Ricoh GX100 Photos

#1 mal smith

I don't know at what ISO these were taken but the noise in most of them was unacceptable. I've always been an admirer of Ricoh film cameras but I'm afraid this just isn't very impressive, and for the same price you could buy a Nikon D40.

1:11 pm - Wednesday, June 13, 2007

#2 Mark Goldstein

They were taken using a range of ISO speeds from 80 to 400. I agree that the noise level is pretty high at ISO 400 and faster.

1:50 pm - Wednesday, June 13, 2007

#3 Zoltan

At ISO 400 shadow noise is rather high, but at ISO 100 it is not much worse than a DSLR. Even at ISO 200 the noise really only affects the deepest of shadows. What do they look like in print?

2:36 pm - Wednesday, June 13, 2007

#4 VEC7OR

Actually the quality is horrid, for example tunnel photo has stripes in dark areas and a LOT of noise, lets just wait and see what Sigma DP1 has, and by looking at SD14 photos (same huge 1.6x crop sensor as DP1) it will slap around alot of cameras, only thing is that low light performance aint that great, its a day camera, period.

3:22 pm - Wednesday, June 13, 2007

#5 Prognathous

You can't judge the noise by by just looking at 100% magnification without considering the pixel count. Try printing the pictures to normalize things to the same comparison grounds.

If 100% magnification was anything to go by, then the output of a 0.5 MP camera would look wonderful (no noise to talk about. ever). The GX100 is a 10 MP camera, and as such should be compared to other cameras only by doing some normalization - either by downsizing its images to the resolution of the other camera in the comparison, or by upsizing the other camera images to 10 MP. Printing removes the need to do this.

In addition, if you compare images with another 10 MP camera, don't forget to check the level of detail too. It's very easy to get a noise-free detail-free picture, but Ricoh chooses to keep the details and let the user remove noise (and details…) afterwards. Other prosumer cameras choose to remove noise and detail (e.g. Canon G7 and Panasonic LX2), and there's no way back. Whatever details the camera removed, are removed forever. It's an irreversible process. To see this logic in practice, have a look at the G7 vs. A640 comparison here:

http://www.imaging-resource.com/PRODS/G7/G7A.HTM
(search the page for the word "peeping").

Prog.

3:31 pm - Wednesday, June 13, 2007

#6 Kendall Helmstetter Gelner

Compare the shot of the tiled surface to:

http://www.pbase.com/kgelner/image/80448123/original

Or this:

http://www.pbase.com/kgelner/image/80167154/original

Compare a shot with flowers in the foreground to something like this:

http://www.pbase.com/kgelner/image/80448123/original

Or this:

http://www.pbase.com/kgelner/image/80448132/original

Or buildings and detail, to an image like this:

http://www.pbase.com/kgelner/image/80397684/original

Those are all from the SD-14, which uses the same sensor as the DP-1 will have...

5:14 pm - Wednesday, June 13, 2007

#7 Kendall Helmstetter Gelner

Or the brick and roof detail from the church with this shot:

http://www.pbase.com/kgelner/image/80025231/original

(Also from the SD-14)

5:23 pm - Wednesday, June 13, 2007

#8 Prognathous

Sure, if you shoot landscapes in daylight, then the DP1 is going to be a great camera. But forget about using it in less than ideal light. Samples and rational here:

http://forums.dpreview.com/forums/read.asp?forum=1013&message=22995258
http://forums.dpreview.com/forums/read.asp?forum=1013&message=23484799

Prog.

5:35 pm - Wednesday, June 13, 2007

#9 Kendall Helmstetter Gelner

SD-14 ISO 1600 shots can look splotchy, but can also look pretty good - it's a matter of good exposure more than simply being outdoors:

http://kigiphoto.smugmug.com/gallery/2583091#136097053
(hover over image to get link for original size)

The interesting thing is that just stepping down to ISO 800, you get a dramatic improvement in quality:

http://www.pbase.com/sigmadslr/image/75384748/original

http://kigiphoto.smugmug.com/gallery/2984950#161651738

And ISO 400 is even cleaner still:

http://www.pbase.com/sigmadslr/image/75756453/original

Those would certainly be better than ISO 400 and 800 results you could expect from the GX-100, given then ISO 400 examples in this set.

Not to mention, that with the DP-1 worked on for some months after the release of the SD-14, it should handle higher ISO images even a bit better than the SD-14. We have already seen some improvement in using ACR to process X3F RAW files instead of Photo Pro (I used Photo Pro to process all these).

