herein2020

I think all camera makers should offer more cameras that can record to a USB-C device, on the other side of the USB-C interface you could have options for nearly unlimited cheap storage and USB-C is fast enough to keep up with the bus speed requirements for even the most demanding video bitrates. That would also solve the encryption wishlist by letting users use encrypted USB-C options and would keep camera makers out of the encryption business not to mention it would appeal to far more users than encryption alone.

Am I the only one who misses the electronic level found in most MILCs and DSLRs? I don't understand why it is missing from the C200 and I would assume it is also missing from the C70. I religiously use the level in my GH5 to double check my gimbal horizon and when setting up on a tripod especially since I offer real estate video tours where a level horizon is critical. It's so annoying to pay $5500+ for a body and not get this basic functionality.

I would just follow their guide below, The bitrate is the most important thing. If you go over their recommended bitrate for a particular resolution they will further compress the video until it matches their bitrate. Bitrate typically won't really make your footage sharper though, it will just improve areas like gradients, fast movement, etc and reduce compression artifacts. https://support.google.com/youtube/answer/1722171?hl=en

Wow, that's terrible to hear, I do hope things pick up for you and everyone else. I am fortunate to still have a full time regular job and I offer such a wide variety of photography and video services that weird quirky jobs keep showing up; everything from shooting a documentary on men's health to creating architectural images for a new airplane hangar. I use my Ronin S nearly daily and honestly AF isn't as important on a gimbal as most people think. First off I know I'm not shooting a Hollywood blockbuster so I'm rarely shooting wide open, I try to stay around F4 or higher on the gimbal and I keep a certain distance between myself and the subject; in post I use a 1080P timeline after shooting 4K which lets me simulate changing distance between the camera and the subject without actually doing so. I've been shooting with the GH5 for years and have never used the AF. I use all manual Voigtlander lenses. So yea, for me AF isn't that big of a deal even on a gimbal because I've never had it, but as you stated, everyone's needs are different. Even when the GH5 is on a static tripod I just shot as stopped down as possible, focus it where I think the action will take place, and hope for the best. It's been my experience shooting events that as long as the audio is good, people can see themselves in the footage, and it looks more cinematic than a cell phone the customer is still happy. For corporate commercial work of course its a whole different story.

I asked the CVP reviewer and he said that EF-S glass will not work on the C70 without severe vignetting and that it does not have an EF-S mode. I do think maybe you are trying to get more out of the camera than is possible. Definitely starting with high quality 4K will be a better starting point; and of course I am sure you already tried adding sharpening in post. I typically keep sharpness turned all the way down in my drones and the GH5 then add it back in post if needed, so the footage out of my gear starts out soft. Upscaling really can only do so much, and I believe Kye has done extensive testing and reached the conclusion that the most you can upscale is 20% before it becomes noticeable which is one of the reasons I stick to 2K vs upscaling all the way to 4K. Exporting to Prores won't improve anything, there's only so much you can do after starting with compressed LongGOP footage, its kind of like JPG, what you see is what you get. You can't take a JPG and convert it to a RAW image and suddenly get all the benefits of RAW....the information simply isn't there.

I recommend first off that you don't buy anything When these cameras come out and are more readily available rent one or a few of them and make sure they meet your expectations before buying them. Second I recommend you try Davinci Resolve, if you are not doing this for paid work even the paid version of DR will be cheaper than PP within a year. Also, I believe the free version does offer upscaling. I upscale my 1080P timelines to 1440P simply by picking the resolution during the Deliver phase. I do this because YT lets you use higher bitrates if you go over 1080P. Have you looked at your footage on a monitor or TV before uploading to YT? Did it look fine there? A little know fact about YT is that more popular channels get less compression (or so I've heard) so it could just be that YT is not trashing some of that top quality footage that you see on YT as much as it is trashing your footage. Last but not least, are you sure the footage is mushy because it is 1080P and not something else? Did you go through Sigma's micro focus adjustment process to ensure the lens is properly focusing? You also have to keep in mind that not all Super-35 cameras will work with all APS-C lenses or even with any of them. The C70 for example will only work with FF glass or so I've been told. I have the C200 and to get it to work with APS-C lenses I had to enable APS-C mode. I have heard that the C70 doesn't offer this mode at all.

