Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

sony

Graeme Nattress said:
Bruce, the point is that Q67 has been aggressively marketed as a sensor design against us "Bayer traditionalists" with some real slight of hand (doubling the pixel count) to "prove" it's a superior design, whereas simple analysis shows otherwise. That's why we compare same number of pixels because their argument is about a sensor design, not a camera. When they have a camera, we can compare that with a current Epic and see how things line up.

As for Q67 simplifying processing, it does so only at the expense of doubling the number of pixels needing processing - I don't see it as a free lunch in that regard.

Graeme
Click to expand...

Has anyone claimed a free lunch? Or simply an engineering compromise with some potential benefit? Seems to me it is at least possible that more pixels - if they are easier to process to get accurate color - could be faster than more complex processing of fewer pixels.
 
Bruce Allen said:
As far as I understand it...
Click to expand...

Bruce,

Thank you very much for that explanation. That definitely makes a good deal of sense. Given the importance of broadcast in Sony's market scheme, I can see why they would implement such a camera to provide quicker, simpler, and probably cheaper (?) instant 4K out the back of a camera. Honestly, I wish the Sony presenter had explained that, and perhaps actually showed how the Q67 makes a difference, rather than stating it and then showing a few "artists rendering" still frames.

I'm still not really sure I like the way that Sony presents it's material in that presentation. At this point, I'm ready to forget about the marketing presentation and actually see the camera and what comes out the back. I imagine I'll have to wait a bit for that. :cheers2:
 
Looking at the bigger picture here. If Sony is now ready bring 4K into the home of everyone in the world, which is something that takes a company as multifaceted as Sony to actually accomplish, and this is the sensor design they say they are going to do it with. Why on earth is anyone from Red/Reduser trying to say it won't work or is a piss poor way of doing it?!?!?

If they are bringing a 4k camera to market you can bet your life they have 4k displays, broadcasting equipment, etc, etc in the pipeline as well. All the things necessary to allow everyone in the world to see the awesomeness of 4k. This can only help Red sell more cameras and allow everyone to get the most out of what Red is offering. I would think that however Sony wants to bring 4k capture to the market is a good thing because you know they will bring everything else along with it.

Selling 7000 or 8000 cameras in the professional camera market is cute and everything but we're talking about Sony, a company that sells 10,000+ cameras in EVERY consumer, prosumer, professional market.

If you want to see 4k delivered in the home of your clients and their target audiences who cares if the sensor that will start that process is rotated, skewed, inside out, or backwards. If it takes a camera that has a 6k sensor that outputs 4k in realtime to every home in the world I say that's cool, awesome, do it already! Stop thinking short term Red/Reduser, there is plenty of room in the market for multiple camera systems and sensor designs.

This is ultimately what Jim wanted to do right? Give everyone in the world incredible images at higher than 1080p resolution? If this is what it takes to make that happen so what? This is one of the ways that will allow Jim's dream to become reality. It might not be exactly how he wanted it to happen but don't the ends justify the means?
 
Adam, with "Why on earth is anyone from Red/Reduser trying to say it won't work or is a piss poor way of doing it?!?!?" you're attacking a straw man. Sony video division have, over the past few years, been very negative towards Bayer pattern sensors. This is their latest negative attack. All I've done is point out quite clearly that such a sensor pattern as they propose is less optimal than a traditional Bayer pattern with the same number of pixels. Think of it as a math exercise.

Graeme
 
Bruce Allen said:
Love Peter Sellers. So, did you bother to actually read what Mike or I posted? It's about designing a sensor around easy-to-decode, "good-enough quality" 4K realtime output.

For example, in broadcast situations, realtime 4K output might be a nice thing to do on-camera. So that you can watch sports AS IT HAPPENS, for example. Radical concept, I know.

Anyone actually want to read my posts and come back with a well reasoned answer about why Q67 is bad from a computational or OLPF design perspective - with the goal being 4K realtime output for broadcast?

