Canon S90 vs EP1/GF1

[chekist]

I have EP1 and this is as low as I am willing to drop as far as camera capabilities. So in general I do not have an interest in PS cameras. But recently my coworker asked what compact camera would I recommend for a long overseas vacation. So I suggested S90 and she got it. And I had an opportunity to play with it.

I was playing with it indoors and so was and could only check its low light performance compare to EP1. I tried both at 3200, 1600 and 800. S90 out of the camera looked cleaner but at the cost of heavy detail reduction due to noise reduction algorithm.

At the same time when I used noise-ninja to get Olympus to roughly the same level of cleanliness there was little left of the detail as well. The point is that there was no apparent advantage from a vastly larger m4/3 sensor. I do not claim that this was a very precise test - it was not; nevertheless, this is a bit upsetting to me.

Of course, there are other reasons why one may want to use larger sensor: color reproduction, dynamic range, depth of field. But I would like to see a comprehensive comparison to see that people like myself are not fooling ourselves and assuming by default that we are getting vastly better performance from a much heavier and expensive system.
--
--
Eugene
http://picture.stanford.edu/Photo

 

[TEBnewyork]

Depends on what is most important to you.

The other aspects you mention for instance controlling DOF is more important to me than the absolute highest ISO performance where my print size being smaller helps in the noise department. I've seen shots between both and it seems like m4/3 does hold the highlights better.

The S90 is VERY slow at the long end of the zoom range. So, the S90 may be OK at high ISO (if you don't like detail) but I can get 2 stops of faster lens for m4/3.

All this being said, the newest crop of high end P&S cameras are getting very good and the manufacturers are on the right (non-megapixel) path. I'm speaking of LX3/D-lux4, G11, S90, GRD III
--
terry
http://tbanet.zenfolio.com/

 

[Robert Deutsch]

Compare images taken at ISO1600 under controlled conditions with the S90 and the E-P1 or the GF1 at http://www.imaging-resource.com/IMCOMP/COMPS01.HTM . If you compare the ISO1600 still life images, you'll see a vast difference in favor of the MFT cameras.

The second comparison is at http://www.luminous-landscape.com/reviews/cameras/g11.shtml . Scroll down to see a comparison between the G11 (which has somewhat better IQ than the S90) and the GF1. Again, the GF1 produces much better results.

I have a GF1 and a G10 myself. At ISO80 or 100 the G10 does quite well (most people say better than the G11/S90), but above 200 there is simply no contest. I tried an S90, attracted by the size, and found that at higher ISOs there was less noise than with the G10, but it was actually not as sharp. Again, no comparison with the GF1. I also found the freewheeling rear dial extremely annoying.

Bob

 

[chekist]

Hold on. You have an m4/3 lens that is F2.5 at 52mm (2 f-stop down)? I would love to get one! Where can I get one? Or you have 14mm F2 lens (actual value)? I think, at least as far as specs are concerned S90 has pretty good lens.

Yes. As I said there are other things as well. DOF does not need to be tested. This is for sure not in S90. But for me DOF is an desirable add on, but not the main reason to get EP1 over S90. The main reasons being High ISO and dynamic range.
--
--
Eugene
http://picture.stanford.edu/Photo

 

[chekist]

Thank you. There is a difference looking at those comparisons.

I had G10, and G11/S90 sensor is a clear improvement.
--
--
Eugene
http://picture.stanford.edu/Photo

 

[TEBnewyork]

I am slower at the wider end of the S90 but I did not make any claim about the wide end. I said the S90 gets slow at the longer end.

My faster lenses that I have to use for m4/3

20mm f1.7 (AF)
25mm f1.4 (AF)
40mm f1.8 (MF)
45mm f2.8 (AF)
50mm f1.4 (MF)
75mm f2 and f2.5 (MF)

--
terry
http://tbanet.zenfolio.com/

 

[Aex]

I have EP1 and this is as low as I am willing to drop as far as camera capabilities. So in general I do not have an interest in PS cameras. But recently my coworker asked what compact camera would I recommend for a long overseas vacation. So I suggested S90 and she got it. And I had an opportunity to play with it.

I was playing with it indoors and so was and could only check its low light performance compare to EP1. I tried both at 3200, 1600 and 800. S90 out of the camera looked cleaner but at the cost of heavy detail reduction due to noise reduction algorithm.

At the same time when I used noise-ninja to get Olympus to roughly the same level of cleanliness there was little left of the detail as well. The point is that there was no apparent advantage from a vastly larger m4/3 sensor. I do not claim that this was a very precise test - it was not; nevertheless, this is a bit upsetting to me.

Of course, there are other reasons why one may want to use larger sensor: color reproduction, dynamic range, depth of field. But I would like to see a comprehensive comparison to see that people like myself are not fooling ourselves and assuming by default that we are getting vastly better performance from a much heavier and expensive system.
--
--
Eugene
http://picture.stanford.edu/Photo

I have checked out a set of comparison samples (in which the pics were taken at supposedly identical environment) and no way can the S90 have the same noise level as the EP1 - even when I'm viewing at 15" monitor size from about 2-3 feet away. I estimate the difference to be at least 2 stops.