I think that's where we are disagreeing, I'll admit, I don't know anything about how much memory cameras have but I would imagine it would not be enough to store the entire file plus the encryption prior to writing to the memory card, especially when it comes to video. I just don't see a 100GB video file sitting in memory encrypted in its entirety before being written to the memory card. So in that case it would have to be an inline HW encryptor which goes back to the previous problem; how to verify the integrity of the file during the write operation if the camera cannot read it as it is writing it. Either that or cut the video into small encrypted chunks in real time based on the available memory which would be another problem entirely when it comes to trying to put the video back together (if its even possible). Even for images, I would imagine the speed hovering somewhere between unacceptable and unusable. What is interesting is the link to Magic Lantern that the OP provided; they tried this and ran into the exact problems I am describing; they didn't try it on video (probably due to the memory limitations) and the users were reporting a lot of corrupt files even for images; and that was using encryption that can be plaintext hacked within an hr. The only way this would work and be reliable IMO would be to either offload the encryption to the storage device like the USB-C device with physical buttons, or the camera makers would need to add a HW encryption module with the private key and keep the volume unlocked during operations. I can see the SW Public Key only solution working for a few images at a time under ideal conditions and maybe for JPGs, but not with any degree of reliability in many real world conditions.

Its great hearing from actual working professionals with this camera and not just another YouTube shil. I have the C70 on pre-order but I also already have buyers remorse so I will probably cancel. The S5 keeps coming back to my attention but your problems with AF remind me that using this camera on a gimbal would be no better than the GH5 that I already own....with the added problem of not being able to use any of my MFT lenses on it. I really think I just need to stick to my original plan which was to give up on replacing the GH5 for another year.

Yes, it looks like they came to the same conclusion that I did....its not feasible for video. Also their implementation of encryption for the sake of speed is very weak. The 64 LFSR encryption method can be reversed in less than an hour with a fast enough modern computer, even they admitted it wasn't very good and had to be downgraded from the typical 256bit standard due to IO speeds and that was in 2014. Your comparison of 5200RPM drives is mixing apples and oranges, the drive read/write speed is not where you will see the most penalty, typically the encryption/decryption is done in the CPU which is where you could get up to a 30% hit when reading/writing to the encrypted drives. Modern CPUs tend to have way more power than most users need so you wouldn't notice the hit to the CPU unless you did a before and after benchmark and modern CPUs are designed to perform cryptographic functions efficiently. The problem of course as I mentioned earlier, is that cameras do not have all of these spare CPU cycles laying around or onboard TPM chips. In the Magic Lantern example the camera's CPU would be needed to encrypt every byte of data in real time which is where the performance penalty and unreliability would kick in. A camera cannot afford to lose 30% of its CPU and encounter up to 20%+ more storage latency when recording video or shooting bursts of images; the buffer would never have a chance to clear. The true hardware solutions like the USB-C drive I sent you manage to skip this penalty because you are technically writing to a storage space after passing through a dedicated encryption chip whose only purpose is to encrypt the data; data which can then only be read after the proper pin has been input into that dedicated encryption chip. Even though this may still incur a small penalty in IO, you manage to get enough remaining IO to conform to the USB-C standard which means it will have enough remaining IO bandwidth to meet the USB-C data transfer standards. Quick Google Search http://cherrybyte.blogspot.com/2012/11/quick-comparison-of-disk-performance.html https://www.phoronix.com/scan.php?page=article&item=2019-linux-encrypt&num=2 https://www.isumsoft.com/computer/how-much-does-bitlocker-impacts-on-hard-disk-io-performance.html

