have said to me is that RED is simply for professionals. Serious professionals with serious budgets.
I think RED is doing some amazing things but I also think they've missed an enormous opportunity.
Christopher... the Ikonoskop is super 16 sized sensor at 1080p so lensless scarlet will crap on it on most every count especially $$ . Also its EVF is a ridiculous 640x480 or something. LOVE the choice but not sure it will have a chance
I like how Jim contacts Stu to clarify why the camera that Stu (apparently) can't afford is so wonderful.
Red caviar vending machines in homeless shelters.
I do give big-time credit to JJ and Red for being so proactive on customer communication. I got a direct JJ response to a thread on Scarletuser where I was speculating the TBD status of the price on fixed lens Scarlet meant bad not good news (2/3" glass prices etc.) and the price would end up much higher that $3k - in fact in EX1 range.
But I again agree with you Stu. I've rented a Red One for shoots and looks like I will be renting these new Reds as well, not buying. If no DSLR or other camera comes forward by the time the fixed lens Scarlet debuts AND it's priced ready to shoot at EX1 prices or better, much less, than I "might" be a Scarlet buyer.
But that's a big step from a few months ago when I was "pretty sure" I was a Scarlet buyer.
Nice post. No "c" in grok, by the way.
well i guess this is in a way, good news for me. i'm a canon stills shooter and now my options have narrowed. red is definitely making tools for the serious film maker, which is great for them. but as a weekend hobbiest i'll now choose between saving up the $4000 (australian) for a 5D mark II body or wait for the next version of the XXXD series which will surely have video and also be as cheap as chips.
i'm aiming for the latter.
I understand that Ikonoskop guys are very busy, but is it so difficult, perhaps in a coffe breack to put the camera (or prototype) on a tripod and shoot something decent?
I have just found this nice video by Hiromi Kusaka, shot with a Canon EOS 5D Mark II.
http://www.youtube.com/watch?v=CdaybWRRDWI
(Dont forget to click watch in high quality)
Here is a trick to link directly to "high quality" http://www.youtube.com/watch?v=CdaybWRRDWI&fmt=18
You only have to add &fmt=18 at the url end.
IMHO this camera will have 24p "tomorrow". The day it also has raw or another low-comp output ....................
(fill in the line)
There can't / shouldn't be any disappointment or heat towards RED's announcement, because it's not an announcement - not close to any product lineup. Forget the technical aspects, take a cold-minded marketing viewpoint for a moment: RED's campaign is a pre-product-marketing teaser to test the market through blog buzz. This is a classic case of so-called 'full factorial test' as marketing people usually call it (i.e. laying out countless myriads of possible combinations and options), instead of performing educated guesses at what the market segments really want before actually designing a product. Or call it a wide-scale, 100%-coverage market survey, which is only prompted by the turning point in technology (causing the established rules being shaken from the ground up).
As have shown the tentative price lists (and the disappointment at the high price tags), the difference in RED is not just the product, it's in the marketing & design process where all possible options are laid out and shared with you instead of keeping that in the labs of product marketing. All possible because it's a privately-held company, hence no share price to be impacted by hasty communication - a whole new concept in handling the user community, much more proactive and transparent - RED's true, disruptive innovation (although much overlooked).
Read again the fine print ("Specifications and Delivery Dates are subject to drastic changes. Count on it and you won't be disappointed") and rest at ease. In this pre-product-marketing phase, you can count on Jim to be currently collecting your feedback to help sort out priorities for the future. I would count on future commmunications with some options dropped, others considerably changed according to buzz feedback. But don't be disappointed: what RED's doing is involving you at a much, MUCH earlier stage of the marketing & design process compared to other manufacturers. Once you realize it's such a different company, it's quite thrilling to be granted such privilege... and no hard feelings when the actual product comes out - whatever that turns out.
just a reminder of what an incredible device the canon is for the price
http://cweb.canon.jp/camera/5dmark2/index.html
wait for the countdown then click on the red box at the bottom of the screen
Thanks Bernie K, grok fixed!
Stu,
Sorry, but I just want low light performance. I know you want dynamic range, too, but it looks like it's RED's not going to make it for under $10K.
We'll be lucky if Scarlet 2/3" will shoot ISO 400 without tons of grain (the photosites are fully 1/3 the size of those of the RED ONE). And anyone seen pictures of the older G1 at ISO 400? Not pretty, and photosites are bigger on that camera than those on the Scarlet 2/3".
The EOS 5D Mk. II shows almost no grain at ISO 1600. Is there anyone else out there that wants this kind of low light performance on a real digital film (video) camera?
How about Jim at RED?