Idea for Canon?

Status
Not open for further replies.
If Canon really wants to bounce back from this whole low ISO IQ and DR thing, couldn't they try to develop a new CCD sensor instead of CMOS?

Correct me if I'm wrong, and I very well might be, but doesn't CCD have a higher potential image quality? Hence why most MF cameras use them? Why doesn't canon make a low ISO monster with huge DR and MP using a CCD sensor. If they are suffering from being blocked by sony patents on how to add all that DR into a CMOS, maybe going back to CCD would help them get around that?

Am I correct in thinking that the only advantage CMOS has over CCD is that it is cheaper and more economical to produce? And current CMOS are so good only because that is what the R&D department has focused on? If they put as much effort into CCD what would have happened?

If I'm crazy suggesting this, feel free to let me know, but I figured it wouldn't hurt to throw that suggestion out there! I'm a photographer, not an engineer. So what do I know. (I first went to school for engineering, then switched over to photography cause photography makes me happy.) You guys on here seem to know your stuff, so, thoughts? :)
 
Canon, like other sensor makes, are constantly trying to improve. Trying and achieving are two different things, though. And achieving and marketing are two separate things on top of that. Canon might have the most bad-ass DR sensor that costs tens of thousands of dollars. Or, they don't give a damn about DR, because people buy their cameras anyway. Or, they just can't achieve the DR, despite several tries. Who knows? The only way they will deliver is when its customers demand it. Most Canon customers don't give a damn about DR.
 
Upvote 0
poias said:
Canon, like other sensor makes, are constantly trying to improve. Trying and achieving are two different things, though. And achieving and marketing are two separate things on top of that. Canon might have the most bad-ass DR sensor that costs tens of thousands of dollars. Or, they don't give a damn about DR, because people buy their cameras anyway. Or, they just can't achieve the DR, despite several tries. Who knows? The only way they will deliver is when its customers demand it. Most Canon customers don't give a damn about DR.

You may very well be correct. Then again, some of us DO care. Canon DOES create cameras for minorities. Example: MOST people don't want something as big and expensive as a 1dx. But, alas, Canon is making it (or so we are told, lets see if it ever gets released).

But anyway, its not only DR. A lot of people, rightly or wrongly, want more MP. Correct me if I'm wrong, but isn't it easier to cram more pixles in a CCD?
 
Upvote 0
poias said:
And MF sensors are decade old technology. Imagine if they have similar tech to modern FF CMOS sensors. They would have 18 stop DR and usable 1million ISO.

That would be a sight to behold!

But why on earth would a $20k digital back use such outdated tech. You would think they would improve it over the years? I'm not saying you are wrong, I'm just really curious as to why! If such a sensor existed, even I would buy one. I would take out a loan if I had to. So if that is true that they could have advanced MF tech but didn't, that seems like the biggest marketing oops of all time, no? They don't want to expand the tiny MF market into something resembling the 1dx/D4 market, which I assume is many times bigger?
 
Upvote 0
Tcapp said:
But why on earth would a $20k digital back use such outdated tech. You would think they would improve it over the years? I'm not saying you are wrong, I'm just really curious as to why! If such a sensor existed, even I would buy one. I would take out a loan if I had to. So if that is true that they could have advanced MF tech but didn't, that seems like the biggest marketing oops of all time, no? They don't want to expand the tiny MF market into something resembling the 1dx/D4 market, which I assume is many times bigger?

Fact is, Leica (S2), Pentax (P645), Leaf, Phase One, Hasselblad, probably put them all together and their sensor R&D budgets wouldn't go anywhere near that of Canon or Sony.
It's unfortunate, but MF technology is going to lag further and further behind, if not least because of their lower budgets. Take the latest sensors, the D800, D4, 5D3, and compare them on a *pixel level* (100%) to an MF sensor, half the time the 35mm outperforms the MF, and that's despite the individual pixels being a crudload smaller and closer together, and despite the MF sensors costing as much as a small house.

