Do you remember the Titanium bladed shutter Nikon used in its Nikkormat line? … Was that close to being perfect, or no? … The cloth shutter used in the other Nikon cameras, has limited use, apparently. The search is on for something better. The electronic shutter used in many cameras, today - solved the quietness thing - but brought with it - a host of new problems. What kind of shutter do we need, to solve all hitches. Is there such a thing? … If so, is there a way to make it affordable enough, so it can be incorporated in all dig cams? … Is there a future, for such a device in today's DSLRs and MILCs, or is getting there still a fantasy?

The titanium shutter bladed shutter had one huge advantage, war correspondents had a shutter curt an that did not rot out in the Viet Nam jungles. BUT, neither did the rubberized shutter curtains in Leica cameras. But as always you are asking the wrong questions. Is there a shutter that is the ideal shutter design for a focal plane shutter? The answer is yes there is. The design was first introduced in 1929, then it was perfected in early 1932 when it went into production in all Leica cameras.

Yes, you should shake your head, this is pretty pathetic. When in 1969 I visited the factory in Wetzlar I asked a stupidly simple question to the presenters, "Does Leica make the best lenses of all camera makers?" I received a smile and was told that Nikon made some superior optics, better than Leitz. Then came the great truth, "If you want the results these lenses are capable of you need to put them on a Leica camera body to get the best results. Their camera bodies are well made but the first shutter curtain slams into the shutter stop creating vibrations that remove the quality of the image."

So the answer to your question is to do what E. Leitz does, look at the total camera then determine if the WHOLE camera system is providing excellence in your images.

A zillion focal points won't solve the global shutter problem. At issue here is how to store the image on the sensor long enough that it can offload that information to the camera.

With a mechcanical shutter, the shutter exposes the sensor to the intended image and then closes. The shutter is back into the dark and what is captured on the sensor remains there while that information is sent to the camera microprocessor.

With global shutter, the sensor is always exposed to the light coming in, but we want to expose it and hold it before sending to the camera microprocessor. As current sensors now stand, they have to spread out the time due to the speed at which the sensor can be read. The Sony A9 has improved the readout speed and has come a long ways toward achieving what is wanted. But even this is not where we would like it.

So what would be ideal? A way to store the entire image on the sensor would be ideal. But what would that take to do it. For a 24 Mpixel sensor, it would take 24 M ADC's to convert the pixel contents into a digital voltage. And then storage to hold 24 M worth of information. But this is an insane amount of circuitry to add to a BSI sensor. The Sony A9 ended up using a compromise solution where not as much circuitry was added, but the effective transfer rate was improved a lot. I fotget the actual figures, but the amount of time that it takes the A9 sensor to move that data to the microprocessor is many times faster than other sensors. But that speed came at a cost of having a much more complicated sensor with lots of supporting circuitry to make it happen.

More and more sensors will take this route. But complexity does add cost. At least initially.

ALL Leica camera bodies from 1932 on have the curtain arrested by a device much like the throw out bearings that adjust the brake system in modern vehicles (similar design, not the same). In a Leica camera body the first curtain does NOT stop, rather it winds around the drum to arrest the motion of that shutter, then reverses to wind back to the rest position. That first curtain can be herd doing this if one listens careful to the back of the camera, the sound is a swishing sound like ball bearings sliding along a trough.

Other cameras with focal plane shutters stop the first curtain abruptly. This is where the great amount of vibration is created causing loos of fine sharpness.

This is not a far fetched idea, the German engineers realized that the main problem was in the first curtain's sudden stopping and so solved the problem with a wind up curtain.

ALL Leica camera bodies from 1932 on have the curtain arrested by a device much like the throw out bearings that adjust the brake system in modern vehicles (similar design, not the same). In a Leica camera body the first curtain does NOT stop, rather it winds around the drum to arrest the motion of that shutter, then reverses to wind back to the rest position. That first curtain can be herd doing this if one listens careful to the back of the camera, the sound is a swishing sound like ball bearings sliding along a trough.

Other cameras with focal plane shutters stop the first curtain abruptly. This is where the great amount of vibration is created causing loos of fine sharpness.

This is not a far fetched idea, the German engineers realized that the main problem was in the first curtain's sudden stopping and so solved the problem with a wind up curtain.

A zillion focal points won't solve the global shutter problem. At issue here is how to store the image on the sensor long enough that it can offload that information to the camera.

With a mechcanical shutter, the shutter exposes the sensor to the intended image and then closes. The shutter is back into the dark and what is captured on the sensor remains there while that information is sent to the camera microprocessor.

With global shutter, the sensor is always exposed to the light coming in, but we want to expose it and hold it before sending to the camera microprocessor. As current sensors now stand, they have to spread out the time due to the speed at which the sensor can be read. The Sony A9 has improved the readout speed and has come a long ways toward achieving what is wanted. But even this is not where we would like it.

So what would be ideal? A way to store the entire image on the sensor would be ideal. But what would that take to do it. For a 24 Mpixel sensor, it would take 24 M ADC's to convert the pixel contents into a digital voltage. And then storage to hold 24 M worth of information. But this is an insane amount of circuitry to add to a BSI sensor. The Sony A9 ended up using a compromise solution where not as much circuitry was added, but the effective transfer rate was improved a lot. I fotget the actual figures, but the amount of time that it takes the A9 sensor to move that data to the microprocessor is many times faster than other sensors. But that speed came at a cost of having a much more complicated sensor with lots of supporting circuitry to make it happen.

More and more sensors will take this route. But complexity does add cost. At least initially.