KOMODO....

[Grant Smucker]

I have chats with people who think Monstro 8K VV should be a $10K camera, but those people aren't identifying the incredible lack of other 8K large format 60p cameras on the market. Even most of the 8K stuff I know is coming, the cheap stuff, tops out at 30p. And I'm not even getting into image quality related issues, features, or workflow. Same could be said about 5K 96fps cameras that RED offers in comparison to the rest of the market.

I have a hard time believing Monstro costs an additional $30,000 (than Helium) to manufacture. Both 8k and same body so the only difference is the sensor. It is a tremendous cam and one of few currently but it's just a novelty fee at this point.

 

[David Johnson]

I believe that Jarred (EDIT: Forgive me if I mis-quoted, can't find the source) said the biggest cost is in the number of sensors that don't meet spec. So for every Monstro sensor that's placed in a camera, x amount has to be tossed.

I imagine there's a considerable amount of $$$ tied up in R&D and other soft costs associated with that camera.

There's a whole lot more than cost of materials that go into the latest tech, especially when you're an innovator.

 

[Phil Holland]

I have a hard time believing Monstro costs an additional $30,000 (than Helium) to manufacture. Both 8k and same body so the only difference is the sensor. It is a tremendous cam and one of few currently but it's just a novelty fee at this point.

I believe that Jarred said the biggest cost is in the number of sensors that don't meet spec. So for every Monstro sensor that's placed in a camera, x amount has to be tossed.

I imagine there's a considerable amount of $$$ tied up in R&D and other soft costs associated with that camera.

There's a whole lot more than cost of materials that go into the latest tech, especially when you're an innovator.

Bigger sensors, no matter what = less yield on the wafer. That's sadly part of the costs that grow more exponentially on that front. I have no idea what RED uses, likely 200mm or 300mm wafers, but if you take into account the format sizes of Helium versus Monstro and potential yield, you'll likely get 2-2.5X more "in spec Helium sensors" off the silicon wafer alone, prior to potential other chaos that can occur in the process, theoretically that's 2X the price right there even on the core sensor cost before electronics are taken into account. This is why things like small cell phone sensors are so economical to make, you get a lot on a wafer (like a couple thousand) and if 1-16 the dies are bad, you still have a very high yield. There's a lot of forward moving effort to improve this when it comes to manufacturing, actually some big leaps just in the last few years I've learned through two of my clients, one of whom makes processors. Finding defects is like science fiction technology these days at the factories that make this stuff.

I'm certain there's other costs in terms of getting faster readouts on larger sensors and all that too. There's a rather large difference between a large sensor that rocks out at 2-5fps versus one that can do 60fps. Tolerances across sensors and QC is also possibly a fun thing to factor in. Don't know all that accounts for it, but it's clear that for a while now it's a unique sensor design that's alarmingly good. In some ways a lot of people don't even commonly look at.

Back when Dragon 8K VV was a thing I remember visiting Jarred at RSH before a rather large shoot and he had an entire wafer sitting on his desk of Dragon 8K VV sensors outlined on the full disc wafer, and the entire wafer didn't pass QC and I don't think that was a unique wafer from the batch. That's a chunk of cash alone. Big and high quality sensors are really, really hard to make. It's certainly one of the reasons many companies will go the route of buying a pre-fab sensor instead of going through the horrors of rolling their own. It clearly sounds expensive and painful.

 

[Paul Leeming]

I received my Blackmagic Pocket 6K today and rigged it up for cinema work. Let me tell you, this is, as pictured with everything aside from the lens, the realisation of 6K for $6K. This thing is INSANE for the money - a Super35, 6K sensor with 13+ stops of DR shooting raw data up to 60p in 2.40:1 ratio and 50p in full sensor use. All to a Samsung T5 SSD attached via USB-C. It takes timecode in for proper work on set (via the audio jack).

IMHO, if Komodo isn't full frame or close to it (maybe 1.2x or so), it's going to have a hard time convincing people to pick it over the Blackmagic. It's just the lay of the land in the latter half of 2019. For $6K I have an entire shooting package including batteries, chargers, sync, monitor/recorder, cage and more. I love Red cameras and have been in the Red world since Red One #378, but this Blackmagic Pocket 6K is a game changer.

