I think that they're (more or less) unrelated. The cameras probably have the legacy OS from the VCR's and TV sets - enough to power the menu but without much computational power per se, with the signal path having a separate system/codecs, processors, etc. Something with a Linux kernel would probably be much easier to crack/mod by the third party like the Magic Lantern. And it kind of makes sense too. A phone OS and the related CPU would be a surplus to requirements.
Sony A7SIII = $3500, 10-bit 4K120p internal, 16-bit 4K60p RAW out
- Thread starter ahalpert
- Start date
I think that they're (more or less) unrelated. The cameras probably have the legacy OS from the VCR's and TV sets - enough to power the menu but without much computational power per se, with the signal path having a separate system/codecs, processors, etc. Something with a Linux kernel would probably be much easier to crack/mod by the third party like the Magic Lantern. And it kind of makes sense too. A phone OS and the related CPU would be a surplus to requirements.
Samuel H
Major Contributor
The OS in Sony cameras works exactly the other way around: there's a high-performance core that's very narrowly tailored to the hardware, and on top of that a high-level layer that runs android stuff for things like showing the menus and programming timelapse shots and things like that (and also probably the wifi and USB and all those other parts that are not "reading the sensor and writing the files")
joema
Active member
I think you can continue using UHS-II cards in the A7SIII, and *each* of two slots will hold either a UHS-II SD card, OR a CFExpress Type A. The type A when available is spec'd at 1,000 MB/sec. That is roughly 3x to 4x faster then the best UHS-II cards.
We don't know what internal codecs the A7SIII will use, but it seems likely it would use XAVC-L for 10-bit 4:2:2 material. That can easily be handled by UHS-II. If it also offers 10-bit 4:2:2 XAVC-I, even that is only about 300 mbps (37.5 megabytes/sec) at DCI 4k/29.97.
Even if it offered internal 10-bit 4k/60 ProRes 422, that is only about 125 MB/sec, in theory UHS-II would handle the bandwidth, but storage size would be small. If CFe-A offers larger sizes that would be helpful.
I don't think Sony can offer internal compressed RAW due to the RED patent. The rumors have mentioned external RAW via HDMI but that eliminates the internal storage issue, since you'd be storing it on an Atomos or similar, probably using AtomX SSDs: https://www.amazon.com/dp/B07JWD1ZYF
sleekmedia
Well-known member
The RED Komodo indeed has phase detect AF. There are examples of it in use in a few easy tracking scenarios online (see below) by Philip Grossman.
https://www.youtube.com/watch?v=mtw1zu4xkzA
https://www.youtube.com/watch?v=5MVrEYSZOGM
It's working quite well in my opinion for RED's first shot at AF, but its definitely nowhere near the accuracy of existing Canon video cameras which are impeccably good at their ability to track faces in difficult scenarios such as when the camera is moving on a gimbal and producing images without focus jumping around and going "in-and-out" that makes it apparent you are on DSLR glass. Jarred Land on Reduser Forum did proclaim that Komodo's currently are running beta software and continuous phase detect AF and face tracking will be updated and further implemented in the primary release of the black colored Komodo cameras to the masses and by software update to the existing beta users. At the rate RED works in the past, I wouldn't expect a proper update with full functionality any time soon as they are currently having trouble even getting the cameras in beta form (custom colors) ready to be shipped to the already long and growing list of custom color orders. I think it is safe to assume we won't see standard black regular order Komodo's shipping till September or October at the earliest and that doesn't count in any implications or hold ups from Covid-19. Similarly, I wouldn't expect a proper software update with full AF implementation until that time or at least another 3-6-months from the time the custom color Komodo's start rolling out and it's not guaranteed to be perfect even then. This also does not take into account any last minute changes by RED to improve the camera in response to the R5, A7SIII, or URSA Mini Pro 12k. Only time will tell.....
