4/3

I've learned quite a bit reading the posts on the E-Volt, including

when to keep my big mouth shut. ;-)

 

So, what do the Olympus fans and general small sensor gurus think the

long term prognosis is for the 4/3 system? For example, in five years,

will the 4/3 cameras be capable of the same high ISO image quality as

the 1.6 crop Canons? How might they compare with the full frame Canons?

 

What technology will be coming along that will enable Olympus and the

other 4/3 players to compete against the larger sensors with their

large photosites?

 

Is it true that larger photosites will always have an advantage in low

noise? Or is there some dramatic change in the works that will rewrite

the rules?

 

Thanks,

I guess I find it hard to believe that larger sensors won't always have an advantage over

smaller sensors. Unless you need a smaller sensor to build a smaller camera, why bother?

Larger sensors can and will use any technology that smaller sensors can, so how can smaller

sensors compete?

It's my guess that in 5 years, the 4/3 digital system will be about as popular as APS film is today.

Keep in mind that large sensors will ALWAYS cost more than small sensors, by roughly the

square of the difference in size. This is simply a matter of manufacturing yield, and has

been true from the first planar transistor.

 

It is more likely that small sensor performance will increase than large sensor relative price

will decrease.

 

Of course, if the general decrease in price is enough, large sensors will win -- who cares

about the difference in price between a $2 or $4 sensor? But for now, it's more like the

difference between $200 and $400.

The Olympus 4/3 format has two problems: the 4/3 format and Olympus. Smaller photosites will always have less light falling on them, so their output must be amplified more in low light, and will yield more noise. Whatever advances are made in separating signals from noise will be available to cameras with larger sensors as well.

 

As was pointed out in another thread, the fact that Olympus abandoned its excellent OM 35mm system, and its users, makes many photographers reluctant to commit to an Olympus system.

 

Olympus tried to get other companies involved in 4/3, but has essentially failed to do that.

Some people wax eloquent about their love for 2/3. Others do for square. Since the 5D has optics and electronics adequate to 3/3 as well as to 2/3, would you buy 2/3 if Canon decided to make the bigger chip?

Hektor,

 

I don't understand that. Canon dumped steaming piles of manure on all of their FD users in the late 1980s, and nobody but me and a handfull of others said more than "Oh Gee" in response. Since giving us the finger, Canon's fortunes have grown tenthousandfold. They've gone from being Nikon's water carrier to the top company in the business. There has got to be more to Olympus's misfortunes than screwing their loyal customer base. Canon has proved to all concerned that you can piss on your customers and they'll love you for it.

Yes, after all, camera manufacturers have a duty to not change their lens mounts ever.

 

Pfft.

The difference is Olympus has only proven it can be a volume seller of first 35mm point & shoots and now digital point & shoots while Canon, having moved to the EOS mount has grown to be the top seller of digital SLR's.

 

The fact Olympus dropped the manual focus line is not as important as they couldn't make it in the 35mm AF SLR market so they had no mount with a set of lenses they could adapt a digital SLR system to. As far as the 4 or 5 major makers go, Olympus had the all-time loser 35mm AF SLR system, so everyone wanting into their digital SLR system had to start all over, and two years ago when I was looking to buy my first DSLR the Olympus 4/3's system with only the E-1 body was not inexpensive. Today they have some good prices, but today is too late as far as swaying most who have now picked one of the other major brands, not just Canon.

The difference is Olympus has only proven it can be a volume seller of first 35mm point & shoots and now digital point & shoots while Canon, having moved to the EOS mount has grown to be the top seller of digital SLR's.

 

The fact Olympus dropped the manual focus line is not as important as they couldn't make it in the 35mm AF SLR market so they had no mount with a set of lenses they could adapt a digital SLR system to. As far as the 4 or 5 major makers go, Olympus had the all-time loser 35mm AF SLR system, so everyone wanting into their digital SLR system had to start all over, and two years ago when I was looking to buy my first DSLR the Olympus 4/3's system with only the E-1 body was not inexpensive. Today they have some good prices, but today is too late as far as swaying most who have now picked one of the other major brands, not just Canon.