Here's a still image sample of a chart test I did today (with my own LUT applied) for pixel peepers: Shot at BRAW Q5 at 24p ISO400 with the Sigma FF Cine 50mm T1.5 at around T4 (60MB PNG): http://www.visceralpsyche.com/misc/web_images/BMPCC6K_CamAlignMatch_WBOK_0.25NR_C_2.3.3.png The image quality is pretty amazing!

Cheers,

Paul

 

[Jeffrey Loewe]

I received my Blackmagic Pocket 6K today and rigged it up for cinema work. Let me tell you, this is, as pictured with everything aside from the lens, the realisation of 6K for $6K. This thing is INSANE for the money - a Super35, 6K sensor with 13+ stops of DR shooting raw data up to 60p in 2.40:1 ratio and 50p in full sensor use. All to a Samsung T5 SSD attached via USB-C. It takes timecode in for proper work on set (via the audio jack).

IMHO, if Komodo isn't full frame or close to it (maybe 1.2x or so), it's going to have a hard time convincing people to pick it over the Blackmagic. It's just the lay of the land in the latter half of 2019. For $6K I have an entire shooting package including batteries, chargers, sync, monitor/recorder, cage and more. I love Red cameras and have been in the Red world since Red One #378, but this Blackmagic Pocket 6K is a game changer.

Here's a still image sample of a chart test I did today (with my own LUT applied) for pixel peepers: Shot at BRAW Q5 at 24p ISO400 with the Sigma FF Cine 50mm T1.5 at around T4 (60MB PNG): http://www.visceralpsyche.com/misc/web_images/BMPCC6K_CamAlignMatch_WBOK_0.25NR_C_2.3.3.png The image quality is pretty amazing!

Cheers,

Paul

Nice Paul! Have you looked into the Lucaadapter? Curious on your thoughts of 120p at 2.8K if you have time to chime in.

 

[Paul Leeming]

Nice Paul! Have you looked into the Lucaadapter? Curious on your thoughts of 120p at 2.8K if you have time to chime in.

Haven't tested 120p yet sorry. I know of the LucaAdapter but coming from a Pocket 4K with the new BMPCC4k specific Speed Booster (which is the best one to date), I can really see the difference without extra glass in the way (extra resolution aside!). Metabones, though good, just isn't optically up there to eliminate the aberrations around the outer area of the coverage. I thought I'd have difficulty adjusting back to 1.55x after my 1.2x SB, but the 6K clarity is so much better that I don't think I'll go back to the 4K except as a B cam. If a proper tier 1 lens manufacturer like Sigma built a Speed Booster though, I might be persuaded...

 

[Savannah Miller]

I believe that Jarred (EDIT: Forgive me if I mis-quoted, can't find the source) said the biggest cost is in the number of sensors that don't meet spec. So for every Monstro sensor that's placed in a camera, x amount has to be tossed.

I imagine there's a considerable amount of $$$ tied up in R&D and other soft costs associated with that camera.

There's a whole lot more than cost of materials that go into the latest tech, especially when you're an innovator.

I would say most of that is sensor design and manufacturing. All the other stuff is peanuts, especially when you spread the cost over many different DSMC2 bodies. One of the clear ways Blackmagic saves money is to reuse as much as the can. Reuse cooling, monitors, body, IR Cut, Sensor, etc. A lot of the same stuff is in all of their cameras.

 

[Sabyasachi Patra]

I have a hard time believing Monstro costs an additional $30,000 (than Helium) to manufacture. Both 8k and same body so the only difference is the sensor. It is a tremendous cam and one of few currently but it's just a novelty fee at this point.

RED is a business. And there is something called premium pricing. Due to lack of other alternatives, they can charge a premium and still get customers. No amount of voicing opinions in this forum or elsewhere will matter as they won't simply care. It is a question of competition and they have an opportunity to charge the premium. Whenever I get a chance to charge a premium for specific shoots which utilise my strengths, I charge accordingly. So I won't grudge RED on that front. If there is enough opportunity in that market segment and if there is capability then competitors will enter that segment and offer a competing product. In the past RED has slashed prices and those were due to impending or existing competition. Obviously they will say that there is more yield. Will we tell people that we are going to lower our day rate due to competition from other DPs? If no, then why do we expect to hear such things from any company? RED mags may be a different story. However, I won't blame RED for monstro.

 

[Hrvoje Simic]

Blackmagic seems to give the most reliable performance of all the brands

Priceless.

 

[Jamie A]

Due to lack of other alternatives, they can charge a premium and still get customers.