But your encryption workflow involves a multi-GHz CPU, SSD or faster hard drives, and no other processes really clogging up the processing pipeline. Also, it sounds like we are still talking mainly about images. Encrypting a video feed in real time regardless of its final destination is no simple feat even for specialized equipment that is designed to do it. We are talking about tiny camera bodies here whose internals are optimized to move data from the sensor to the storage device very quickly. UncleBob's public key approach still has the problem with verifying the written file prior to clearing the buffer and does not account for video. In my opinion the only reliable way to account for both photos and video would be to have a dedicated chip in the camera with the private key; not just blindly encrypting files on the way to the storage medium and hope they write successfully. This approach would probably also be able to keep up with the read and write speeds the camera needs to function; the only downside being that the files aren't actually encrypted until the camera is turned off, the memory card is removed from the camera, or a an encryption button is pressed. BTW this is also how cell phones and computers do it as well, they have a dedicated chip like a TPM chip which is where the private keys are stored. I still think the best and simplest solution that is available right now is the encrypted USB storage solution. Now how many cameras can write directly to USB-C storage and what the adapter options are for this is a different story. You could have a device like this one in your pocket and no memory cards in the camera or dummy ones that will look empty to law enforcement, the only missing piece is a camera that will write to USB-C but I'd imagine one exists: https://www.zdnet.com/article/this-is-the-worlds-most-versatile-hardware-encrypted-usb-flash-key-with-prices-starting-at-59/ If no existing cameras will record directly to USB-C then THAT is what I would ask of camera makers, that would have much wider appeal and would solve your encryption problems.

I have the GH5 with the XLR adapter and a MixPre6 as well as a C200. I think the XLR adapter or XLR straight into the C200 is great for a single mic or a mic and a backup mic, but I learned the hard way just a few days ago that if you have multiple speakers the Mix-Pre is way better. It was a pain to manually audio duck in Davinci Resolve the two speakers that I had; the stereo mix out of the MixPre for 4 speakers was less work than manually mixing two speakers from the onboard audio. For the C70 I really don't consider the mini XLR a big deal, as you mentioned it is not as bad as mini HDMI. One thing I wouldn't do though is leave the adapters plugged in, I always feel like things sticking out of a camera port will weaken the circuit board over time as they get wiggled and moved in your bag. I'd just load up on about 6 adapters, two go in my emergency kit in my car (along with memory cards, random power cables, etc), two stay in the bag with the camera at all times (since they are so small) and two go in my dedicated audio case where I keep all of my cables, the MixPre, etc. Also, for my particular use case, this camera would primarily be a gimbal camera, audio would not be that important to me.

My wish is that all of this so called H.264 and H.265 hardware accelerated encoding actually worked. I'm sure it helps but it never seems to be enough to make actual editing smooth and I always end up making proxies anyway. With that being said I have pre-ordered the C70. I figure its better to be on the list and cancel then to wish I was on the list. I have been thinking more and more about cancelling though, although this is my dream gimbal camera the more I think about the type of jobs that I do the more uncertain I am that I would be willing to use a camera this costly to do them. Weddings, events, music videos, and corporate work simply do not need a $6KUSD camera on a gimbal. The R6 is what I truly need and the camera that is in the right price bracket for my work but we all know how that turned out.

That does sound like the best solution then, at least for images; ironically the cell phone still sounds like the best solution for video. Fortunately in my line of work the most hostile environment I may be in is the father of the bride dislikes who the bride is marrying.

That would work but wouldn't be a great option, in addition to not being able to review the footage the camera would not be able to verify the integrity of the file after dumping the buffer unless it also signed each file then checked the signature prior to clearing the buffer. Any way you look at it, the encryption process would incur overhead; processing, heat, HW, development, etc. and include compromises (complexity, possible lost data, additional costs, etc) for a very small target client base. Also, if you really are in a scenario where the government seizes your camera; if they can't read the data and you are physically unable to provide the decryption key on the spot they will now have an excuse to search the rest of your gear or they will simply destroy the storage medium. Once again phones have the advantage here, they already have network connectivity, they already have data at rest encryption options, they can already livestream events as they happen, etc. Trying to reverse engineer all of that and put it into a camera then trying to market and sell that camera simply isn't something any for profit camera maker would do. All of the encrypted SD cards that I have seen require a full OS or some way of loading their decryption/encryption application onto the device; something that would be impossible to do with a camera. There may be one that I am unaware of (entirely possible), but I haven't seen a simple SD card you can just plug into a camera and have it encrypt everything that is written to it. I believe they all need an app, and you need to enter the proper authentication information into the app prior to using writable space on the card. I have seen USB drives that have physical PIN pad buttons which is how you enter the volume password like this one: https://www.amazon.com/Apricorn-Validated-256-bit-Encryption-ASK3-120GB/dp/B00W2EN8CE/ref=sr_1_1?dchild=1&qid=1601314672&refinements=p_n_feature_keywords_browse-bin%3A6813186011&s=pc&sr=1-1 but for that to work the camera has to support writing to a USB drive or SSD drive. The FlashAir SD card looks like the best option, but it looks like it would be complicated to setup and get to work reliably. Just turning on and off the camera would require the offloading device to reliably reconnect to the wifi, the app to automatically grab the files as it detects them, then move those files to a cloud backup solution (and we are talking large JPGs here). It's been my experience that getting that many different technologies to work together reliably and repeatedly is uncommon.

