Olympus E-3 preview

[Bob Ross]

Advertisement (gone after registration)

5 stops of image stabilization and a superfast 14-35mm f/2 or a 35-100mm f/2 zoom are a very interesting antidote against "noise". AF is able to get precise focus with very low levels of light. Now consider the most effective dust reduction system. This is a very well designed and balanced camera.

I think the specs list the effective range as 5 EV steps. The point that confuses me is are they using 1/2 EV steps or 1/3 EV steps or whole number EV steps (for which they should use the term "stops"). So, in actual stops we might have 1.66, 2,5 or 5:o

Bob

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Guest malland]

The doubts I have about the 4/3 system is the huge size of the lenses, although I like the sensor being smaller than the DX system because I like pictures with some "bite" rather than grainless results are higher ISOs. In any case I wonder how the E3 will compare with the Nokon D300, given that the two are about the same size and weight.

 

—Mitch/Paris

http://www.flickr.com/photos/10268776@N00/

[Bob Ross]

The camera looks very interesting, but it is almost the size and the full weight of a 5D, which to me is too large (and the reason I bought an M8).

 

Hi Carsten,

The M8 is the champion for the biggest sensor in the smallest body and the E-3 might be a contender for the biggest body with the smallest sensor, though the FZ-50/V-Lux1 class is probably the winner there. I have used 4/3rds, an E-1, for four years and find it good stuff, but then I don't print posters every day.

Bob

[ptomsu]

Peter, it is not a secret on the Internet ... everybody knows that I don't like 4/3.

 

It's just the digital version of Olympus Pen F half frame camera ... and they've never been shy to admit that it's a "huge failure" in their own words.

 

Read this ...

 

Olympus Olympus History : the Semi-Olympus I - the Pen Series

 

Of course there will be always diverging opinions about certain solutions, so this is also the case with 4/3. And I even must admit that I had this system when the E1 was introduced, was very happy with it and then sold it because the E3 did not come (this was 2 years ago).

 

But now after having done work with the D2X and also with R9/DMR I came to the conclusion, that if some thresholds are reached by the E3 I might want to go back.

 

A lot of these thresholds were actually reached and even exceeded as I could see during the press conference of Oly I attended today, as well when I took the E3 into my hands and got a first feeling of it.

 

1) ISO 100-3200 with low noise

2) larger and brighter viewfinder

3) much better AF

4) 10 MP (although I would have loved a bit more - maybe 14MP)

 

So I would say, that Oly overall has finally done their homework and the E3 can become a big success. This is what I am finally happy about and also given the price of this system compared to some other semipro DSLRs.

 

The sytsem is designed for up to 23MP (and even more, so once the sensor technology will allow we will se much higher resolution then. I personally belive tis will be in another 4-5 years, but what the hack, just sell the E3 then and buy the E4 or E5 and be happy with that system.

 

For my personal needs already the 10MP and ISO3200 are very much ok, but of course for my ego it might be nice to have a future higher resolution as well.

[sdai]

It's not really about poster printing but about "cropability" (not sure if there is such a word).

 

With so many pixels crammed into such a tiny sensor, you just don't have the flexibility a larger sensor will give you.

 

I've no doubt that the E-1 is great camera at its time. If Oly has adopted even the DX format, they could have done a lot more and a lot faster.

[sdai]

The sytsem is designed for up to 23MP (and even more, so once the sensor technology will allow we will se much higher resolution then. I personally belive tis will be in another 4-5 years, but what the hack, just sell the E3 then and buy the E4 or E5 and be happy with that system.

 

For my personal needs already the 10MP and ISO3200 are very much ok, but of course for my ego it might be nice to have a future higher resolution as well.

 

It won't hurt if you simply wait and see ... but voting with your wallet is a different matter altogether. If the camera could handle ISO3200 well then I must admit that it's really something.

 

The E-3 just reminds me of the D2H, which is a great camera in almost all aspects but plagued by the lousy sensor in it. We keep hearing people saying that cameras are used for taking pictures ... what's the most important thing then? image quality.

[ptomsu]

Advertisement (gone after registration)

I saw today an A0 print (portrait) taken with the E3 - it was just marvellous. You can see no pixels, it was crisp and sharp (actually a bit too sharp for a portrait) and had wonderfull colours and tones.