--
=============================
My flickr: http://www.flickr.com/photos/testdasi/

 

[chekist]

This is not quite fair. If you are willing to carry 6 lenses and 3 are MF and they cost another number of $1000s of dollars on the lenses you can get faster. But then I would just take my D700 and F2.8 lenses. Gives the same depth of field as F1.8 4/3 and ISO compensates for lack of light with great high ISO performance.
--
--
Eugene
http://picture.stanford.edu/Photo

 

[chekist]

Looking at this comparison:
http://www.imaging-resource.com/IMCOMP/COMPS01.HTM

I am not sure there is 2 f-stop difference. What is true is that Canon relies on heavy noise reduction be default. So images have little noise and little detail. Olympus and Panasonic do not have such an harsh noise reduction. So there is more detail and more noise. I am not sure if it is possible to compare with about equal noise reduction, but this would be interesting.
--
--
Eugene
http://picture.stanford.edu/Photo

 

[TEBnewyork]

This is a useless argument. My only comment was the lens on the S90 was slow at the long end and you came back at me at what lenses do you have that can get you fast on m4/3. I gave you a list - and never said that I went out and bought the MF lenses for this purpose - I had them. I certainly don't pack all of that into any bag that I take out and around.

In reality I just went away for 10 days with a GF1 and 20mm and 45mm macro and was fine for what I was shooting. If I want a bigger kit I add the 7-14 and the 14-140. But those lens choices are based on my normal shooting. The point is I have the choice to go for fast glass; with the S90 Canon made the speed choice for me.

I too can take an A900 with me and I have the Zeiss fast glass for it. But that wan't the point of the whole post. We are comparing small cameras and you were commenting on their performance vs. m4/3.
--
terry
http://tbanet.zenfolio.com/

 

[chekist]

My point us that m4/3 did not make those such lenses that you are talking about. Yes, for outrageous amounts of money you can get them and make them work with m4/3. But you loose all the convenience of a modern camera and in fact its mobility - its sole advantage over a DSLR.

So you do not practically have all those lenses when you take EP1/GF1 with you. If I know for sure what I am planning to photograph I have whole arsenal of equipment to choose from. Why stop at the lenses? I have bring dozen of flashes to light up the scene.
Tripods. Assistants. But the point of compact camera is lost somewhere there.

--
--
Eugene
http://picture.stanford.edu/Photo

 

[Teski]

the 20 1.7 an 45 2.8 Terry said went on the last trip are both Panny m4/3 lenses...No adapters needed.

--
Thanks,

Teski
http://www.tedescophotography.com

 

[chekist]

Well yes. But neither one of them is 2 f-stops faster. Not only that, but with 2 of those you do not have wide angle. And if you have one, it for sure would be slower than Canon, by over 1.5 f-stop. So I am not sure that you are in clear loss as far as aperture value with S90.
--
--
Eugene
http://picture.stanford.edu/Photo

 

[Bjrn SWE]

You can compare the RAW image files from these cameras (and many other cameras) at DXO Mark:

http://www.dxomark.com/index.php/eng/Image-Quality-Database/Compare-cameras/ (appareil1) 335|0 (appareil2) 327|0 (onglet) 0 (brand) Canon (brand2) Olympus

Copy the whole link above or pic the cameras from the drop-down menues.
The Olympus m43 is better in every aspect apart from dynamic range at low ISO.

 

[Teski]

But with the larger sensor, I get better DoF with my GF1 than I do my S90. I wouldn't have had the same look on these with my S90..





--
Thanks,

Teski
http://www.tedescophotography.com

 

[TEBnewyork]

As I said in my very first post I was talking about LONG end. Point two. When I wanted wider than what the 20 gave me I shot a pano.

--
terry
http://tbanet.zenfolio.com/

 

[Serhan2]

After seeing Canon S90 size, I was checking the samples also. From what I have seen Canon used a heavy noise reduction in higher ISOs in the jpegs vs less noise reduction with the LX3 or EP1/GF1. Also you can adjust the noise reduction with LX3 or EP1/GF1 where Canon doesn't have that option. So using raw is only option to get comparable results. You can check Amin's comparison at serious compacts:

http://www.seriouscompacts.com/2009/11/iso-shootout-canon-s90-vs-ricoh-gr.html

Serhan

 

[Teski]

I own the S90 and had a G11 before I returned it and ordered my GF1. The S90 is awesome when you need a truly pocketable camera (even moreso than that GF1), but it does have limitations with the sensor size and NR applied, and I know that. I returned the G11 after a week because the S90 had everything I wanted in that type of camera, and the GF1 gave me more in a package not much bigger...I would carry the G11 on my shoulder just like the GF1.

--
Thanks,

Teski
http://www.tedescophotography.com

 

[roblarosa]

http://www.dxomark.com/index.php/eng/DxOMark-Sensor/DxOMark-reviews/DxOMark-review-for-advanced-compacts

 

[Summitron]

I bought the S90 and the E-P1. I shot a few hundred shots with each and then I returned the S90. I would definitely have kept the S90 even with the annoying thumb wheel if the IQ were sharper. The E-P1 was better with the kit lens in all respects, and with the 20mm pancake I'm sure it will even be better.

The S90 is a beautiful but flawed camera. Images are a tad soft. It is, however, a terrific restaurant camera. Even at lower ISO the S90 just had this sort of pervasive, subtle softness that I found annoying, even though it was quite flattering on portraits. I preferred the G11 to the S90 in all respects, including the built in flash that doesn't try to make you drop the camera. Still, the S90 is a great restaurant cam. Consider the Fuji F70 as well. I'm still keeping my F31fd which I preferred outdoors to the S90, and indoors with some custom lumiquest material taped over the flash for a soft, fill flash effect. I expect either the next Fuji or Canon will for sure bump off the F31fd, and indeed, were it not for the subtle softness of the S90, that or the G11would be my pocket cam. If only they could slim down the G11 a tad, and bring back the faster lens, that would be perfect.