I think it is a great idea, but the reality is the market that is requesting it isn't large enough for the manufacturers to listen. Also, I disagree that this could be added to existing cameras via firmware. To create a reliable encryption / decryption process for the massive video and photography files that cameras create you would need a dedicated hardware chip to do so, even iPhones and Androids have dedicated hardware for encryption; Apple spent millions developing their encrypted data at rest and fingerprint unlock solution, and so has Google, Microsoft, Linux, and Unix. This would also incur the headache that current security providers encounter which is constant patching and updating as hacks and workarounds are found. So in a nutshell, you are asking for what would be considered a niche feature requested by a small group of customers, yet that feature would require a team of hardware and software security engineers to maintain, integration into the rest of the camera's processes, a new testing regimen that would have to be added to the test matrix, possibly millions spent on the new hardware and software testing, entrance into the security whackamole world of hacks, reverse engineering, patches, etc. etc. all for a feature that despite your opinion; probably would not sell more cameras. If the general population truly cared about security, Facebook, Instagram, Google, (name your social media network here), wouldn't exist. Lets not forget that cameras are built for speed, from the minute you turn them on to the minute you press the shutter button to the minute the image/video is written to the media; how many customers do you really think are going to want to try to enter a password into the screen before taking a picture or recording a video, lets not forget for a complex password you will need a complex keyboard which is another challenge on a tiny MILC/DSLR screen and another feature no current camera has that would have to be developed. It is easy to say something is easy to do until you try to do it. Your best bet would be a camera that recorded to an external USB drive that had some sort of encrypted data at rest solution, or just get a cell phone with the best camera available. There are two more things to consider that you probably haven't: For the data to be truly secure it would need to be encrypted using asymmetric encryption and the only way to do that in a camera scenario would be to have the second half of the key stored on that dedicated HW chip that I talked about earlier. So what does this mean? It means that the memory card could only be decrypted while the chip was in the camera which means that now you will need to leave the card in the camera and use the camera's file transfer system to get the data off of the card (another inconvenience that most owners will not tolerate) Data corruption - do you have any idea how long it takes to encrypt large files that are already compressed? The camera would have to stay on for the process which would eat into the battery life and the media would have to be very reliable. I can see a much higher chance of data corruption during the encryption process due to the time it takes to encrypt compressed files. I could go on..I have an entire list in my head..but long story short, what seems easy and inexpensive to you is not even close to being easy or inexpensive so I think it is safe to say it will never happen; no camera maker would ever get their return on investment.

I agree 100% with you. I have the C200 and due to the file sizes have never shot raw for a paying job. For the types of jobs that I do, as long as you WB and expose properly the files sizes of raw don't justify the benefits. YouTubers seems to think raw is the holy grail that will instantly turn their cat videos into cinematic blockbusters. With that being said, most people when discussing raw seem to think its only benefit is its ability to have more latitude in color grading. I had a long discussion with Blackmagic Design and the real benefit of raw is the fact that it is designed for editing. For this reason alone I wish the C200 and C70 had compressed raw options. I have no problems with the image quality out of the C200, C300, and C70, but I cringe thinking about trying to edit those files after adding titles, color grades, Fusion effects, and transitions.

I'm thinking definitely not DGO, probably the C200 sensor, no 4K120FPS, no RAW of course, probably same form factor, mini XLR, fewer ND stops, etc. With Canon the sky is the limit on how far they can throw the cripple hammer. It probably won't be less than $4K either which would slot it in right above the R5, but below the C70. At least they can't get away with no 4K.