A few other things too consider about Komodo versus its competition coming from an existing RED owner who has a decent amount of knowledge on the Komodo / RED platform and has an interest in purchasing one:
1) Data Rates - It has not been confirmed whether the Komodo shoots the "exact" same type of REDcode Raw as an existing DSMC2 camera (i.e. Monstro, Helium, Gemini, Dragon, etc.). This is very important for a few reasons. If the Komodo utilizes a new "DCT" version of REDcode Raw which is not as efficient as the "Wavelet" based Format it may end up giving you less compression for the same quality of footage compared to a DSMC2 camera. Therefore in order to shoot high quality (i.e. low compression at 2:1 - 4:1 etc.) it will end up giving you double the data rates of an already data hungry DSCM2 camera. At 5k full height / 24fps (2:1) compression on my Gemini I can record 26 minutes of footage and at (4:1) compression I can get a nice boost up to 52 minutes on a 480gb mini mag which is quite good in my opinion in terms of recording time VS. image quality. So far it has been confirmed by Phil Holland on Reduser Forum that Komodo records 1hr of 6k / 24fps at "high quality" compression (rate has not been confirmed) to a 1TB CFast card ($1000). It should also be noted that Komodo currently does not have the same options for compression as a DSMC2 camera (2:1, 3:1, 4:1, 5:1, 6:1, 8:1, 16:1, etc.) and instead has only 3 options (HQ, MQ, LQ) which RED said was supposed to simplify compression for unfamiliar users but to most existing RED users this was alarming as it takes more options out of the hands of the user and forces you to use a compression rate that may not always be perfect for your shooting scenario. According to what is known about Komodo, a 512gb ($500) CFast card would give you approximately 30 minutes of recording time at 6k / 24fps which isn't a lot for say an interview and definitely doesn't take into consideration of high frame rate b-roll and still we don't even know if this is 2:1, 4:1, or 6:1 compression. Seems like you may be filling up CFast cards pretty fast which defeats the purpose of cheaper media options for Komodo compared to a DSCM2 camera and mini mags (if they are higher priced but more efficient in terms of compression). Komodo is already double the price of the R5 and A7siii (assumed) so when you take into account the additional price of (4 x 512gb or 2 x 1tb) CFast cards you can add another $1500-2000 to the purchase price not including monitoring, batteries, rigging, etc. Not to mention the additional cost in long term media storage on HDD's and short term SSD volumes for editing. Considering my Gemini already pumps out some pretty heavy files this should be a concern for anyone looking into Komodo. Obviously with REDcode and 16-bit raw vs 10-bit codecs from the likes of Sony or Canon, more data in most cases means higher quality image but there is a distinct line that needs to be assessed as to whether heavier files gives you their full worth and weight in IQ in comparison to the costs of managing them.
