Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Please, support PV!
It allows to keep PV going, with more focus towards AI, but keeping be one of the few truly independent places.
New RAW / 4k / 2k / Prores / DNxHD camera
  • 69 Replies sorted by
  • I don't know if we are not over analyzing on this thread. If @peterosinski is not asking for money why do people have to be so negative. If he was doing a digital Bolex, I could understand the hostility, but until now it is not the case.

    I know that his story is a bit too good to be true, but we never know. Perhaps other companies are seeing some potential with the interest in the digital Bolex or now the Black magic camera. It is not that complicated to all those machine vision camera companies to do a head with uncompressed sdi or hdmi output that can be recorded by an external recorder.

    In fact I find the idea to be very very good. You buy a camera head and as your need grow you can buy another recorder to get raw or 4k. People on this forum should also have a little sense of perspective. A lot of people are buying a muddy barely 720p. 4.2.0 camera with low bitrate for $ 3500. I think a $ 4500 full frame 10 bit 422 dnxhd camera is still very very competitive and it can grow as you have money to buy another recorder.

    It is not because we can have the high bitrate gh2 for $ 700 that everything must be compared to that. What the gh3 lauch has showed me is that be prepared to pay a heck of a lot more for the next step from companies like Panasonic for things like 10 bit etc. They even crippled the hdmi output so that it stays at 4.2.0. Until a hack come and brings 422 and 10 bit and a log output, I don't think we will see anything revolutionary comming from the gh line.

    Lets not talk about Sony, which have invented a new feature in there camera called aliasing and moire. With all the model they have realeased in apsc thay could at least have developed a low end native 1080p sensor or even use the f3/fs100 one. No, they just put a dslr sensor in video camera so that they can differentiate them. So don't expect too much from these big Japanese companies. At most we will could see an apsc version of the BMC in two years. As for peterosinsk, I hope he does not get discouraged by the response here and continue in his endeavor, even if I won't believe until I see a working prototype.

  • @danyyyel This is the most constructive post so far, thank you.

    One thing I am finding out for sure is that my biggest cost is going to be the sensor. The size of the chip is large and therefore the yield of sensors from a wafer of silicon is low which drives up the price.

    I will keep you all posted on what I find.

  • @peterosinski

    I think that some negative tone of my comments and some other comes just from some facts you provided and prior topics on various forums.

    I could fully understand such posting in case you made prototype, made business plan and come to ask or make some marketing research.

    May be I just do not understand something.

  • May be I just do not understand something.

    I think you do understand :)

  • May be I just do not understand something.

    The word "my" put in front of things ("my camera") that don't exist, with the price of that thing (that don't exist) being the main subject launched (and relaunched again and again) by Op... well, it's a little bit difficult to understand something in that... thing. =)

  • @vitaliy_kiselev prior topics on various forums by myself or someone else? Also, I am trying to find opinions on the general idea. Due to cost it would be too late in the game to ask opinions on the general concept of a camera once I have a prototype

    @atraban would you prefer I say "the camera I would like to make"? What do you mean by relaunching? I was responding to comments and looking posting further details?

    Further details such as this: 1) Due to the cost of silicon the sensor alone will be around $2000 2) I am going to need to use an off the shelf sensor because the cost of developing a brand new sensor is in the 1-2 million dollar range 3) My research shows that the only way current full frame dslrs are able to produce 24fps and higher at their current costs is due to line skipping.

    Also, is it so wrong to discuss concept ideas for a project on a forum that speculates and creates wishlists for cameras before they are officially announced? Would this topic have been better received if I had named it "I hope someone makes a full frame 4k machine vision style camera that can use readily available external recorders. What do you think?"

  • Due to the cost of silicon the sensor alone will be around $2000

    Can you show us source of such information?

    My research shows that the only way current full frame dslrs are able to produce 24fps and higher at their current costs is due to line skipping

    Same question.