I like the codecs from a size standpoint and have never shot RAW with the C200 except to test it...my only problem with the codecs is they are not editing codecs they are playback codecs as Blackmagic Design made sure to point out to me when I filed tickets over why Resolve stutters on MP4 and MOV footage. Since they are playback codecs editing with them is a real pain; everyone thinks color grading is the only benefits to RAW or compressed RAW; the real benefit with an editing codec is you don't need to transcode first. I have a 14 core CPU, and an RTX2080TI video card and I still need to transcode if I want smooth playback in Davinci Resolve after color grading, adding some Fusion effects, transitions, motion graphics, speed ramps, etc. With something like BRAW or ProRes I could save a lot of time.

I think picking up a C70 used for around $4K USD would be worth it vs. the S5. That way I'm not giving Canon my money, I'm getting a gimbal camera that can do 4K120, internal NDs, internal XLR, Canon color, and Canon AF. The problem of course is holding out long enough for a used one at $4K to come along.

The codecs are pretty annoying...strong for Canon but below avg for the competition. Editing those types of files nearly always need to be transcoded first. I do think it would do great for corporate work and it would be nice to use a single camera for both gimbal work as well as interview type work. Right now I use my C200 for interviews and tripod work and the GH5 for gimbal work and b-roll. For lower budget clients I could use this camera for both. If I need RAW I can use the C200 or rent another body. The URSA 12K looks great but it is over twice as much and the gimbal for it will cost more than this whole camera, not really considered a run and gun camera due to the extra rigging required, it will need proper lighting and rigging to get the most out of it, and the storage needs are astronomical. It definitely has its place but the URSA and C70 simply aren't meant for the same target audience. I like to stick with bodies that can shoot 90% of my work and for that last 10% I can always rent the gear needed for the project. One thing I can't tell is if APS-C lenses will work with the C70. I know they have the speedbooster, but no one has said if the speed booster will only work with FF or will it work with APS-C lenses as well. My favorite C200 lens is the 18-35 Sigma APS-C lens.

I have never even looked through my GH5's EVF, I always use the flip screen. It is way steadier to tuck your elbows and hold the camera with a 90 degree bend in your elbows close to your body than to try to use the EVF. From there I can do pans, tilts, simulate slides, etc. all handheld. If you mean using the EVF for shoulder work then yea they want you to buy a bigger model for that, but for handheld I hate the EVF, I completely disabled it in my GH5 to save the battery life. I love the idea of the S5 but after buying that and a few L mount lenses it would cost me more than the C70. When I travel I actually just use my 5DIV, and the Mavic Pro, since my travel work is not paid and the 5DIV has really good 1080P which is enough for me to use as b-roll clips later on for things like music videos, nature etc. If I am traveling on a paid assignment then usually it is just video and I take the GH5. If I'm traveling for a paid hybrid gig then I lug the 5DIV with the 24-105 and the GH5 with a few lenses...so yea this camera doesn't help at all in that scenario. I'm thinking even local hybrid shoots though, I would still need to lug around the C70 and the 5DIV. Many times I just need photos and some b-roll video clips...I don't trust either the R5 or R6 to do this right now. For long form stuff I still have the C200.

As mad as I am at Canon right now, this camera looks amazing. Now we all know why they crippled the R6 and R5, they really wanted you to buy this camera. This could be the GH5 replacement I was waiting for (dual slot recording, XLR audio, good AF, RAW 4k60FPS output, 4K120FPS in body, no overheating, no recording time limits, small enough to fit on a one handed gimbal, uses my existing Canon glass). My biggest problem though is I still would need to lug around my 5DIV for hybrid shoots....so annoying.

I'm waiting for the GH6 as well. I do think an L-mount GH6 would be a bad idea, they already have that with the S5, and the reason I am not interested in that is specifically because it would mean I have to get L mount lenses to have a native system. If the GH6 has 4K 120FPS and some form of in body compressed raw options that would be good enough for me. Not having to buy new lenses or buy into a new system is the main thing I care about. I don't think Panasonic will ever get PDAF, there's rumors Sony simply won't license it to them, how true that is I have no idea but I have to believe that if Panasonic had it available to them they would have implemented it by now. With that being said, my C200 has DPAF and I still almost always shoot with MF, I just don't trust AF for critical shots, so if the GH6 has no useable AF that's not a deal breaker for me. The Canon C50/C70 looks pretty fascinating but I'm still at least a year away from getting over my disappointment in the R5 and R6. I already have a C200 and a GH5, I don't need another body that can't take pictures as well as video, what I need is an R6 that actually lives up to its marketing materials.