Hi Peter, good to see you from the other thread... :^) I don't have a crystal ball, so I won't venture an opinion on the future of any particular system, but I thought I might elaborate on the statement I made "over there" about one of the more interesting evolving technologies out there. There are surely others that exist and are probably at least as interesting and relevant. The one I mentioned is abbreviated MEMS = Micro-Electro-Mechanical Systems. One link that talks about it further is http://www.memsnet.org/mems/what-is.html. Basically, it is a collection of technologies which concentrate on mass-fabricating extremely small mechanical systems using IC fabrication type processes. It is possible to make nano-level motors and mechanical switches, all mounted on silicon substrate. There is a lot of interest out there in this sort of technology, but here's how I think it might apply to our parochial interest - cameras. Keep in mind that I'm free-wheeling here, so there is probably a show-stopper of which I'm unaware.

 

Imagine if you will a CCD or CMOS imaging sensor that is covered by, in addition to an anti-aliasing filter (as most sensors used by cameras are these days), a large collection of small, independently controllable, light-tight "flap" that is mounted over each photosite. With the appropriate control circuitry, it should be possible to expose each photosite for a different amount of time! So, areas of shadow would be exposed for longer times, while bright areas could be exposed for much shorter times. This would mean that you might strive on each "picture" or "exposure" to make sure that the resulting image from the CCD is a uniform field. If you measure the times that each individual pixel is exposed, those times should be directly but inversely proportional to the amount of light that strikes each photo site. These times could then be translated to an exposure value for each pixel. The example I've just given would only work for black-and-white, of course, but I suspect that a Bayer type filter could be made to work with such an arrangement. Anyway, this type of sensor might be interesting, because it effectively does away with the noise issue. Each CCD site integrates to a level that far exceeds the noise level. The tradeoff here is that since each photo-site is exposed for a different amount of time, subject motion will appear differently in each pixel. This might be a show-stopper. On the other hand, there is software even now freely available that attempts to deconvolve subject motion or the effect of camera shake thereby "sharpening" a photo. These sorts of algorithms might be of use in a future sensor I described. Right now, the processing power to accomplish this is far too much to live in a camera body, but that will not always be the case.

 

Like I said, free-wheeling here...so it's probably a lousy idea...but interesting all the same. I would not be surprised to hear that one of a number of camera companies are looking closely at these various technologies in order to perfect manufacturing to drive costs to a level suitable for consumer markets.

 

That said, bigger silicon area = lower noise. That's physics and it won't change. But I suggest from my above comments that there are plenty of interesting games that could be played with significantly increased processing power and novel technologies. So who knows what the future might bring?

I am new to the whole digital world so take this with a pinch of salt.

I have just printed two 8X10's taken with my E1 & they are wonderfull.

Why so many photographers argue crop factors beats me.Why the fuss on sensor size? Does everyone print to poster size? Does everyone need high ISO capability? I have made photos for many years & have learned to live within the limits of all of my equipment; which consists of a lot of Nikon film gear. My refurbished E1 & 14-54 lens plus a mint used 50-200 (both excellent lenses) cost me $1740 USD. I shudder to think what a Nikon or Canon ,similar, outfit would cost.Had I purchased similar Nikon or Canon I would have to hire a Sherpa to lug around the equipment for me.Maybe Olympus will make it & maybe not; but that will not stop the E1 from being an excellent image maker.

<i>It's my guess that in 5 years, the 4/3 digital system will be about as popular as APS film is today.

</i>

<p>

You really think the popularity is going to increase that much?

Peter, there isn't a single major 35mm camera maker that still exists that hasn't changed

their mount. m42 became K, L39 became M, Minolta's mount went Maxxum, OM died, and

Nikon went from F to G.

 

Canon had the balls to come out and do it in the open, while Nikon did it behind your back.

Go ahead - mount your new G lens on your Nikon F. I hope you enjoy shooting at f/22.

Yes, Melvin, you have a couple nice zooms there.

 

What if you need a tilt shift lens, or a fast prime, or an inexpensive 300mm lens? The

Olympus cameras and lenses are nice, but they aren't exactly a system yet. Despite what

Olympus tells you, their 300mm lens isn't a 600mm lens, and shouldn't be priced as one.

It's as sensible as Canon charging $6000 for a 300 f/2.8 IF you have one of their cropped

DSLRs, but only $3500 if you have a full frame DSLR or a film camera. I'm sure Olympus is

losing money on those 300mm lenses, as they certainly can't sell enough of them to pay

for the tooling and design work.