    Also, is it so wrong to discuss concept ideas for a project on a forum that speculates and creates wishlists for cameras before they are officially announced?

    It is not wrong, but you need real ability to produce camera :-)

  • Kerry Van Iseghem of isgchips.com to answer question number 1 and 2 in addition to other sources

    This of course does not apply to the chip in the 1d-x and 1d-c but of course those are 6 and 12k cameras

    I'm not asking people to put faith in me to produce a camera. I'm asking a group of people with similar interests, some of whom with needs which may be similar to mine, what they think of the concept

  • Kerry Van Iseghem of isgchips.com to answer question number 1 and 2 in addition to other sources

    And you really saw tech specifications? I mean other than just plain talks from guys who sell very costly cameras.

  • It was a discussion with someone who has a reputation in his industry. Also, the cost of silicon came up when we discussed price. The point he was trying to make was that for a sensor that size it is too expensive for him to even be able to mark up properly and he advised that I could save cost by purchasing it myself and having it shipped to him. That way he doesn't have this risk of taking it in as stock as it is a specialty item

    I would compare it to some of the information I get on this forum. If you tell me that there are some things that the gh2 simply cant do such as 4:2:2 output. At a certain point it is easier for me to take someones word than look into it myself. For example, if you were to tell me that it was impossible to hack the gh2 to be able to do something, I would say "well if he says so I guess I'll go with that because it's not like I'm going to hack the camera" The same applies to this. I will seek a second opinion but when it comes down to it I'm a filmmaker not an engineer

  • @peterosinski The concept shows potential. How attractive it ends up being in the end will depend on the specifics and the competition at that time.

    If you can (essentially) do a 4k full-frame BMDCC for less than a price of an Epic, there could certainly be market interest if it is done well. Just make sure to take a look at what the sales numbers in the price bracket look like. And get a sense for whether your product offers enough to make people convert or bring in potential new customers with the income and inclination to spend, before you invest yourself heavily in the project.

    I think that one of the things that people in this thread have been trying to emphasize is that the market is a lot larger for lower priced cameras than high priced ones but that the lower-priced market is also already very competitive.

  • @thepalalias true, but there are some things that can't currently be accomplished at certain prices.

    For example, below is an quote from the CEO of black magic talking about why they chose the sensor size that they did. Note the cost he says it would require to find an affordable full size sensor (>$10k). I am finding some sensors that could work but many of them could only attain 24fps or higher with line skipping. Others that fit the specs are actually closer to the s35mm frame size which would defeat the point of the project (4k vistavision) and the s35mm market is already saturated.

    " It’s basically a 4/3rds sized sensor… the thing we’ve had to kind of explain to people and on and obviously I think we all, whether it be Black Magic or when I speak to potential customers customers about the camera we always seem to try to get back to the same place which is: it’s an amazing camera – Specs or cost or whatever all of it’s to do this exciting product there’s always that “Why couldn’t you” or “Couldn’t you have done” and I think the important thing on this answer that you have to understand is two or three years ago when we said we could build a camera, one of the things we had to do was find a sensor. It wasn’t like we were going to go out and design a whole sensor from the ground up – that’s not who we are or our forte. Nor were we going to be ringing up people asking for quotes on sensors because you know we can’t just tip our hat tip our hand that way. We had to go shop for the sensor we thought we wanted. There were several different things on that shopping list, how big could the sensor be, obviously now much would the sensor cost -we knew we wanted a really high dynamic range, we knew we wanted higher than HD resolution… \

    And the sensor we eventually found was just an off the shelf part and we got most of what we wanted on there to build you know kind of incredible camera that we have there.

    So one of the things we hear a lot of is could you do a full size sensor or a 2/3rds sensor… certainly we could if we could find that one discreetly that would be less than ten thousand dollars and do higher than HD resolution and high dynamic range with a large degree of stops – that would have been great. Would have been a 25 grand camera maybe…"

    you can view the entire interview here http://filmmakeriq.com/2012/07/exclusive-blackmagic-answers-your-questions-about-their-new-cinema-camera/

  • @peterosinski

    Good thing to understand. Your only potential advantage is raw recording.