 

This shot was taken with ISO 800, no pstptocessing or whatever, as it came out of the camera - NO NOISE at all was visible!

 

What more do I expect?

[sdai]

This shot was taken with ISO 800, no pstptocessing or whatever,

 

That can't be true, Peter ... you could dare them to take a picture of you on the scene and print straight from the camera using pictbridge port ... not from a computer.

[ptomsu]

That can't be true, Peter ... you could dare them to take a picture of you on the scene and print straight from the camera using pictbridge port ... not from a computer.

 

This picture was taken from a professional photographer here in Vienna, who works with Olympus cameras and I fully trust him. He is also giving seminars for Olympus, so attendees can do the same thing in his studio and see themselves.

 

But I will let you know once I have done my seminar, will not be before January 2008 unfortunately :-))

[audidudi]

Isn't it target to ship on the 23rd of this November?

 

FYI, Amazon.com is taking orders at $1,699 with a delivery date of December 15th. I must admit to being tempted ... very, very tempted.

[Riley]

It's not really about poster printing but about "cropability" (not sure if there is such a word).

 

With so many pixels crammed into such a tiny sensor, you just don't have the flexibility a larger sensor will give you.

 

I've no doubt that the E-1 is great camera at its time. If Oly has adopted even the DX format, they could have done a lot more and a lot faster.

 

if you are discussing these terms in equivalent value then for the same DOF in a given light situation (for same shutter speeds) :

FF @ iso 1600 + 50mm @ f/4

APS-C @ iso 625 + 31mm @ f/3

4/3rd @ iso 400 + 25mm @ f/2

 

The smaller the sensor, the lower the iso level can be. The difference in noise levels between these alternatives will not be all that much noticeable.

 

The advantage of bigger sensors is that they typically come with larger lens mounts, enabling larger apertures and more shallow DOF, which can be useful in some situations ie. portraits to avoid going into even higher iso values. They also typically have brighter VF. Downside is portability and, also typically, cost.

 

Conversely, If Im looking for more DoF, I fight an escalating iso and subsequent loss of DR and this presupposes that noise is not of influence, where at iso1600 I would contend that it is. This in a 35mm system already inhibited by optical aberrations like vignetting and fall off, and worse with WA/UWA where soft edges and corners reduce resolution. The only way for FF to fight this, is to stop down. 4/3rds doesn't share these issues, and usually performs excellently 1 stop from wide open.

 

Should E3 have made inroads into noise processing at the pixel level (which is what C&N do after all anyway), and should that lead eat into the 2 stops difference by say a stop+, it actually has advantages in noise performance, not the other way around.

[sdai]

Should E3 have made inroads into noise processing at the pixel level (which is what C&N do after all anyway), and should that lead eat into the 2 stops difference by say a stop+, it actually has advantages in noise performance, not the other way around.

 

Canon and Nikon can integrate peripheral electronics on board because the larger sensor could provide enough workspace, not so with the 4/3 sensor.

 

All other issues putting aside, noise alone would eclipse every other advantage ... of course, they could run firmware in DSP to do noise reduction, that has been proven not to be a better solution by almost every other company.

[Riley]

Canon and Nikon can integrate peripheral electronics on board because the larger sensor could provide enough workspace, not so with the 4/3 sensor.

 

All other issues putting aside, noise alone would eclipse every other advantage ... of course, they could run firmware in DSP to do noise reduction, that has been proven not to be a better solution by almost every other company.

 

well its already happening in digital microscopy

where sensors are typically 2/3" 12.5Mp (considerably denser than 4/3rds)

they dont field a 35mmFF microscope with good reason

 

then theres this

Olympus Microscopy Resource Center: Digital Image Processing - Median Filters for Digital Images - Interactive Java Tutorial

[sdai]

You didn't read it carefully ... man.

 

The Dp71 camera used on Olympus's digital microscope actually only has 1.45MP on its CCD, to increase its spatial resolution, they've adopted a pixel shifting technology to interpolate the image to 4080x3072 pixels. It's not real pixels ... LOL

[FastFashnReloaded]

You didn't read it carefully ... man.

