Multi-shot on future Alpha camera ... Why not ?

Discussion of all digital SLR cameras under the Minolta and Konica Minolta brands
Forum rules
No more than three images or three external links allowed in any post or reply. Please trim quotations and do not include images in quotes unless essential.
SilviuV
Acolyte
Posts: 5
Joined: Wed Aug 27, 2008 7:44 pm
Contact:

Multi-shot on future Alpha camera ... Why not ?

Unread post by SilviuV »

I'm new to this forum. Actualy this is my first post here and I would like to learn, discuss and share ideeas. I apologize for my future spelling erorrs.

Here's the ideea: a while ago, I saw on a Hasselblad (can't remember exact the model) that they had something called multi-shot. What that meant was thaking the same image but with shifting the senzor 1 pixel at a time to determine de exact amount of light in the RGB spectrum. That actualy takes 4 images, 3 for the R, G and B and another for exposure (or another parameter, corect me if I am wrong). This multi shot offers the same color accuracy per pixel as a Foveon X3 sensor. We all know that the image we get by thaking a photo is now the result of post-processing of the Bayer arrangemet.
I was also very impressed by Sigma's Foven X3 4.7mp sharpness at this resolution. Upscaled to 10 mp it was as good as any Bayer 10mp image.
I have seen on the Hassleblad homepage a side-by-side comparison of the two modes, and the multi-shot was waaaaay better than the single-shot: imagine that on a 30-something mp Hasselblad :).

This will not only give the photographer better color accuracy, but unparalelled per pixel sharpness. I don't know how that Hassleblad worked in that mode, if it required four times the exposure time or if it was forced to use a tripod to get the best result (this frankly is best in any case). Or if it can work with SSS on.

So WHY NOT ? I'm not saying that it should be implemented in the A900. I'm saying why not in future generations. They allready have the potential of the SSS to shift the sensor. It might be a in camera software update (i'm a hopeless dreamer) or future in-camera design criteria.

I want to hear what you think about this ideea ... and maybe one day Sony will take it in consideration.

P.S. And another thing ... dust. Why not implement something like Olympus's ultra-sonic dust removal ? It is THE ONLY WAY to the dust problem.
Proud architecture student, A100 SAL 18-70mm, Hoya Pro1 Digital CR-PL, ExpoDisk Neutral, Lowepro Daypack Backpack.
David Kilpatrick
Site Admin
Posts: 5985
Joined: Sat May 19, 2007 1:14 pm
Location: Kelso, Scotland
Contact:

Re: Multi-shot on future Alpha camera ... Why not ?

Unread post by David Kilpatrick »

The idea of shifting the sensor 1 pixel in four captures, or 1/2 pixel with two captures, was raised by many journalists including me when the KM 7D was launched. Despite Peter Ganzel's comments on Dyxum about the magnetic centering device and the accuracy of the system, my own tests have shown that the AS/SSS system is just not capable of repeatably positioning the sensor even at rest to within 1 pixel. The system is affected by gravity (the orientation of the camera) and by random small errors. AS/SSS shake correction is a dynamic process, not a matter of calculating positions to within 1/200th of a mm.

While the AS/SSS carriage and piezo mechanism could perhaps be adapted to do the same as Hasselblad multishot (which uses much finer pitched fixed expansion piezo shift) the existing design would not be accurate enough. I agree with your comments on the Foveon system and I have used all the SD series digital cameras - we own an entire set between my company and Richard (son) privately, and they sometimes come back to me, sometimes to him. The problem is that even the SD14 is not enough. The same density of pixels across full frame would be amazing though.

David
SilviuV
Acolyte
Posts: 5
Joined: Wed Aug 27, 2008 7:44 pm
Contact:

Re: Multi-shot on future Alpha camera ... Why not ?

Unread post by SilviuV »

Ok. So it will not work with the curent version of SSS.

In any case, I would consider this to be a very big improvement in image quality and must be implemented. Maybe the market demands the LV, video mode, bigger-better LCD, high frame rate, high ISO performance, many MP, but the final image will not be that much improved. I find myself shooting near base ISO, mostly at no more than 1 frame every 5 seconds with the viewfinder in apx. 80% of the time. It's like a car: we often drive at 70 mph on straight roads and don't need much more than that to reach the destination.

It will bring better DR and color redention.

A image taken with multi-shot could be easily up-scaled 2 or 3 times bigger. The 25mp FF sensor could posibly make usable 70 mp images. It would shatter all competition and bring the FF and APS-C near to the MF territory. Multi-shot will be a real break-through, even if it only comes to the benefit of landscape, studio, macro photography by beeing tied to a tripod.

I hope ... :)

Thank you !
Proud architecture student, A100 SAL 18-70mm, Hoya Pro1 Digital CR-PL, ExpoDisk Neutral, Lowepro Daypack Backpack.
Post Reply

Who is online

Users browsing this forum: No registered users and 8 guests