2) High Frame Rates - Anyone interested in Komodo should be aware it only shoots 2k / 120fps with severe crop factor. A 6k sensor windowed down to 2k will produce much higher noise and will make wide angle lenses become more like medium telephoto in terms of angle of view (AOV). It would be safe to assume the Komodo sensor is windowed down from slightly larger than S35 when shooting 6k to somewhere around S16 size when shooting 2k. This would produce somewhere around a 3X crop factor in terms of AOV on lenses. So a 21mm may appear to have the same AOV as a 63mm. http://phfx.com/tools/formatCompare...odelB=Komodo+6K&formatB=2K+WS&focalLengths=21 So if you are trying to shoot HFR with Komodo on a wide angle lens expect it to come out looking like at least double or triple its focal length when shooting 2k on Komodo. This creates issues in lots of scenarios on set where you may need HFR but also a wide shot. I am going to guess that both the R5 and A7Siii also have a sensor that windows when shooting 4k HFR but how much crop is there is up for debate and should be assessed by someone that has access to the cameras. My Gemini shoots 5k WS @ 120fps with practically no crop (obviously at a higher cost than Komodo) and the URSA Mini Pro 12k has a symmetrical sensor that allows the use of various resolution choices without cropping so this method of utilizing the entire sensor area for multiple resolution recording formats is the future and cropping to shoot lower resolution HFR is about to become a thing of the past. If you publish and deliver in 4k like I typically do then 2k/120 fps from the Komodo is not going to cut it unless you want to deliver 2k or 1080 files. With 12k and 8k cameras becoming the "mainstream" I think its easy to see 4k delivery becoming the new standard which is evident in the fact that Arri has been bumping up Alexa resolution beyond 4k. So you have to ask yourself is 2k/120fps with severe crop going to even be usable (outside of great lighting scenarios) and will it last the test of time when there are so many other camera companies putting out 8k capture with greater than or equal to 4k HFR (i.e. Canon, Z-Cam, Black Magic, Kinefinity, etc.) for similar or even cheaper prices. I mean even the Black Magic Pocket 6k did 6k/60fps and 2k/120fps over a year ago and that camera can be had for under $2000 today which is less than 1/3 the price of Komodo built, which only does 6k/40fps and 2k/120fps respectively. Not to mention with the release of the UMP12K I would be willing to bet a new 8k pocket cinema camera release isn't far from upon us. This is definitely something to consider if HFR is a staple in your workflow and you plan on delivering 4k files to clients.
3) Bit Depth - Something definitely to consider about Komodo as a benefit over most of its competition is its codec and 16-bit color depth. C-log from Canon is a good codec which is also heavy in data rates (like any high quality Raw) but the R5's recording of 8k and 4k / 120fps appears to be limited to 10-bit depth even when recording to external recorders. With the Panasonic S1H allowing 12-bit 5.9k external recording and the BMP6k allowing 12-bit recording to SSD's I am unsure why Canon is limiting the R5's ability to record higher bit depth to external recorders. Similarly, I think we will see ProRes Raw as a flavor with the A7Siii and external recorders but I am willing to bet bit depth will also be limited to 10-bit. 10-bit is perfect for quick turn around time doc work, weddings, vlogging, interviews, real-estate, travel videos, etc. and in controlled lighting scenarios 10-bit won't look much different than 12-bit or 16-bit but in more challenging lighting you will really see a difference in terms of IQ. 10-bit sounds great but it is not really acceptable to professional colorists on jobs where IQ is at the forefront of importance and coming from a 16-bit camera I can tell you this does matter and gives you so much more flexibility in grading without the image breaking up. This is one area where I see Komodo as a big win over the R5 or A7Siii in terms of overall IQ. Not to mention Komodo and Panasonic S1H are approved cameras for Netflix broadcast if that tells you anything about their image quality where as R5 and A7siii if only 10-bit will not make that list. I believe Canon and Sony limit their prosumer DSLR sized mirrorless video cameras (i.e. A7siii & R5) to 10-bit in order to keep the value of their professional line of cinema cameras high (i.e. FX-9 and C300ii) which can shoot 12-16 bit flavors. Where as RED's philosophy is to never have a compromise when it comes to IQ. So instead of limiting bit-depth and overall IQ on Komodo to reach lower-price points, they instead lowered its ability to record HFR instead as a compromise in comparison to their DSMC2 line of cameras.