The fact is that Red doesn't complete the 4K puzzle at the moment. The moment we can get an easy 4K broadcast pipeline is the moment 4K TVs stop costing stratospheric prices.

Sony's (Q67-based) sensor seems to be a good compromise to solve those problems. Epic with a built-in RED ROCKET-style module sounds like another. But instead of talking about that, everyone is running around comparing Q67 to a 6K Bayer sensor, making stupid diamond analogies and completely missing the point.

Bruce Allen
www.boacinema.com
Click to expand...


Well I was the first one to point out that the reason for the 45 degree rotated array is that would be much faster to read out this way and this is possibly the primary reason for the arrangement. I think people are still comparing apples and oranges when they talk about speed. CCD architecture is completely different from CMOS. When I say fast I mean fast for a CCD. If you look at any high resolution industrial sensor that are 16MP +, the frame rates are abysmally slow. What I’m trying to say is that there is something radically different about the architecture of a 17 mp CCD sensor that can globally shutter at any kind of decent frame rate. I am very impressed with this.

However I do take issue with the Q67 “dodgy high school new-math” that was presented as a marketing tool designed to mis-lead. I’m with Graeme on this one as the math speaks for itself.

Exactly as Graeme says very concisely …
Graeme wrote

“the point is that Q67 has been aggressively marketed as a sensor design against us "Bayer traditionalists" with some real slight of hand (doubling the pixel count) to "prove" it's a superior design, whereas simple analysis shows otherwise. That's why we compare same number of pixels because their argument is about a sensor design, not a camera. When they have a camera, we can compare that with a current Epic and see how things line up.

As for Q67 simplifying processing, it does so only at the expense of doubling the number of pixels needing processing - I don't see it as a free lunch in that regard.”

///////

The fact that the thing is called a Q67 is really bad, as what they present is less than a mathematical half truth. They could call it the sQew 45, because that’s what it is.

It’s not the camera it’s the marketing that’s dodgy…
 
Sony video division have, over the past few years, been very negative towards Bayer pattern sensors. This is their latest negative attack. All I've done is point out quite clearly that such a sensor pattern as they propose is less optimal than a traditional Bayer pattern with the same number of pixels.
Click to expand...

So you have to stoop to their level by saying negative things towards them back? Is this kindergarden? I thought you guys wanted to raise above the stuff the "big boys" do.

They are saying they will have more pixels btw, so I don't know if it's fair to take a smaller sample and say it's equal. This sounds like that 3:1:1codec with a 1080 sensor produces the same amount of color information as a 4:2:2codec with a 720 sensor argument all over again. And you know what? You could still get a good chroma key out of either of those cameras. That argument is even more moot today since there are so many quality post tools out their. Sure we want the best quality we can get but I think it's hard to say they will produce the same quality image without even seeing what their system is capable of. You guys like using the term "magic pixies," well maybe they have some of those themselves. And you should know better than anyone that "magic pixies" can be used to do wonderful things that nobody thought was possible primarily when it comes to a surface level presentation of their tech.

So what if it's really a bayer pattern? Who cares if they have said time and time again that it's inferior. If they are doing it, and calling it Q67 or Q908123, obviously there is something to it. So I guess that means Red was right and is super cool. If they are agreeing with you and doing what you guys are doing, feel good about it.

What I'm harping on is you saying they won't do as well in aliasing, demosaicing, resolution, etc, etc. You're speaking negatively towards them about those things without actually seeing what the end result will be. Don't go down to their level, it says that you're not confident in your product. I know you are, so let the final products speak for themselves, like the real professional you are.
 
Please let's ease up on the "Sony bashing" claims: Typically, the trouble lies not so much in the motives behind the line of reasoning, but with the subsequent and downright exaggerated claims made about the motives behind the line of reasoning. If it isn't Sony bashing, it will soon be proclaimed as Sony bashing. And, for all to see, a shiny statue will be erected in the name of Sony bashing. "Let's put an end to that rumor right here and now, Barton."