    After this check other BM staff interviews and notice how long it took for this company to design and start producing cameras (having good resources). And note that they already missed shipping dates.

    Whole point here that in the time your camera will be available any of major players can introduce relatively small camera modification allowing recording of raw data from sensor. And this means end of your game, with still very big money you'll need to return.

  • I believe that it took about a year and a half from the point in time the committed to building a camera. Also, while you are correct the have the advantage of good resources you are incorrect that this is a disadvantage to me. I will be taking advantage of the resources of black magic and many other companies by using their external recorders.

    This saves me the trouble and the end user the cost of getting licensing and integration with prores and dnxhd. Also I have been trying to contact cineform regarding their raw format but unfortunately it has been impossible to get anyone to answer their business development line.

  • @peterosinski I agree that there are things that cannot be accomplished yet - I think the thing to keep focused on is whether the same people that it would be most desirable to are the people that can and will pay for your product.

    Personally, I welcome RAW recording options at every sensor size and would almost certainly rent the camera you mention, possibly frequently - but I could not guarantee that I would buy it.

    As far as Cineform, use a different department to get a referral - the business development line may be understaffed.

  • I have done some more research and am looking for some more opinions.

    I was able to find out from one of the sensor manufacturers that I have looked at that they will have a 5k x 3k sensor capable of 30fps coming out at the end of the year. They currently only offer it in monochrome but the new version comes out in december. The sensor itself costs 2.6k usd but that goes down when buying in bulk.

    The next thing I have been trying to research is how to get cineform raw recording in-camera. I have done some testing of the cineform raw codec by converting cinemadng to cineformRAW and the file size/quality is great. You have tons of flexibility and the included LUTs in cineform lightroom are great for making dailies. That being said the cinemadng to cineformraw is a command-line-pain in the ass. I think that it would be better to record this in camera. That would add to the cost of the camera but remove the need for an external recorder to achieve RAW recording.

    The original concept involved getting audio into the file by using the external recorder. If I go this route it will require either a MOS approach or I will have to add audio connectors/headphone jack. I don't think people would go for a MOS camera these days, what do you think?

  • You are INCREDIBLY underestimating what it takes to make a camera, not to speak about the camera you are talking about. Do yourself a favour and quit wasting your time.

  • @stip fair enough, even still, there is no harm in having a discussion. Not to mention I never said it would be easy, or cost less than a house to make.

    I just find these projects interesting. It's funny that there is no discussion here, just nay-saying. I don't see that same mentality in the DIY focal reducer thread.

    Anyone care to comment on the concept of a cineform raw camera itself rather than the logistics of making a camera?

    I have been playing around with some cineform raw avi's and they playback like a dream with cs5.5 on my laptop (im working on a laptop in a hookah bar right now) at full resolution 2.5k from black magic cinema camera. The data rate for 2.5k seems to be about 30MB/s. By rough estimate that would put 5k at about 128MB/s. You would still need a RAID to pull off those data rates in real time but it's still more reasonable, not to mention easier, than working with full res cinemadngs

  • @peterosinski Cineform RAW is one of my favorite recording formats in terms of the size to quality ratio and the flexibility (and more importantly, in terms of timeline playback on older systems). As a consumer, I would like to see it implemented.

    The audio side of the equation is a complicated and big deal. If you get further along and show that you are committed to the project, I can refer you to companies that have a good history in terms of either converters or headphone amps (or both) and to some guys with a lot DIY experience if you choose to go that route.

    At this point, I would keep focused on the video side exclusively - it's a big enough challenge on its own and solving the audio side will seem simpler by comparison (because honestly, what is considered "good enough" on a camera is usually pretty poor compared to what is considered "good enough" in a recording studio).