 

This is what keeps a lot of buyers away from Olympus. People who are happy with f/

2.8-3.5 'pro' lenses which give insane DOF wide open (a normal 25mm at f/2.8-3.5 has

very deep DOF) and folks who are more than willing to shell out twice the market price for

telephoto primes (price out the 150mm f/2 against a Canon 135mm f/2 L, or their 300mm

against a Canon or Nikon equivalent) are sure to be content with Olympus, but those of us

who saw them disappear from the SLR market and abandon their beautiful and excellent

line of OM cameras know that, despite how nice or wide their system becomes, Olympus

may pull the same stunt again. Hell, after their terrible performance the first half of this

year, and seeing the kind of bullshit products they pull out like Ferrari cameras with

engine noises or digital camera / mp3 player hybrid devices (which are mediocre at each

of their functions) Olympus seems to be worse off than ever.

 

Some senile old Japanese guy is at the helm of this Titanic, and he thinks that the iceberg

is actually smooth glassy water!

Peter, the 4/3rd-line does not compete with neither Canon or Nikon over sensor size, nor resolution (at the moment). Olympus has however, created capable cameras with no similar features: a dust proof, splash proof design in a small package-- C.A. NONE existent, a superwide zoom (14-28 eq.) and a stellar telephoto (100-400) all for more affordable prices. You don't have to like it, and you don't have to buy it, but you can do MORE with an EVOLT kit with two lenses ($750 now) than with any other camera in this price range (or actually, even for $500 more).

 

Andrew, if Leica can sell a 50mm for $3200, Olympus can sell a 300mm for $6000. Who cares how they price their lenses? are you an E1 user?

 

Melvin, people (and mostly forums lurkers) enjoy scrutinizing over pixel size, sensor format, quantum physics, and ISO performance alike. This is all good and fine, but has no real link to actual photography. If your tool serves your purpose, be it a pinhole or 45mpixel digital back, then you got it made. I have owned the E1, and use both Canon and Nikon. I use medium and large format, film and otherwise, and every one has its purpose. The E1 has superb lenses, a dust shaker, an INCREDIBLE interface in a small robust size. Regardless of its sensor size and high-ISO performance-- it beats all Nikon and Canon DSLRs in the aspects I mentioned above. Those who contest this have never held it or used it.

 

I don't know what would happen in 5 years from now. Maybe Nikon would by then ship their D3X? Perhaps Canon would switch to a new lens mount to accomodate sensors LARGER than the 35mm format? Perhaps the brick-bodies we're so used to would be replaced by cellphones? Years ago, everyone KNEW that the Earth is flat, that the universe revolves around us, and that People can't fly.

What does it matter? because very few will be using the camera they have today in three to five years time. With annual new models with developments the idea of keeping a digital camera like my possibly 100year old Thornton Pickard is rather remote :-) or my 45yo Rolleiflex.

Peter, I did some calculations the other day. Basicaly, if you cut a 4/3s size sensor out of a 20D sensor, you end up with a 5.9MP sensor with the same noise quality. Do the same with a D2X sensor and you end up with a respectable 7.8MP.

 

So I would say that with improvements, a low noise 10MP is certainly possible on the format, I guess it just depends on who makes the chips.

 

The future of the system depends, I guess, how well the "E-3" does when it comes out, together with it's pair of f/2 zooms. My guess is it will be an 8MP model, but with more speed and also a much improved sensor over that found in the E-300, otherwise they probably wouldn't have waited this long.

 

For a long time, full-frame will be better, it's simply physics. But there will come a time when better becomes a moot point. Is anyone really going to have arguments like these if the choice is between noise-free ISO 32,000 and ISO 128,000?

 

In any case, I think the popularity will be higher than "what APS is today" simply because APS sucked and never had any good support in the SLR world and for consumers the cameras were more expensive, as was film and processing. I believe there will be a fair amount of people that, like me, prefer the lower cost and smaller package. I mean, Leica are still in business with just a niche following, aren't they? And they are actualy more expensive than any other 35mm brand!

Bas, did you know that Nikon, Canon, and Minolta all had APS SLR systems? The Canon and

Nikon APS SLRs took the same lenses as their 35mm big brothers and the Minolta had its

own system of splashproof APS SLR(s?) and lenses! Contax and Leica produced luxury high

quality APS cameras too.

I feel that it will continue to be a niche market, supported only by a few Olympus enthusiasts.

 

The 'standard' 2:3 format has never really been standard at all, or at least not among pro photographers. Basically pros have used 10x8", 4x5" 6x7cm or 6x4.5cm - all much the same ratio. As far as I can remember 2:3 has only appeared as 6x9cm and has never really been popular. Let's face it, the shape doesn't really suit either standard printing papers or most brochure shape requirements and needs to be cropped, reducing image quality. It was Oscar Barnack who made the 2:3 shape popular, and he used it because of convenience. 35mm cameras then became almost standard for amateur photographers and the likes of Canon and Nikon then started describing some of their models as 'professional'

 

 

It seems to me that the 4:3 format may be much better than 3:2 for most people, but that it will never catch on. A pity really, because Olympus make good cameras with superb lenses - but it seems that no other major manufacturers want to play with their toys. I feel pretty confident that if the mighty Canon had introduced this format it would have been accepted by many people.

