10 replies to this topic

[james]

Check out the official website:

http://www.four-thir...om/index_01.htm

Comments?

Cheers
James

[MikeO]

Interesting read. Not ready to give up my Canon L lenses just yet, though. It is of particular note that there are only a few existing lenses for this system and Oly doesn't have a wide angle lens in the initial set shown at PMA.

[scorpio_fish]

I used to think the whole concept would be still born. Maybe I'm wrong. This is definitely being led by Olympus, the company that abandoned the 35mm film market. I still have my 20 year old Olympus gear.

How excited are Kodak and Fuji now that they have the 14n and S2, respectively?

Where will the price point be compared to the current crop of D-SLR's?

Will it appeal to only those who currently own no Nikon or Canon glass? Will it only be for consumer upgraders?

What about the glass? How long will it take to develop a complete line of glass? Will all three players make complete lines of competing and interchangeable lenses? If so, somebody is going to win and somebody is going to lose. Imagine, an open industry standard which can be good for the consumer! But how open will it be? Just those three? Or will we see 4/3 mount lens from Tamron, Tokina, Sigma, Vivitar, Nikon and Canon?

I'd like to see the business plan on this one. I'm sure it looked much better a year ago. We now have a Canon 10d at $1500. Are the 35mm based cameras that bad? Are its shortcomings that bad? I don't think so.

Me thinks the bodies need to come in at $600-800 to get any attention. So the lenses will be smaller, but will the price be that much smaller?

[MikeO]

You're right. Though Olympus has a great argument for the purist in that the lens really does need to be designed for the sensor, as long as a "focal length multiplier" is acceptable, the 35mm format cameras and the same old lenses will be fine. After all, the 4/3 sensor is still quite a bit smaller than the D10's and D100's and the S2's. However I think the manufacturers are finally starting to realize that the "full frame" 35mm sensor is a pipe dream as long as you're using the same lenses you used for film. Maybe there's a way to bend the light one more time in the camera or process the image a different way . . .

Mike

[tshepherd]

Ironically enough, in manufacturing, smaller means more expensive. The smaller the product, the tighter the tolerances, so the higher the cost. Obviously that's not 100% applicable, because we're not talking about lenses that are an order of magnitude smaller, but I think you're right, they may be smaller, but I doubt they'll all that much cheaper.

I do really believe that this idea will be more targeted at non-traditional SLR users, not those of us who are already there. If the bodies and lenses come in at a reasonable price, they'll be attractive to consumers who would normally buy cameras like the Coolpix 5700. It's very hard to guess at what the potential market might be though, especially as we get entries into the market like the 10d at it's lowered price.

[james]

OK, here comes a pretty strong opinion:

Canon and Nikon are going to make 4/3 a hard row to hoe.

Canon just announced a beautiful DSLR at $1,499 and have 10,000 of them ready to ship. So if you don't have any lenses, would you buy an E20 or its 4/3 big brother when you can buy a Canon 10D instead?

Nikon announced a line of DX lenses specially designed for their DSLR's so the argument that only the 4/3 system lens/sensor are matched and (therefore superior) has been squashed by Nikon. Not only that, but a Nikon 12-24DX lens is already being manufacturered and people are ordering it. Not to mention that it has an incredibly wide end of 18mm (Equivalent) and the 4/3 lenses only go out to 28mm equivalent.

The biggest problem for me though is the sensor. Here is a graphic from the 4/3 website:



As you can see, the sensor is roughly 1/2 the size of "full frame" 35mm format. What that means in practical terms is that in order to fit in the same number of pixels on the 4/3 sensor, the pixels will need to be 1/2 as small. Any time you shrink the pixels, because of straight up physics you lose dynamic range and noise goes up. Again, just due to plain old wave mechanics - this can never be an advantage.

On the other hand, smaller sensors actually ARE cheaper to make. The size of the sensor is smaller, so a number of them can be made from the same wafer perhaps - whereas only one FF sensor or even a part of an FF sensor can be made from the same wafer. Other advantages are that the cameras can be smaller, but WILL they? The prototype Oly shown looks about the same size or bigger than the E20 - which is about the same size as the D100 and Canon 10D.

My opinion only,
James

[bobjarman]

Agreed James, plus the simple fact is that Canon on Nikon have huge market share advantages in the SLR market and would never agree for that reason alone. Without commenting on the merits of the format, to me this is really a clever marketing idea by companies who want a piece of the pie from the big 2.

I cannot imagine under any circumstances that canon and nikon would agree to a universal lens mount. As good an idea as it is, they have way to much invested and way to big a capture of the market to do that voluntarily.

The only way I could see tis happenning is if the new format came out and started digging into the market share. And even then I would imagine each would keep a proprietary lens mount.

I would!

[wetpixel]

As you can see, the sensor is roughly 1/2 the size of "full frame" 35mm format. What that means in practical terms is that in order to fit in the same number of pixels on the 4/3 sensor, the pixels will need to be 1/2 as small. Any time you shrink the pixels, because of straight up physics you lose dynamic range and noise goes up. Again, just due to plain old wave mechanics - this can never be an advantage.

Isn't 1/4 the size? Or do we speak in squared-lingo when talking about areas?

[MikeO]

Or do we speak in squared-lingo when talking about areas?

Squared? I thought it was a footnote!

(Math vs. English major joke, sorry . . .)

Mike

[james]

Ya got me there Eric!

Ok, so for the FF sensor, let's assume that the pixels are 8 microns by 8 microns. Then according to the graphic, the 4/3's sensor's pixels would be 4 microns by 4 microns.

So the surface area to "detect photons" is roughly 1/2^2 or 1/4 the surface area.

Cheers
James

[wetpixel]

Increased noise in denser sensors (har har har) is well documented. That's one reason our D60s produce such buttery-smooth images vs. 5MP consumer cameras that produce images that are almost the same resolution.