4) Price - On first glance Komodo appears to be an excellent option as it's supposed to be available in black at $6000.00. Like any camera, this is before you take in to account the necessary gear to make the camera function. With Komodo you will at least need (2 x 512gb CFast with card reader), (4 x Canon BP-955 batteries), (1 x Dual Battery Charger), Canon RF to EF adaptor (unless you already have RF glass), Top Handle / Cage and or Rails for Rigging, and a monitoring Solution (i.e. Atomos or Small HD). This puts Komodo in an $8000.00 minimum price range and if you really wan't to build Komodo out as a more capable rig with a more reliable battery solution (i.e. v-mount or gold-mount) that has D-tap ports for powering wireless follow focus and video transmission similarly to a DSCM2 camera, I think it's safe to say we will see builds approaching the $10,000 price range by the time they are all said and done. That is not far off of the soon to be released Kinefinity Mavo Edge which is full frame 8k 16-bit and shoots 8k/75fps, 6k/100fps, and 4k/160fps with beautiful color science as well. Simply put there are a lot of awesome cameras in the $10,000 price range and when you really consider how much Komodo is after its built up I am not sure it is better than some of its competitors or even the option of picking up a used Scarlet-W or Dragon-X which offers higher frame rate recording and better (no cables) modularity. The whole point of a tiny camera is to keep it tiny and a lot of the builds I have been seeing of Komodo thus far make it as large and much more clunky than a seamless modular DSMC2 build with cables, non-proprietary monitors hanging off arms or ball mounts, and v-mount battery adapter plates on rails, etc. which are very reminiscent of the BMP6k or even bigger Alexa Mini builds with v-mounts you see on rails. Remember that the screen on Komodo is really for adjusting menu settings and is not really designed to be used for monitoring despite the fact I think many people will attempt to use it this way, not to mention the fact that you can't shoot eye level with a camera that has a screen flat mounted on top of the body as you won't even be able to view it. So don't think you won't be adding at the very least a cage/top plate/top handle, monitor mount, and small monitor with batteries to Komodo. Not to mention if you decide to power a monitor and FF off the camera you absolute will need to convert Komodo to v-mount or gold mount batteries with an adaptor plate. This is where I really see the R5 and a7siii shine as they are under $4000 and work right out out of the box and are no larger than a typical DSLR for photography and you still have the option of adding additional media, batteries, cage, and monitor in a package that still comes in under $6000.00 built. The only question to remain is are there any benefits to external recording with these cameras at this time and if not the barebones "easy to travel" DSLR form factor sounds like a win for internal recording. Also, don't forget about the fact that the R5 takes 45mp stills and will be one of the worlds first cameras that does high resolution and high frame rate video recording while simultaneously being great as a high-res option for commercial photography, which is something I have wanted out of a DSLR I can throw in my bag for a long long time.
As an existing Gemini user I don't see Komodo as a true A-cam that could replace my Gemini in the working professional world. With it's lack of modularity and more reliable battery solutions without add-ons or expansion modules for running accessories, you really have to sacrifice spending more money, form factor, and weight to build it out in a way that it quickly becomes bulkier than a clean-lined and truly modular DSCM2 camera which gives you dual iso (Gemini), much better HFR (Gemini or Dragon-X) or 8k resolution options (Helium). I personally see Komodo as more of a B-cam to existing RED camera users and as an A-cam to individuals who cannot afford higher priced cinema cameras however many should take a look at all the options out there in this price range. It was my understanding that Komodo was originally designed to be a tinier than DSCM2 option to record REDcode in demanding situations as a B-Cam or C-cam on set (i.e. car mounts, gimbal, second camera for interviews, etc.) which would allow for better workflow as you are not constantly re-rigging your A-Cam for each scenario. But due to internet popularity over its name, cool looks, tiny size, and matchbox car color options, it has become an A-cam in many minds to the flooded market of Indies who always wanted a RED but couldn't afford one until now. Many are not taking in to consideration what it will take to get this camera to function the same as a true A-cam for recording audio, ports for wireless follow focus and wireless video transmission for directors monitors, etc. as there are many add-ons to make this level of functionality happen which directly impact its size, weight, cost, and form-factor and ultimately what the camera was originally designed for when you compare it to its bigger brothers. For me the camera was supposed to be the opposite of my DSMC2 camera in which I wanted to keep it as small and light as possible in which it could live on my gimbal or be packed up in my carry-on for inconspicuous shooting and easier travel to jobs where a fully rigged A-cam isn't required. It seems in looking at all the accessories that are coming out for Komodo that they are going to opposite direction in building it up to the likes of a fully rigged Alexa mini which isn't the greatest form factor for travel or quick changes on set if you have ever used one.