If Canon or Sony took what they'd learned from making APS-C and FF sensors, and put the same tech into an MF, that would be phenomenal. Take the low-light performance of the 5D3 and make those pixels twice the size again? Take the DR of the D800 and double or triple their full-well capacity and increase another 1 or 2 bits? Even take an MF sensor now and add gapless microlenses would bump up performance a stop or two.
I don't care how expensive that sensor would be, it would be phenomenal and leave even the IQ180 for dust.
 
Upvote 0
dr croubie said:
Tcapp said:
But why on earth would a $20k digital back use such outdated tech. You would think they would improve it over the years? I'm not saying you are wrong, I'm just really curious as to why! If such a sensor existed, even I would buy one. I would take out a loan if I had to. So if that is true that they could have advanced MF tech but didn't, that seems like the biggest marketing oops of all time, no? They don't want to expand the tiny MF market into something resembling the 1dx/D4 market, which I assume is many times bigger?

Fact is, Leica (S2), Pentax (P645), Leaf, Phase One, Hasselblad, probably put them all together and their sensor R&D budgets wouldn't go anywhere near that of Canon or Sony.
It's unfortunate, but MF technology is going to lag further and further behind, if not least because of their lower budgets. Take the latest sensors, the D800, D4, 5D3, and compare them on a *pixel level* (100%) to an MF sensor, half the time the 35mm outperforms the MF, and that's despite the individual pixels being a crudload smaller and closer together, and despite the MF sensors costing as much as a small house.

If Canon or Sony took what they'd learned from making APS-C and FF sensors, and put the same tech into an MF, that would be phenomenal. Take the low-light performance of the 5D3 and make those pixels twice the size again? Take the DR of the D800 and double or triple their full-well capacity and increase another 1 or 2 bits? Even take an MF sensor now and add gapless microlenses would bump up performance a stop or two.
I don't care how expensive that sensor would be, it would be phenomenal and leave even the IQ180 for dust.

Canon should make a 5d2 of the MF world. Something that is affordable enough for the masses, but super amazing. I don't see why they can't, but I know they wont.
 
Upvote 0
Tcapp said:
Canon should make a 5d2 of the MF world. Something that is affordable enough for the masses, but super amazing. I don't see why they can't, but I know they wont.

Isn't that the Pentax 645?
ok, $10,000 is debatable as to whether it's "affordable", but it's half the price of a Hasselblad, you get nearly 3 for the price of a Leica S2 and probably the whole range of P645 and lenses cost the same as an IQ180 back-only.
And it's still the same price as a D3X with a good lens.

I'd certainly like to see them do it too, it would mean a whole new range of lenses though, and they have trouble enough building the ones they've got. If they wanted to make a digital back that could mount onto a Hassy/Contax/Mamiya body that is even better but less likely, the 'MF for the masses' would follow the P645 and be non-interchangeable methinks...
 
Upvote 0
The Pentax 645 is about $10k, yes. So yeah you could buy 3 for the price of a leica S2.

The base hasselblad is just over $15k (I believe in a kit), and the phase one starts at about $13k in a kit - a case with a leaf 22mp back plus 80mm leaf shutter lens (plus odds and ends). Im not sure as to how much the actual phase one backs are, but I'd suggest the base model would be about the same as the mamiya backs too.

And um... the S2 is like $28k for the body alone, no lens, no nothing but obviously the 'back' is part of the body as opposed to the blad or phase.

Thats the prices here in AU at least. I do wonder if canon or nikon for that matter will release a real MF eventually. I'd be interested, as I am in the above. Just can't afford them... until they're a tax write off.
 
Upvote 0
What about a Forevon like sensor? I can see why Canon would not be interested in putting one in the 5D3 or the 1DX given the issues with FPS and video but if a new camera was specifically looking to target the high resolution market it seems like a good option to me.

Wasnt there some talk of a Forevon like patent from Canon last summer?
 