Yes Resolve is incredible, but it can be pretty hard to find some of its features. I think the key to my color grading workflow is I don't add the adjustment layer until I have a base grade for all of the clips, that way I don't accidentally end up with it selected. It's pretty annoying that DR does not provide an easier way to do this, and I accidentally start working on the wrong layer all the time if I have already added say a logo or watermark at a higher layer. I now try to color grade before adding text/titles/callouts etc. but if a client wants changes after seeing the project that's where things get time consuming. I learned the memories trick from the following video, and out of all the different ways to copy a grade to another clip I liked it the Memories trick the best, if the clips are side by side I just hit the = key but it usually doesn't work out that way. For speed purposes I just put everything into one node unless I'm doing something complex like power windows, selective color, green screen work, etc. I used to do multiple nodes and all of that but it takes too much time; create a base grade, throw a finishing LUT into an adjustment layer....done. I literally did not leave Premier Pro until DR added the adjustment layer feature. It is a huge time saver.

My entire process is set up for speed, so it may not yield the quality that you are looking for but my workflow is as follows; some of the steps may help you speed up your workflow.... 1 - Import all of the footage from the memory cards to the PC into a folder called RAW clips (sub folders separate the clips by camera) 2 - During the import separate them by camera using the sub folders (I typically can have up to 4 cameras) 3 - I have a second folder called selected clips 4 - I open the RAW clips folder in one window and the selected clips folder in a second window 5 - I preview each clip in VLC then drag it to the selected clips folder if I will use it. Any clip that looks totally unusable (false start, etc. I drag to a discards folder). 6 - From the selected clips folder I drag it into DR 7 - I build out the timeline do all of the edits etc until the project is done. I dump all of the cameras on a single layer and only add layers for text, multi-cam scenarios, overlays, special effects, etc. I don't bother labeling the cameras, naming the clips etc (unless I am going to add it to my stock footage library). Every min I spend on a project is $$ lost is the way I treat projects. 8 - I move to the Color tab and when I encounter the first camera I apply a grade that brings it to the Rec709 standard based on the WFM. Once the grade is complete (LUTS, manual adjustments etc) I save it to Color memory slot 1 > Color > Memories > Save Memory A (or just press Alt 1). Each time I encounter that camera I press Ctrl 1 to apply that camera's grade. For the rest of the cameras I repeat the process. By the 4th camera I have base grades for all of them saved. 9 - After pressing Ctrl + camera number to paste the grade on the next clip from that camera, I make small adjustments in case it is the same camera but a different scene (i.e. indoors vs outdoors etc). 10 - After all of the base grades are done I go back to the edit tab and add an adjustment layer over all of the video clips but under any titles/text/logos etc. 11 - I go back to the Color tab and apply a finishing LUT to the adjustment layer. 12 - Color grade done With my process my entire color grade for a typical 2-5min project with 4 cameras and 100 clips takes less than 10min. I have optimized my workflow to minimize the throwaway clips in my NLE, maximize my editing speed, and with the added benefit that if I start running out of storage space in my archives I know I can go back and delete unused footage from the Discards folder without affecting the finished project. Also for speed, my entire folder structure for photography and video is set up as a templates in a read only folder so I just copy that folder structure each time into a new project folder when I start a project. I hate spending 1 additional unnecessary second on a project if I can help it. I tried DR's Use Remote Grades feature but for me it was confusing and annoying to work with if a single clip had multiple exposure changes (say walking from outdoors to indoors or the sun went behind the clouds) so I like the memories slot approach better. I also tried the apply grade by camera approach using bins but there is no single camera base grade that will cover every scenario so I found it faster to build a base grade for that particular project for each camera by hand or via a LUT with manual adjustments. I have also had multi-day projects where the next day was a completely different venue and there is no way I would want to use the same grade from the previous day even though it was the same camera. My process I'm sure would fall apart for really large projects but for me where the typical project never exceeds 4 cameras or 100 clips it works great.