Sun Mar 25, 2018 7:11 am
John, I do not wish to talk with you. You do this a lot. You misinterpret things into something else, miss accusing me. You talk about what 'you think' despite what was said that was on a different track and to answer different points than those made. You apply religious like belief in faith of how a company should act to suit yourself, rather than the majority. To bind up the expanse of what they can do, limiting them to one small market, rather than allow them to do best for the price, or to include practical industry standard acquisition. So, nobody gets a 4k pro video recording head that doesn't overheat because ... Nobody gets a cheap 4K pocket pro video camera that doesn't overheat because .... No, it has to be only be what you say, and the competition can steal millions away. If they can do a $500 4kp50/60 12 bit 4:2:2 pocket camera for the type of filming you dont want, then let them. If they can do a $1000 8k camera to do the sort of filming you don't want, then let them. If they can fit a raw codec at those resolutions in, even better. I'm not limited. Did it ever occur to you that even if they could only support the full resolution with a lesser codec, they might be able to manage 4K or 2k with a prime codec in the same camera, and next model might get the full thing then. But that means they are in the market with the best they can do 1-3 years earlier, keeping the competition at bay. Not doing this, classically, only invites somebody else to fill vacuum. Which is how SI, RED and BM got started. But the raw Bayer quality BM archives can be done at lower price. So in business you can't just sit still and get overun by the new kid. Besides, Red is coming down the top end as well into 8k cheaper cameras. Red likes price, so there is likely to be room under them. Practically it is extremely costly to go it alone and make technology of their level outright from scratch (though there are weaknesses that this mass market product is likely overcoming. ASIC requires high numbers to make this chips cheap. FPGA has even more liability, as they run hot, and are costly. That is where mass market Application Specific Integrated Circuits, as processing solutions come in, as the cost of development and production is shared among all manufacturing customers into the millions, making them incredibly cheap for what they can do compared to fpga or much low volume asic).
You also don't understand compression, you set up an acquisition codec different from the delivery, they also have various levels under the same name. So delivery quality, or the quality from another camera, does not matter. The camera you are on can be setup by the manufacturer with certain qualities, or desirable qualities. The initial prepping of the image can be setup to be more aggressive on the contrast and colour, or less. But the reality of compression itself is the higher the data rate, the less difference between compression routines. So, at 1:1 many should be the same. At 2:1 some will testing lossless compression, at 3:1 many will not, and maybe a few above 4:1 these days. I imagine that BM still uses compression at below 2:1 to maintain lossless. Red, has access to very good technology, and so 3:1+ should be possible, even above 4:1. To look at a similar performance codec, visually lossless on cineform Bayer is around 6:1, but goes up to 10:1 from memory (though maybe I've got those values shifted after all these years). Visually lossless is the range you want fur cinema. How much BM does visually lossless at the moment, I didn't know, but don't see that high. Maybe 4:1. But, it is likely that BM's raw compression ratios goes beyond what is needed to maintain visually lossless. 300 mb/s h264 on a good camera should be visually lossless (as this is how the industry sets these rates for rntry level pro video cameras in the digital era). Saying it always has to be more is indexed.
However, I strongly suspect, that if you look at the performance curves in the colour filter, you will discover why it's colour is so good. Any tests out there on pure primaries and complementaries to test the colour gamut?
Hmm, here is another missaplication and miss accusation. I showed that there is a new better version of the zcam coming out. I've also shown over time that it uses a bigger sensor etc without heat issues. All in answer, as per only usual, to miss information being spread by others. Doesn't at all say BM should or could make something as low end as the E1. The with is at least one or two grades down in sensor and more so in compression rate, before I think it's good enough to use even as a recording head. The most bazaar accusation. But side tracking to prove a point not raised. If the new one has the hc3 chip in it, with pro video modes (12 bit 4:2:2 +) running at full speed, then that would be different. However, if we can hook it up to a VA with a good sensor that would be good.
Wow! You really went to town on zcam. I remember an issue of Your Computer years ago, which had a drawing of the editor looking into his crystal balls, predicting future technogy. The past does not predict the future. You have no idea of what he will release, and I'm the only one of us that have even bothered to talk with him. The 300 mb/s statement does not even warrant another response.
Visually lossless, is where pro video production cameras start. Why do you think they had 10 and 12 bit 4:2:2 vussuslky lossless on a 8 bit 4:2:0 delivery of half the data rate or less, to colour grade and correct the image, that's what it was designed to do. An Apple doing visually lossless, you don't mean the standard datrates rates, that are far too low do you? How familiar are you with 300-600? After you get rid of artifacting it goes towards picture quality. You must be thinking some rubbish sensor, put through a compromised processing path to a 100 mb/s h264 codec is the same as a good sensor with an uncompromised compression path at 300 mb/s. Also, h264 has dramatically improved too. You should also calculate the numbers on the old cinema delivery furnats, to see how low they are. There were a lot if cinema production people that near miraculously saw lossless compression quality on Reds 7-9mb/s delivery codec. I think there might have been drinks around at midnight. It likely was just as poor as a cinema delivery. That codec is generations ahead of h264, but it shows, it is not the bit rate, but what you do with it. Also, there is a dramatic difference between h264 between cameras. Done are pooir, and you shouldn't make rash judgement. They were exceedingly poor before ambarella came out. Ambarella was using a higher h264 profile on those little cheap cameras, which was much better per bit than any of the h264 from camera makers I remember. So much so, the better camera makers licensed their technology. So, tools and camera makers not using prime h264 are going to likely look lousy. So, the question is, what are we looking at when we see these things. I also would like to state, I have no reason to believe that a top camera maker won't sabotage the h264 quality on an entry level pro video camera, compared to their higher end model at the same bit rate. Again, one example dies not predict the other (except you know that they want you to spend up big on the more expensive model). I would suspect that BM could make it also do the max.
Now, I forgot in the reply before. That not everybody wants raw. That it adds dramatically to the costs of storage and computational power, as these days GPU cards have circuites that favour h264. On the street, prime h264 is more than enough often. So, the h264 option makes sense on the lowest end, but nobody I know of is making prime h264 at the low end (it's too good). So, there is a frantic two in the market there (which boosting phone video might touch, but those phones are expensive and often a little sensor with a little lens).
But I notice rather than acknowledging, you often shift the argument away from the actual points made, my points. If you fellows want to argue against some odues points, I suggest you stick to them and only address them.
Now enough of this ranting against me.
I've wasted 6 hours giving reasoned replies I didn't need to.
aIf you are not truthfully progressive, maybe you shouldn't say anything
bTruthful side topics in-line with or related to, the discussion accepted
cOften people deceive themselves so much they do not understand, even when the truth is explained to them