There's quite a few alternatives to the Red Monstro.

 

[gabrielB]

Blackmagic seems to give the most reliable performance of all the brands

Priceless.

I think she meant of all the brands using the same Sony sensor. Not *ALL* brands including Arri and RED.

 

[John Brawley]

Bigger sensors, no matter what = less yield on the wafer.

That's true of any sensor. The larger the sensor, the more likely it is to have faults. The rejection rate can be astronomically high.

I'm certain there's other costs in terms of getting faster readouts on larger sensors and all that too. There's a rather large difference between a large sensor that rocks out at 2-5fps versus one that can do 60fps.

Yep, the clock speed of the whole system drives everything. The front end of the sensor (the rolling shutter time) through all the way to recoding a file. If you up the clock speed for a small impreovement in RS time it has massive flow on effects for diminishing returns in improvement.

and the entire wafer didn't pass QC and I don't think that was a unique wafer from the batch. That's a chunk of cash alone. Big and high quality sensors are really, really hard to make.

Yeah 80% or 90% rejection rate wouldn't be unusual for LF and bigger sized sensors.

It's certainly one of the reasons many companies will go the route of buying a pre-fab sensor instead of going through the horrors of rolling their own. It clearly sounds expensive and painful.

And all of the above issues also apply to more finished sensors. They're described as "off the shelf" but that is a total misnomer as you still need to do a lot of customising. Even developing your own sensor it's years of R&D. You try it out, you see how it performs and then make a few hardware adjustments at the sensor level for another spin of the sensor and it's months of waiting and more millions to roll the dice and see how much better you can make it. And that's before you even get to manufacturing it.

I was peripherally involved during the development of the Aaton Penelope / Delta and they were working with Dalsa. Dalsa are true pioneers and leaders in sensor design, they do a lot of work with NASA and a lot of the sensors are on remote probes and explorers.

https://www.teledynedalsa.com/en/products/imaging/custom-design/

Well, they had an initial batch of five sensor samples to work with and JPB was trying to get his head around the huge sensor to sensor variation. They ran out of money trying to figure out how to make a camera that could deliver consistent results when there were such huge variations in each individual sensor. And that's from a huge name in sensor design / fab.

Making cameras is hard because the way a camera makes pictures is ultimately subjective. This is the so called "colour science" but there's these kind of subjective decisions that get made at a hardware level that affect the "look" as well...

Panavision were working on their own camera for years as a successor to the Genesis. Part of that camera was a sensor that they designed. The camera never made it to being finished because they faced some insurmountable hurdles and got leap frogged. But they spun the R&D off they did on the senor into this.

http://shootdatapost.com/blog/2010/3/16/panavision-dynamax-35-ultra-video-sensor.html

I'm sure as some here already know, sensors are truly freaking hard to do. And it's a huge endless dump of money when you're doing your own design. Add in low yields for larger sensors and you can see how the costs will escalate rapidly.

It sent Aaton Broke.

Panavision gave up making their own camera.

JB

 

[Christoffer Glans]

Due to lack of other alternatives, they can charge a premium and still get customers. No amount of voicing opinions in this forum or elsewhere will matter as they won't simply care.

But Hydrogen failed to "revolutionize" mobiles and no one really buys this phone unless they are super fans of it, so that's a fail in premium. If Komodo doesn't meet what BlackMagic Pocket 6K delivers for its price, then that will be a fail as well.
Premium prices works when talking about cameras like Monstro, Helium, Ranger and DXL2, but if we're going to talk about cameras in the range of the Pocket 6K, then the competition is pretty harsh.

Komodo needs to be better for it's price. If ZCam delivers 8K full frame for the same amount and BlackMagic delivers a Pocket 6K at s35mm, then Komodo needs to have a special juice to compete with that. Doesn't matter what we say on this forum, it matters how it actually perform both in specs and practical application. If it fails, it's gonna die. The mid-range priced cameras like these live or die through what they deliver at their prices. If you can afford a Red DSMC2 at any range, with all the side-gear needed plus lenses, then you would probably not care about $10K differences and some small spec changes between cameras or brands because all would be enough for most kinds of projects. But at the Komodo, Pocket 6K and ZCam price range, the competition either kills or gets killed, there's no in-betweens as there is at higher ranges.

 

[Rakesh Malik]

That's true of any sensor. The larger the sensor, the more likely it is to have faults. The rejection rate can be astronomically high.