I can understand quiet clearly what Graeme and Eric are saying, and Bruce has helped a great deal in qualifying Sony's claims. M Most, too, if a little anecdotal at times. Until the camera is out on the streets, this is what we have to discuss.
 
Adam, read through my posts again carefully. They're a mathematical analysis of the sensor design. I'm not the one playing childish games - I don't double pixel counts and pretend that a new arrangement of pixels is adding value to the image when all they've really done to improve things is add more pixels.

As for "won't do as well in aliasing, demosaicing, resolution," if you read through my analysis I based the traditional Bayer numbers off current data from working systems, and I gave every advantage to Sony's arrangement and the numbers then say what they say. Feel free to discuss the technical issues here, but from my point of view, the only bashing going on was in the Sony video.

Graeme
 
Bill Anderson said:
Please let's ease up on the "Sony bashing" claims: Typically, the trouble lies not so much in the motives behind the line of reasoning, but with the subsequent and downright exaggerated claims made about the motives behind the line of reasoning. If it isn't Sony bashing, it will soon be proclaimed as Sony bashing. And, for all to see, a shiny statue will be erected in the name of Sony bashing. "Let's put an end to that rumor right here and now, Barton."

I can understand quiet clearly what Graeme and Eric are saying, and Bruce has helped a great deal in qualifying Sony's claims. M Most, too, if a little anecdotal at times. Until the camera is out on the streets, this is what we have to discuss.
Click to expand...

Just to set the record straight,

I didn’t say that the Sony de-bayer would be inferior in any way, or that there has been a radical speed and quality trade off, and I pointed out that possibly the 45 degree arrangement creates much faster read out times… and for my applications a global shutter IS very important.. and as I have said I have tremendous respect for SONY as a corporation.

However as an engineer I take issue with someone in a suit trying to pull the wool over my eyes with “con-math”, and the fact that the instrument is named for it, I find really offensive.

This is not good, I’m not saying Sony is bad and I’m not saying that the camera is bad, I’m saying whoever is behind this type of marketing should seriously re-think their approach.
 
Believe me, Eric, the counter "bashing" claim wasn't aimed at you. Nor was the post meant to imply that your position is anything less than fair, professional, erudite... ( :
 
Bill Anderson said:
Believe me, Eric, the counter "bashing" claim wasn't aimed at you. Nor was the post meant to imply that your position is anything less than fair, professional, erudite... ( :
Click to expand...

No I know, that was for some of the others (they are OK too),

Bill, that really is embarrassing, you didn’t have to say that, in all your posts you always come through as a really nice guy with interesting fun and smart things to say.

I have to admit I am ashamed of how much of the past two days I have spent on this thread; and no doubt I will meet the appropriate Sony executives at some point (assuming I don’t get myself into trouble in the mean time…)

Cheers,

Eric
 
Adam S said:
So you have to stoop to their level by saying negative things towards them back? Is this kindergarden? I thought you guys wanted to raise above the stuff the "big boys" do..
Click to expand...

Adam.. Or should I call you Blaire? or more accurately Dwight, or Mr. Tudor?

Real names only please.. specially if your gonna call my buddies into a fist fight.
 
I'm certainly not looking to get into a fist fight, especially, not with you or any of your friends. You're a big dude. I was just suggesting that we all here try to treat other companies as we would like to be treated. I don't like reading so much negative stuff from everyone about things that haven't even been delivered yet. I don't like listening/reading to all the Red bashing that has taken place over the years. I know you guys don't like it either. So when I see Red doing it I feel like I'm back at NAB2007 all over again.