Upvote 0
That sensor has very poor high iso performance, and despite the claims of great color, from the tests I've ever seen the colors are pretty much awful. I don't know if it can handle video either, and i doubt canon or nikon/sony would be interested in coming to market without that feature. Sure it has more resolution than it's numbers might suggest, but i don't think anyone really would want to be in the position of explaining all that to customers.
I second the vote for the pentax 645 being the 5dmk2 for MF. That thing looks pretty awesome.
BTW, i think leica is looking at moving to a cmos for the S3.
 
Upvote 0
dr croubie said:
If Canon or Sony took what they'd learned from making APS-C and FF sensors, and put the same tech into an MF, that would be phenomenal. Take the low-light performance of the 5D3 and make those pixels twice the size again? Take the DR of the D800 and double or triple their full-well capacity and increase another 1 or 2 bits? Even take an MF sensor now and add gapless microlenses would bump up performance a stop or two.
I don't care how expensive that sensor would be, it would be phenomenal and leave even the IQ180 for dust.

If any of the manufacturers took the sensor tech from their compact cameras and enlarged it to 35mm full frame, it would blow the current sensors out of the water. It seems that a lot of the technology doesn't scale up cost effectively (like back-side illumination), at least for now.

Interestingly, there are hints that the Leica S3 may use a CMOS MF sensor:

http://leicarumors.com/2012/05/01/rumor-leica-s3-with-new-sensor-at-photokina.aspx/
 
Upvote 0
Tcapp said:
But why on earth would a $20k digital back use such outdated tech. You would think they would improve it over the years? I'm not saying you are wrong, I'm just really curious as to why! If such a sensor existed, even I would buy one. I would take out a loan if I had to. So if that is true that they could have advanced MF tech but didn't, that seems like the biggest marketing oops of all time, no? They don't want to expand the tiny MF market into something resembling the 1dx/D4 market, which I assume is many times bigger?

For the same reason FF sensors lag. The smaller the sensor, the faster the product cycle. The sensors that go into cameras and PaS cameras are generally a generation or two ahead of the crop and FF sensors.

Until they figure out a way to do tiling in a cost effective way, we are stuck with 'one big piece of silicone' sensors and are unlikely to break this cycle.
 
Upvote 0
Tcapp said:
poias said:
And MF sensors are decade old technology. Imagine if they have similar tech to modern FF CMOS sensors. They would have 18 stop DR and usable 1million ISO.

That would be a sight to behold!

But why on earth would a $20k digital back use such outdated tech. You would think they would improve it over the years? I'm not saying you are wrong, I'm just really curious as to why! If such a sensor existed, even I would buy one. I would take out a loan if I had to. So if that is true that they could have advanced MF tech but didn't, that seems like the biggest marketing oops of all time, no? They don't want to expand the tiny MF market into something resembling the 1dx/D4 market, which I assume is many times bigger?

For what it's worth, CMOS and CCD are practically the same age (1967 vs 1969). CMOS is also a bit more appropriate for pixel-based output in my opinion. Keep in mind, many OLD digital cameras used CCD and have since switched to CMOS. There must be some logical reason for that. Many modern video cameras use CMOS now as opposed to CCD (ah..the days where I would dream about having a 3CCD camera). Cost is a factor in this, but so is complexity of design. I believe many MF backs would still be based on CCD because it would be insanely difficult to manufacture a CMOS sensor at those sizes (I'm sill impressed that we manage to make so many FF CMOS sensors).
 
Upvote 0
This seems like a strange thought.

Canon. Nikon, and Sony have switched to CMOS over the last few years because it is better. Sony and Nikon could not compete with Canon's large CMOS based sensors, so they switched.

Do a little research.
 
Upvote 0
I don´t agree in "better".
CMOS advantage are the production costs!

it requires not that much energy like CCD
and the high ISO performance is better.

but color deph and DR, sharpness are definately better at CCDs
 
Upvote 0
Status
Not open for further replies.