And since in a sensor intended or a high end camera, a bad pixel = a bad sensor, you don't get the same forgiveness that you do with something like a GPU or a modern processor. There's a reason that even though it's using a fab whose volume makes Intel's look small, AMD is still manufacturing the Zen2 lineup as multiple chiplets in a single chip carrier rather than as a monolith. And that's in spite of the fact that in a CPU chiplet you can get away with disabling cache and shifting it to a lower bin rather than chucking the entire die.

I'm sure as some here already know, sensors are truly freaking hard to do. And it's a huge endless dump of money when you're doing your own design. Add in low yields for larger sensors and you can see how the costs will escalate rapidly.

And never mind that a modern semiconductor fab involves equipment costing billions of dollars. Starting up a new fab is by itself enough to drive companies out of business. Having its own fab nearly killed AMD and Global Foundries has stopped trying to progress from 14nm to 10 (let alone 7) because the costs of all the equipment would drive it out of business.

It's crazy expensive... so there aren't that many companies doing it, and the most modern fabs are huge because that volume is what allows them to amortize their costs over both a lot of wafers and a lot of customers.

It's really not a surprise that Sony is so dominant in sensors... the technology it's developed in recent years like back-side illumination and stacked sensors are impressive, and possibly partly because of Sony's volumes.

And a lot of that volume is tiny cell phone sensors. It's also no surprise that Sony pilots its newest sensor technologies with those tiny cell phone sensors... the first back-side illuminated sensors, stacked sensors, etc...

 

[Alain M]

Imagine a Komodo with pins so you attach monitor (unmade new model 2020) RED Touch cable less (with batteries Lithium) attached on the back like ATOMso it does not drain KOMODO little batts and including hard drive....

it's like shrinking Helium setup giving us a flex smaller camera for projects not so up there, run and gun, among others yet maintaining the BEAUTY of RED Quality, elegance, Sleek Module approach and keeping RED's beautiful workflow and science under the hood.

It would be EPIC imo

I am too attached to RED workflow with my helium 8K and before I get a smaller alternative cam like Pocket 6K, I would love to see full spec on KOMODO and keep my cams to stay in the RED Family... Let's see when Jarred spills more frijoles. !!!

 

[Michael Tiemann]

RED is a business. And there is something called premium pricing. Due to lack of other alternatives, they can charge a premium and still get customers. No amount of voicing opinions in this forum or elsewhere will matter as they won't simply care. It is a question of competition and they have an opportunity to charge the premium. [...]

Um, this directly contradicts the founding principle of RED, which was to be a disruptive force in the industry and breaking the stranglehold of those companies who were too much using premium pricing in a way that discriminated against the little guy with big dreams.

I will not disagree that making high-quality cinema kit is not nearly as cheap at the end of the day as one might believe, and I will not disagree that certain cost-cutting decisions can fundamentally re-categorize a product's position.

It could be that I and other have been duped into believing that RED is all about disruption and independent filmmakers but that they are in fact merely exploiting our self-delusion by charging premium pricing and laughing all the way to the bank. But I don't think that's RED's strategic intent. I believe they are truly trying to be innovators, willing to bleed themselves as much as they bleed their customers in the pursuit of products that really do open new doors to an ever-expanding market of content creators.

 

[Phil Holland]

Dalsa are true pioneers and leaders in sensor design, they do a lot of work with NASA and a lot of the sensors are on remote probes and explorers.
Panavision were working on their own camera for years as a successor to the Genesis.
Panavision gave up making their own camera.

Fun and moderately off topic. Growing up in the valley and eventually living in Woodland Hills I used to visit PV and Dalsa when I was younger (a few blocks away from each other). Also the majority of the first decade of my career in features were mostly PV jobs, so dipping in and out was sort of a thing for a long time. Saw so much cool crap at Dalsa, a whole bunch of stuff I "shouldn't have seen" due to their clients. But it was a much different world back then and they didn't mind entertaining a young dude hungry to learn. Some of my earliest curiosities and learning experiences surrounding sensors came from the big D and the custom CCD based system we had at the older studio.

