Rangefinderforum.com - View Single Post

CJP6008 · 11-01-2005

An interesting debate, the issue re full frame sensors is a vexed one. Irwin Puts write intelligently and with some authority on the subject (I suspect he knows more about these things than the rest of us put together!) See http://www.imx.nl/photosite/comments/c014.html and http://www.imx.nl/photosite/comments/c011.html. I recommend a read of what he has to say - he is a good myth debunker and there are plenty surrounding this subject.

Re Zeiss's attitude to digital - another reading of what they say in the brochure is that they are waiting for a leap forward in sensor technology, more than just a fiddling with the current sensor technology. I am no expert in these matters, but I know a man who is - he makes thermal imaging cameras and they use the sort of chips now current in digital cameras, let's call them CCDs for ease of reference (in fact that is what CCDs were originally invented for, sort of, by (among others) my mum's cousin's hubby - brain the size of a planet! Anyway the thermal imaging guy explained how they work - boy it's crude. They are in essence transistors with the tops shaved off. To get anything vaguely meaningful out of them one has to put a load of filters in front. What they produce is pretty much electonic gibberish. The clever bits are the programs that unscramble it all. They do an amazing job considering the rubbish they start with (I am told). It may be easier to explain in analogue terms - image a system that produced negs so bad that you could hardly see what it was - a complete mess, like it had been shot through heavily frosted glass. The only way you could get anything out of it was to scan it and employ some specially designed software. In essence that is what goes on in a digital camera.

Good as those programmes are, there are limits - you cannot make a silk purse...

The foveon chip is a radical departure from the CCD idea - controlling it is touch though. My relative wondered (when it first came out) if they would ever manage to get the best from it for that reason. It took them nearly 30 years to get to grips with CCDs!

Perhaps Zeiss have in mind some fundamental leap forward in the direction of a new sort of chip entirely. After all, if what these guys tell me is true (and I have no reason to doubt them) building CCDs with load more pixels or whatever is not a quantum leap forward - a medium format neg shot through frosted glass to follow my earlier analogy. Something that sees more clearly is needed. After all you cannot polish a turd!

11-01-2005	#22
CJP6008 Registered User CJP6008 is offline Join Date: Sep 2005 Location: Guildford UK Posts: 119	An interesting debate, the issue re full frame sensors is a vexed one. Irwin Puts write intelligently and with some authority on the subject (I suspect he knows more about these things than the rest of us put together!) See http://www.imx.nl/photosite/comments/c014.html and http://www.imx.nl/photosite/comments/c011.html. I recommend a read of what he has to say - he is a good myth debunker and there are plenty surrounding this subject. Re Zeiss's attitude to digital - another reading of what they say in the brochure is that they are waiting for a leap forward in sensor technology, more than just a fiddling with the current sensor technology. I am no expert in these matters, but I know a man who is - he makes thermal imaging cameras and they use the sort of chips now current in digital cameras, let's call them CCDs for ease of reference (in fact that is what CCDs were originally invented for, sort of, by (among others) my mum's cousin's hubby - brain the size of a planet! Anyway the thermal imaging guy explained how they work - boy it's crude. They are in essence transistors with the tops shaved off. To get anything vaguely meaningful out of them one has to put a load of filters in front. What they produce is pretty much electonic gibberish. The clever bits are the programs that unscramble it all. They do an amazing job considering the rubbish they start with (I am told). It may be easier to explain in analogue terms - image a system that produced negs so bad that you could hardly see what it was - a complete mess, like it had been shot through heavily frosted glass. The only way you could get anything out of it was to scan it and employ some specially designed software. In essence that is what goes on in a digital camera. Good as those programmes are, there are limits - you cannot make a silk purse... The foveon chip is a radical departure from the CCD idea - controlling it is touch though. My relative wondered (when it first came out) if they would ever manage to get the best from it for that reason. It took them nearly 30 years to get to grips with CCDs! Perhaps Zeiss have in mind some fundamental leap forward in the direction of a new sort of chip entirely. After all, if what these guys tell me is true (and I have no reason to doubt them) building CCDs with load more pixels or whatever is not a quantum leap forward - a medium format neg shot through frosted glass to follow my earlier analogy. Something that sees more clearly is needed. After all you cannot polish a turd! Last edited by CJP6008 : 11-01-2005 at 14:12.