Andrew, there is more wrong than with lack of bodies. Yes, there were a few SLRs but no "EF-S/DX" lenses, which the system would have needed badly. Add to that that APS didn't support slide film; in fact, I don't think there were any professional emulsions at all. Plus it also didn't suit custom/hand printing. So there wasn't much point in it for professionals or even enthusiast, and we all know that when it comes to SLRs, "the people" take cue from what pro users or their entusiast friends use. On top of that, it was more expensive.

 

And when it came to compacts, it was more expensive, which doesn't sell very well when you don't have much extra to offer over 35mm.

 

4/3s is cheaper and has none of the other drawbacks of APS. (mostly because, well, it's digital and doesn't need film)

Bas,

 

So perhaps Olympus would have a better camera if they didn't push the resolution so high? If they aimed for a 6MP sensor using the equivalent of Canon's 20D sensor technology, they could have a fine camera for making 8x10s?

 

For many pros and most enthusaiats, that would be plenty.

A bit more:

 

We've always accepted that any given format has an upper limit on how large a print you can make at a gioven quality. 4x5 prints larger than MF. MF prints larger than 35mm.

 

So if 4/3 users accept that they'll never print as large as 35mm, they can have a more compact and lighter system?

 

Of course the lenses can be smaller and lighter, but the camera really can't, since so much of the camera is fixed in size, being the user interface. The grip doesn't get smaller simply because the sensor is smaller. The LCD is what it is, not due to sensor size, but the ease of reading the information on it, etc.

You Olympus amatuers talking about the future of professional dSLR's is kind of like a bunch a white guys watching NFL football on a Sunday afternoon and yelling at the coach after a controversial play like they know something better. :-)

 

The reason that 4/3 and APS based cameras exist in the first place is to fit the budget needs of the remaining view_finder_obsessed photographers. The only reason Olympus went with 4/3 was simply an engineering mandate to produce a cheaper sensor than being used in Nikon/Canon dSLRs while keeping quality between Canon/Nikon and the more advanced digicam market. Hell, if I worked at Olympus I'd likely come up with the same plan. I made a joke previously about Olympus challenging the Canon 5D by taking light cone efficiency to it's logical limit and dumping 4/3 in favor of a square 3.5/3.5 sensor. Heck, I woulnd't be surprised if they made a round sensor.

 

You guys don't want to believe it, but the APS and 4/3 market segment has already been signed a death warrant by the rapidly improving quality of the digicam market and evolutionary apex of the dSLR. With increasing financial pressure from China and professional photographers reaching their saturation point with dSLR gear, the only way for this niche' segment to go is nowhere. Who gives a royal flip about decreasing sensor noise. We're already starting to see the same trend as in the decline of the 35mm SLR market with increasing competition and production of cheaper and less costly to build amatuer dSLRs. The 4/3 and APS sensor based cameras are sitting in a niche' that will sooner than later no longer exist. Share-holder demands and market pressure is more important than anything else to the companies producing these cameras.

 

None of the professional and advanced dSLR and Oly owners I know personally (and I know a lot of them) are planning any more investment in their gear aside from a lens or two. They either want to move up to full frame, or down to the $300-400 digicam segment, or both.

Peter, yes, although an 8MP or 10MP sensor is easily possible, but there will also become a point where the sensor outresolves the lens, and at this size, this will obviously happen sooner than at full frame, so much more than 10 probably isn't all that usefull anyway.

 

I agree with you for the most part, except the max 8x10". For a magazine read close to your face, especialy on glossy paper, at least 200PPI is a must, I wouldn't expect to see an E-1 image on a two page spread!

 

But wall prints at viewing distance can go much lower; the larger the print, the further away the viewer will be. The 150PPI 16x12 prints I have look absolutely fine. Olympus was even showing off A2 size at Photokina which looked great (well, the ones shot raw anyway, you could tell the JPEG shooters) but would require more preparation and a better printer, like a Lightjet.

 

This fact makes it a very good choice for social snappers as well and indeed, amongst the professionals who use it, it seems most popular with wedding photographers and photojournalists.