There is an awful lot to consider with so many options available today but you have to ask yourself "how do I intend to use this camera" and really be honest with yourself in assessing your own needs versus how much you gain from each camera platform and their inherent price. If you are trying to shoot a cinematic feature film and are on a budget, Komodo might be the camera that gives you the high quality IQ and 16-bit image of RED but may be in your price range. If you are looking to replace a BMP6k I am not sure Komodo is going to offer you much more than you already have other than keeping up with the Jones's. If you would be coming from FS7 and typically shoot corporate ad's and commercials I think that may be a nice transition if you don't like the FX-9 or C300ii and want something smaller that can still be built out into a full ENG styled rig. If you are looking to replace your A7III or Canon 5DIV for your vlogs, travel videos, creative content or weddings it's just me but I probably wouldn't jump as high for Komodo and I think the R5 and A7sIII would be awesome for your needs and honestly a lot more practical in terms of size and travel. My point is don't buy a RED just because you think it will make your images better or will get you into a realm of higher priced clients and bigger jobs because it will not. How you use it in combination with your overall production value and knowledge of cinematic lighting is much more meaningful. With so many camera companies releasing new technology and updates every 6-months now is an amazing time to buy new or even used gear but it brings a lot more questions when comparing options and assessing their long term viability of a camera platform within the industry. Overall, I just wanted to share some thoughts to think about when assessing the true value of Komodo vs similar competitors as these are some of the questions I came across while I determine whether I still want to pick one up.
The biggest things to consider are....
- Do you need super-reliable and advanced autofocus? (R5 / A7SIII) or do you typically shoot on cinema lenses (Komodo)
- Do you need 16-bit IQ (Komodo) or is 10-bit (R5 / A7SIII) fine for your line of work and client demands?
- What kind of codec / data rates can your media storage and laptop handle? Have you ever tried editing 16-bit REDcode on your current editing laptop and can it handle it?
- What resolution and crop factor is acceptable for HFR? 2k/3X crop (Komodo) or 4k/1.5 crop (R5 / A7sIII)
- What is your total budget? $4,000 (R5 / A7sIII bare bones), $6,000 (R5 / A7siii with recorder and cage), or rigged out Komodo ($8,000-$10,000)
- What is your limitation on camera size? Fit in a glove-box or small pouch or around your neck? (R5 / A7sIII)
- Do you mind having to assemble your camera with monitors, handles, etc. each time you get it out? (Komodo)
- Do you intend to also use the camera for Photography? (R5 / A7sIII)
- Are you already highly invested in E-mount Glass (A7sIII)?
- Do you intend to shoot anamorphic? (Komodo)
Last edited:
ahalpert
Major Contributor
Yes, I know this. I'm unhappy if it uses Type A cards because I expect them to be an overpriced bad investment. No one is selling them right now. So you'll have to buy them from whoever sells them first (Sony). And they'll be overpriced because seller competition brings down card prices. And they'll be a bad investment because other cameras are already not using them in favor of a faster card. Type B cards will continue to dominate and your Type A cards will be useless for most other cameras.
I like the fact that it has matching card slots (not 2 different slots) and that I can still use my SD cards. Perhaps that is why they needed a card that could fit the SD slot so neatly. I don't like having to buy Type A cards for whatever the biggest bitrate recording options are (4K120p XAVC-I?)
puredrifting
Major Contributor
Amazing that it's 2020 and we are still dealing with incompatible, proprietary media throughout the industry. For while it seemed like SxS, P2 and the like were going away
and we were settling in on SD cards or CFast. But now with CFast 2.0 dead (amazing BMD still put it into the 12k camera!), SD cards are pretty much reaching their write limit with
6k, 8k and 12k recording and now we have to deal with two incompatible CFExpress formats? The recurring theme in camera progress is one step forward, two or three steps back it seems?