PV had a really interesting journey post Genesis. Outside of the defunk Dynamax stuff, what we loving refer to as the NASA White 70mm camera (other parties were involved as well) was a project plagued with major hurdles that in the end were to large to leap. Specifically power and heat related if I remember correctly and "other complications". Interestingly there were a few byproducts of their efforts. Firstly was the development of the rather pretty Primo 70 Prime lenses, which were actually designed for this camera. Though a lot of things were going on, that project died and seemingly the hopes of a future path for Panavision's own take on digital cinematography and not merely Panavising something. ARRI eventually more or less created a rather similar style (similar-ish to the targeted specs even) camera with the Alexa 65 with a slightly larger format sensor than the PV NASA camera would have been. However, they kept it in the realm of rental only and mainly through them. The growing relationship between Panavision and RED as well as Light Iron the led up to what we now know as the DXL and DXL2. Bringing back a whole lot of PV's intentions for a Panavision/RED co-developed camera that they could actually build, support, and have state of the art technology as well as globally supported workflow to boot. With the NASA cam they were more or less reinventing the wheel and truthfully there was no path there to be a "preferred leading tool" in this industry, with the DXL there was a lot of obvious groundwork not only done, but widely adopted. All of that in my mind was perfect timing from everybody I know at Panavision. Some of that crew is still as excited today if not more so about DXL and the future of the platform.

It's very interesting to see how the industry developed from there. This is obviously not the price point of Komodo or anything like that, but more just professional industry happenings. With ARRI Rental now expanding out to more major markets globally, which has been a moderately criticized move by rental houses and certainly owner/ops. RED's own efforts on "boutique" cameras like the previous iterations of Ranger and Fincher's Xenomorph, which after DXL led to the further evolved full blown Ranger for rental houses. Meanwhile we "still have" the more flexible modular solution in the form of DSMC2 cameras that use the same sensor tech, provide the same level of image quality.

That's all the past, but it helps me keep an eye of the future. Komodo is a bit of an interesting case as I believe where this project is now headed is overall a better path than what it was originally and something a bit more useful when you look at it with a wider lens. Some are pretty critical of the change of form factor, but that really was going to be a limiting concept in my mind when it comes to actually using the thing. That's something most professionals can see pretty clearly and some of that conversation seems to be muddled down as more interesting features are revealed. I see a tiny cinema camera, Soderberg likely sees the tool he's been looking so hard for.

 

[David Rasberry]

That's all the past, but it helps me keep an eye of the future. Komodo is a bit of an interesting case as I believe where this project is now headed is overall a better path than what it was originally and something a bit more useful when you look at it with a wider lens. Some are pretty critical of the change of form factor, but that really was going to be a limiting concept in my mind when it comes to actually using the thing. That's something most professionals can see pretty clearly and some of that conversation seems to be muddled down as more interesting features are revealed. I see a tiny cinema camera, Soderberg likely sees the tool he's been looking so hard for.

Yes! Komodo is a much more practical and appealing solution all around, however Hydrogen fits into the mix. Intelligent, even brilliant, design choices from what we know so far.
It is becoming possible to build much higher levels of processing into lower power and thermal envelopes.
I read a preview of Intel’s 10th gen mobile i7 processors built on 10nm fab. Quadcore hyperthreaded version with significantly better performance than my comparable 6th gen has a maximum power consumption of just 17 watts vs 95 watts.

 

[David Johnson]

I see a tiny cinema camera, Soderberg likely sees the tool he's been looking so hard for.

This is the camera that I've been looking for, too. Most of my production is in the outdoors & marine industry. I need light weight, rugged gear that can easily travel and be stored in small hatches. Image quality is critical. Affordability is also very high on the list. Even with the best preparation and safety measures in place, accidents happen. It's painful when gear takes a plunge. Less painful when the gear doesn't cost $25k+.

Even better when a camera shoot 6k, matches well with my other Reds, and I can get 2-3 of them... a fleet of "B" cams. I couldn't be more excited.

 

[Rakesh Malik]

This is the camera that I've been looking for, too. Most of my production is in the outdoors & marine industry. I need light weight, rugged gear that can easily travel and be stored in small hatches. Image quality is critical. Affordability is also very high on the list. Even with the best preparation and safety measures in place, accidents happen. It's painful when gear takes a plunge. Less painful when the gear doesn't cost $25k+.

Even better when a camera shoot 6k, matches well with my other Reds, and I can get 2-3 of them... a fleet of "B" cams. I couldn't be more excited.

I have a feeling that you're not alone in that... I'm eyeing Komodo both for b-cam (i.e. the gimbal cam on narratives) and a mountaineering camera...