 

The Dp71 camera used on Olympus's digital microscope actually only has 1.45MP on its CCD, to increase its spatial resolution, they've adopted a pixel shifting technology to interpolate the image to 4080x3072 pixels. It's not real pixels ... LOL

 

Of course it is "real pixels". You build up a picture of the world around you by moving your eye, or head around. Shall we just cut off the top of your head and replace it with a big wide angle lens so you don't have to do that any more?

 

THis sounds like a pretty neat bit of equipment. Olympus DP71 Camera Offers Faster-Than-Ever Display Rate, High Sensitivity, Low Noise, Outstanding Color Fidelity Oh, and by the way, man, if it works, and works well, who cares how many pixels there are in the sensor?

 

An NTSC, CRT TV doesn't display a full picture all at once. It's called SCANNING. Seems to have worked pretty well for fifty years in TVs, RADAR, LIDAR, computer monitors, and for a lot longer in carbon-based life forms.

[FastFashnReloaded]

Canon and Nikon can integrate peripheral electronics on board because the larger sensor could provide enough workspace, not so with the 4/3 sensor.

 

All other issues putting aside, noise alone would eclipse every other advantage ... of course, they could run firmware in DSP to do noise reduction, that has been proven not to be a better solution by almost every other company.

 

"Canon and Nikon can integrate peripheral electronics on board because the larger sensor could provide enough workspace, not so with the 4/3 sensor. "

 

Wrong again. Just because a sensor's light sensitive bits are a certain size doesn't mean that you are limited on the space BEHIND or BESIDE the optics. Just because a microprocessor is a certain size doesn't mean that you can't have support chips and such right next to it, does it?

[sean_reid]

I've never understood what they mean by this. If I can normally handhold at 1/30th will this camera magically let me handhold at 1 second? Is it that stable for that long a time?

 

That's what they mean. That said, I'm sure that this claim (when its true at all) will depend on focal length, etc.

 

Cheers,

 

Sean

[sean_reid]

I'll be at PhotoEast on Friday and I'll take a look at the E3. There are bound to be samples there. My wife, a professional photographer, is very interested in the new Oly because its not only weather sealed but also, possibly, very quiet. Her big question, and mine, is how the camera does at high ISO.

 

I think we should wait to see the actual file quality before making any pronouncements about what the sensor can or can't do, etc. The rest is theoretical. I can say that the 4/3 Oly lenses I've tested have been excellent.

 

Cheers,

 

Sean

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[AlanG]

That's what they mean. That said, I'm sure that this claim (when its true at all) will depend on focal length, etc.

 

Cheers,

 

Sean

 

I think what they mean is that if you are a shaky photographer and need to use 1/500th to get consistently sharp results with a 100mm lens without IS, then you'll have a fighting chance of getting sharp results with the same combo at 1/15th. (5 stops) But even this seems pushing it a bit. I think that it would only be valid if in the above example you'd always get images that are as sharp at 1/15th with IS as you would at 1/500th without it.

 

So let's take a wide lens like a 20mm (or 10mm in 4/3rds world.) Even a very shaky shooter should be able to get sharp results without IS at 1/125th or 1/250th. So with IS they could expect equally sharp results consistently at 1/4 or 1/8th, even if they are real shaky?

 

So what if you are a steady photographer? Of course the steadier you are, the less the IS will matter because it is unrealistic to think you'll be so steady that you can shoot real long exposures.

 

I have IS on 4 lenses so I am pretty familiar with what it can and can't do. Allowing one to use exposure times that are 32 times longer than without IS seems incredible. But I'll withold judgment until I see some tests.

[Bob Ross]

Canon and Nikon can integrate peripheral electronics on board because the larger sensor could provide enough workspace, not so with the 4/3 sensor.

 

All other issues putting aside, noise alone would eclipse every other advantage ... of course, they could run firmware in DSP to do noise reduction, that has been proven not to be a better solution by almost every other company.

Oly has said that they have changed the support electronics by using bare chip technology and eliminating a common unit. This was to reduce read noise and to keep the fill factor (photo diode area) the same as lower MP sensors.

There is a link to samples of ISO noise of the E-3 on dpr. Quite good and it is probably from a pre-production unit.

Pixel size E-1 = 4.7µm and for the 300d it is 5.5µm and if we knew the fill factors, it might even be a smaller difference. Adjust the 300D sensor to 4:3 and you get 10.7MP vs. E-3's 10.1MP......so, we aren't talking about big differences in practical terms.

Bob