NorBro
Major Contributor
No crop/window with R5's 4K/120p in 8K RAW and 8K/4K DCI modes. Full-width of sensor.
10-bit ProRes 422HQ, XF-AVC, XAVC-I and AVC-Intra 4K are all approved.
You'll see the R5 on there within a month.
NorBro
Major Contributor
No XF-AVC on it. But their list will eventually expand to include HEVC (and others) as modern times continue to march forward.
In short, if the Canon C500 and C300 Mark II are on there, the R5 should DEFINITELY be on there. (And this assertion is just from working with a few random HEVC and CRM files, and also knowing how the aforementioned perform.)
Now if many would use it (at that level) if overheating is truly extreme is another story. Definitely should be on gimbals and drones.
In short, if the Canon C500 and C300 Mark II are on there, the R5 should DEFINITELY be on there. (And this assertion is just from working with a few random HEVC and CRM files, and also knowing how the aforementioned perform.)
Now if many would use it (at that level) if overheating is truly extreme is another story. Definitely should be on gimbals and drones.
I'm not sure how this would be possible, since full sized SD cards (32.0×24.0×2.1 mm) and CFexpress Type A cards (20.0 x 28 x 2.8mm) have different dimensions.
CFexpress Type B cards have the same physical dimensions as XQD cards, which is likely part of the reason why that particular size (of the 3 sizes available) has gotten the most attention:
https://www.compactflash.org/cfexpress
There is an SD Express card standard which has the same physical dimensions as traditional SD cards, but uses a PCIe based NVMe interface for bus speeds of up to 985 MB/s. The problem is that the SD Express cards use the same pins for this new interface as were previously used for UHS-II (this is the second row of pins on the back of the card). Which means that if UHS-II cards are used in a SD Express slot, they will only operate at UHS-I speeds of up to 104 MB/s (using the top row of UHS-I pins) rather than at UHS-II speeds of up to 312 MB/s.
There's some more background on CFexpress vs. SD Express here:
https://progradedigital.com/sd-express-a-comparison-to-cfexpress/
CFexpress Type B cards are currently available from at least some manufacturers, but I don't think anyone is really selling SD Express cards yet, and for that matter there is actually a UHS-III spec (624 MB/s bus speed) that no one seems to be using yet either (even though it was announced in 2017). Nonetheless, the SD Card Association has plowed ahead and announced an updated SD Express 8.0 spec that uses a PCIe Gen 4 interface (and 3rd row of pins) to boost bus speeds to 3840 MB/s:
https://www.newsshooter.com/2020/05/25/sd-express-delivers-new-gigabyte-speeds-for-sd-memory-cards/
Samuel H
Major Contributor
The cards are completely different in size and shape but they found a way to make a combo slot. It's pretty clever actually. See the graph here.
https://www.sonyalpharumors.com/rea...kely-going-to-be-used-by-the-new-sony-a7siii/
ahalpert
Major Contributor
That seems clever. Ok if the slot works well and they had to use type-a cards so they could have 2 matching slots, I will accept the fact that it uses type-a and not type-b. And I will keep using sd cards until type-a cards come down in price
joema
Active member
The only info we have so far is a rumor that the A7SIII will have two "combo" slots, each of which will accept an CF Express Type A or an SD-size card, possibly including UHS-II. There is also a speculative slot diagram: https://www.sonyalpharumors.com/rea...kely-going-to-be-used-by-the-new-sony-a7siii/
The CFe Type A card uses end connectors and the SD UHS card uses side connectors, so in theory a single slot which accommodates both from a connector standpoint might be possible.
The Type A card is narrower and thicker, and the SD card is edge-guided. It should be physically possible to design an SD slot with a recessed aperture in the side which accepts and guides the CFe-A card. In that scheme the CFe-A card might not have full circumferential guides and would be partially stabilized by the end connector. How mechanically secure that would be, I don't know.