I respect everything you guys are doing, just like I respect a lot of what Sony, Arri, others have done. I find it difficult to listen to representatives from those companies negatively discuss the various design aspects about you guys. So when I see you guys doing the same thing it's disheartening. Those other companies don't listen to their customers, you guys do, which is amazing. I was hoping you would take my posts as a Red supporter and constant user who is saying they are tired of reading/listening to all the bickering and trying to bury each other before a product is even released, because that is honestly what I was intending. Ban me and erase my posts if you must, I understand this is your house, but please try to understand where I am coming from.
 
honestly man, nobody is really bashing anyone here. we're talking about facts and theory that we know, you can't really argue against that. the people that are getting upset, like myself, are upset because of the dishonest marketing schemes coming from sony. a reasonable man cannot argue otherwise. they've spent the last three years convincing their customers that a 4k bayer pattern is inferior to their damned tape decks or whatever, and now they release one and try to pretend like that's not what it is. that's exactly what they've done: historical facts. no debate, no bashing. that's just what it is, and they've lost my respect for it.
 
There are a lot of problems with this graphic and the verbal explanation offered.

Can anyone explain how you get 100% green area coverage of the sensor, when the green photosites only populate 50% of the sensor?

There was no indication of multi layered photo sites or the use of prisms?

[Especially given that the comparison is made to a 3 chip prismic system, or hypothetical true RGB]

(I have to do some work today so I’m not really here).

Cheers,

Eric
 

Attachments

  • SONYQ67i.JPG
    SONYQ67i.JPG
    61.7 KB · Views: 1
Eric Lange said:
Can anyone explain how you get 100% green area coverage of the sensor, when the green photosites only populate 50% of the sensor?
Click to expand...

This is another bit of the marketing that I find a bit...questionable. You are right. Or at least you appear to be right. The green pixel that is present in every black-outlined box represents roughly 50% of that area. The red and the blue each represent 25%. Meaning for every green pixel there is 0.5 red and 0.5 blue. Or if we want integers, for ever 2 greens, there is 1 red, and 1 blue. Or 2:1:1.

Which is odd, because the same ratios can be easily recognized in bayer pattern sensors. It seems to me, what they are really saying is that for every black-outlined box there is one whole green pixel, and half a red and blue. But the actual ratio is the same. Which is a bit sketchy, in my opinion, because if you cut the bayer pattern in half spatially so that the pixels being compared were actually the same size (and the sensors the same resolution) - you'd also get one whole green per box, and half a red and half a blue.

Which leads me to my question - on the Q67 sensor, what is the black outlined box? Is it supposed to represent a single pixel? In other words, are they filtering each pixel with different colors? Or is it supposed to represent the relative size of the much lower resolution bayer they are comparing it against? Or is there just to illustrate how they are counting the contribution of each color towards the image? I don't get it - whenever I've seen sensor diagrams they don't have such boxes.
 
I'd check again Michael - it looks like Rot45 Bayer to me.

Graeme
 
We're all working through Sony's diagrams together to see what they're up to. No need for apologies - Sony are making confusing statements that are hard to follow. I'm betting that most people watching that will take away:

RGB = 100%, they'll get that.
Bayer = 33% (Most people don't really understand how Bayer works)
Q67 = 67%, the number is in the name, they don't understand it either, but they'll think it's 2x good over Bayer!

All I've been saying all along is that look at the pixel counts! It all boils down to "if you double or triple the pixel count you'll get a better resolution image" - like wow. That's earthshattering!

Graeme
 
Yup, it is identical to a rotated Bayer pattern, (at least in terms of layout); but the way its marketed they say standard Bayer is only 33% and Q67 is 67%. What they seem to try and mis-lead people by, at least in the presentation, is that if you have a rotated Bayer pattern that this brings about some sort of “magical packing factor” with a performance enhancement of 34%, … and a 100% green channel that’s just the same as a true RGB system!

I wonder if the head of marketing for this product would be able to answer the simple mathematical question posed in post 154 “Pompi’s Hall way”?

Either a vital piece of information is completely missing or we live in rather desperate times?





Maybe they have a green, brown and turquoise filter array!
 
You must log in or register to reply here.
Back
Top