If that is what Sony actually does, it would bypass the backward compatibility issue of CFe-A (none) and the lower data rate of SD Express in compatibility mode. It would be exploiting the mechanical and connector differences between CFe-A and SD to use two different incompatible card types in a single physical "combo" slot.
sleekmedia
Well-known member
Thanks for the update. I had no idea 10-bit ProRes 422 or XF-AVC and XAVC-I was approved for Netflix. I didn't know they approved "codecs" rather I always thought it was a list of camera bodies?
https://partnerhelp.netflixstudios.com/hc/en-us/articles/360000579527-Cameras-and-Image-Capture
Honestly, I don't think the image that ever came out of my A7rii with my atomos ninja in 10-bit 422 was any good and definitely not good enough for mainstream broadcast or Netflix for that matter. Still much better than recording 8-bit XAVC internal but I never felt like the images had enough latitude and would get grainy and looked very digital when pushing heavy grades. I am guessing that is due to the design of A7rii sensor and probably not the codec. Similarly, I always felt that my old FS5 (before I sold it) even when running 12-bit raw out to my shogun Inferno to capture ProRes Raw always left me wanting more compared to the image of a friends Scarlet-W and even an FS7 I would occasionally use at the time. Again, this is most likely due to the sensor, probably not the codec but that is why I am pretty sure the FS5 or A7rii despite being able to shoot greater than 10-bit 4k, never appeared on the Netflix list even though the FS7 did. Both of those cameras shot 10-bit 422 or 12-bit ProRes Raw but were not approved, which I would have had to agree with Netflix's decision as it was quite difficult to get super immaculate images with either cameras consistently. External recording and the ProRes Raw codec was quite new at the time and I don't believe it was until the Panasonic S1H that we saw a camera that can shoot ProRes Raw to an external recorder get added to the Netflix list. I agree Netflix's list is nothing to tout or make a camera purchase on but its something that is very interesting to watch as new camera models come out to see what makes the cut and what does not and to assess "why" a certain body or codec might fall short in that realm.
If you look at the list today you will see the Panasonic S1H is still the only "DSLR styled" video camera to ever make Netflix's list and it's actually quite ironic that the Black Magic Pocket 6k is not on there despite its ability to shoot 6k 10-bit and 12-bit Braw and Pro-Res directly to SSD's and has very good color science. This still leads me to believe that Netflix bases their decision off of multiple factors as well as industry and company feedback and not just solely on bit-depth and codec. In my experience I have seen lots of lower end Canon Cameras (i.e. C200) push out footage I think looks as thick and creamy as my RED footage today (friend shoots on c200) and I have watched countless docs on Netflix shot on C300 that really were great. It's funny that the C200 despite being very similar to the C300 (especially when getting the most out of the camera by using an external recorder) is not on the approved Netflix list. Personally, I think factors that go beyond image quality such as industry feedback, etc. are all taken into account as well as the companies right to preserve the retail value of their higher end cinema line of cameras by not allowing their lower price point prosumer cameras to be utilized on big productions at least not in front of mainstream eyes. Like it or not, I doubt the R5 or A7SIII will ever make Netflix's list as it just doesn't make sense for the companies to cannibalize their FX-9 and C300iii in the eyes of prosumers and professionals. The R5 and A7siii are or will really be no different in terms of bit-depth, codec, and overall IQ than what a BMP6k did 2-years ago despite the addition of increased frame rates / resolution and that camera did not make the list. I really do hope the R5 or A7siii make the list, but I just don't see it happening no matter how good they are as too many people would look towards those platforms instead of the companies higher price-point cinema cameras.
I think it is really great that the R5 doesn't crop or window it's sensor to shoot 4k/120 fps as I did not know that capability was confirmed. That is a huge deal to me as it has been a limitation of HFR shooting on many cameras (including RED) for the last few years. Personally, at this time I think Komodo is off the table considering it will come out somewhere close to $8,000-$10,000 "built-out" and it will offer worse frame rates than BMP6k that came out a year ago for less than a 1/3rd of the price and it will have severe crop when shooting HFR. As much as I don't like the BMP6K form factor you could purchase 3 of those cameras or 2 x R5's for the price of one Komodo. RED image quality is great but I don't think its worth 2-3 times the price of an R5 or a7siii (which both can double as a really great photography camera) if these cameras push out footage reminiscent of their bigger brothers (FX-9 / C300ii). I think at this point in time, the price of Komodo I can just not justify considering I already have spent lots of money on my DSMC2 Gemini (which depreciates a lot) and I am fairly certain Komodo (non-upgradeable like a DSCM camera) will be outdated by DSMC3, which will come out sooner than later in response to the UMP12K within 6-months to a years time. Komodo is a nice way for RED to capitalize on their "existing" technology by providing a price-point camera to Indies who have always dreamed of shooting "RED" while marketing the tinier, cool-colored, cube body design, which realistically Kinefinity and Z-Cam came up with a few years ago. Z-Cam has been putting out cameras exactly like Komodo but with higher specs and lower cost for a while now, but again for some reason the industry just does not take to unfamiliar smaller brands like Kinefinity and Z-Cam despite their insanely affordable powerhouse specs.
Who knows what the R5 or A7sIII cameras will be able to do with external recorders as well and maybe we will see 12-bit Raw to a flavor of Pro-Res. If the R5 has image quality comparable to c300iii I will probably be looking to pick one up for sure but I am definitely going to wait to see what the A7siii brings to the table.
Last edited:
joema
Active member
No current Sony Alpha mirrorless camera can record 10-bit material via HDMI to Atomos. It is 8-bit 422. My doc team has used those a lot in low-light situations, captured as ProRes 422 on an Atomos, and it works OK but with a flat profile it's very easy to encounter banding in post. I would never use Slog3 on an 8-bit camera, and even the less-flat Slog2 can be problematic. I can understand a higher-end media company like NetFlix or BBC not accepting that for an A camera. The dilemma is they typically don't want a baked-in color profile, so request flat/log, but for an 8-bit camera this constrains grading in post.
The FS5 12-bit ProRes RAW looks pretty good, but up until very recently required FCPX, so could understand not accepting it for that reason. Today ProRes RAW is supported in shipping versions of Premiere Pro, and supposedly will be supported in Avid.
When I've tested 12-bit ProRes RAW in FCPX, it looks and feels good, but when I use HSL color masks I don't seem to get the same hue and sat resolution as luma. I don't understand that because with 4444 material I thought spatial resolution was the same as luma. There are downloadable test clips here: https://filmplusgear.com/prores-raw-testfiles-for-download-nab-2018/
Re Blackmagic PCC4k and 6k not on the NetFlix list, that is interesting. My team uses those plus Ursas, and the image is good but IMO the latest generations of mirrorless cameras like S1H (and maybe R5 and A7SIII) can match those in dynamic range yet provide IBIS stabilization and vastly better low light. Before these latest mirrorless cameras our approach was use an Ursa or similar for interviews and narrative material and a mirrorless camera for low-light or mobile shooting. However in theory the newest mirrorless camera might be able to do both, and we'll be inspecting that closely. They don't have internal ND which is a known issue.
sleekmedia
Well-known member
Great info here... I never knew that the A7 series couldn't output 10-bit for recording. I swear it was marketed that way a few years ago when Atomos became popular. So in reality there isn't much gain from an external recorder for the A7 series. I agree about the BMP6K being very good. I have seen excellent grades / IQ come out of those cameras. The more I read about the R5 the more I hear about it's overheating issue which is now a concern. Will be interesting to hear what A7siii will bring to the table.