8:32 pm - Wednesday, June 13, 2007

#10 Kendall Helmstetter Gelner

One last ISO 400 shot to compare to the other flower examples...

http://kigiphoto.smugmug.com/gallery/2928685#157951154
(as before, hover over image for link to Original size image).

8:35 pm - Wednesday, June 13, 2007

#11 Prognathous

Don't forget that the GX100 f/2.5 lens is 1.5 stop faster than that of the DP1 (f/4.0) and that the Ricoh has an image stabilizer that is worth at least another stop.

You should be comparing DP1 at ISO 800 to GX100 at ISO 160 and DP1 at ISO 400 to GX100 at ISO 80. Who do you think would win that match?

Prog.

10:05 pm - Wednesday, June 13, 2007

#12 Kendall Helmstetter Gelner

The DP-1 would still be a little better, given the pictures shown.

Not to mention the image stabilizer is only good if you are shaking, not the subject. So you get perhaps a 1.5 stop advantage for most subjects, not very significant.

10:28 pm - Wednesday, June 13, 2007

#13 Kendall Helmstetter Gelner

On second thought, one aspect you are ignoring when talking about lens speeds is the best quality possible from the camera. Sure if you go in a dark room, the two cameras might produce more similar results in everyday use (though I'd have to see indoor samples from the GX-100 to say for sure!).

But the gallery we are all responded to is filled with outdoor shots, where the higher speed lens and image stabilization simply wouldn't matter - because the shutter speed would be fast enough that either camera could be easily handheld to get the sharpest shots possible for that camera. And in those cases, I can say unequivocally that the current sensor in the SD-14 (and soon to be in the DP-1) can produce much better images.

I think a person shooting mostly indoors with tricky subjects, like someone shooting a young child, might still be better off with a Fuji. But if you are getting an expensive compact with a wide range of manual controls, you'll be shooting a wide range of subjects including a lot of outdoors stuff, and you'll want the highest possible quality output from said camera.

10:40 pm - Wednesday, June 13, 2007

#14 Stef

I bought a GX100 a couple of weeks ago after a lot of thought and research because of its wide lens, raw capability and viewfinder option. The noise levels at 100% view are high but so what? If I could have I would have bought an 6-8mp version but marketing requirements prevent that. How many people take pictures with a compact that really require 10mp and which other compact produces 10mp images that look perfect at 100%? If you downsize images from the GX100 to 6-8mp equivalent they look just fine and are plenty big enough. Yes I could have bought a D40 for the same money but I can't fit it in my pocket. The GX100 is unashamedly a street photography camera and it is well suited that task. You're also going to need to put some work into post processing but the kind of people the GX100 is aimed that expect to do, want to do that. I wouldn't recommend it for general snapping or the highest possible image quality and anyone spending £300-400 on a compact camera expecting that is either spending too much money on a compact or buying the wrong kind of camera

12:02 am - Thursday, June 14, 2007

#15 Prognathous

Mark, I just noticed that you shot many of the pictures using very small apertures (in terms of small-sensor cameras). That's too bad, as diffraction kick in real early with such cameras. If you test the actual resolution at different apertures you'll see that unlike with DSLRs, the GX100 and its like peak at around 1~1.5 stop down from wide-open aperture, certainly not at F8.0~F11 like in many of the pictures. Compare picture 6 (f11) with picture 3 (f5.6) and you'll easily see what I mean.

Everyone else, try downsizing the 10 MP pictures of the GX100 to 4.7 MP (like the DP1), then tell about the noise

Prog.

12:55 am - Thursday, June 14, 2007

#16 Stef

Spot on Prog. I was flicking through some photos I've taken with my GX100 and wondering why some of the images reproduced here look so crummy in comparison. The resolution of my GX100 excellent (to my eyes anyway) all the way out to the corners. Clearly, a drab day and poor lighting didn't help but I think you're right at least some of the test pictures may have suffered from diffraction.

1:23 am - Thursday, June 14, 2007

#17 Kendall Helmstetter Gelner

I agree that a number of these images are drapping some sharpness due to diffraction, though in image 6 the problem was more it was focused too close I think (the walls on the side are much sharper and then it gradually less sharp as you move to the back of the channel).

Yes you can certainly resample them a lot smaller, and that helps - but it will still be nowhere near the quality of a larger sensor. The advantage of the Foveon sensor to be in the Sigma DP-1 is that images from it should upsample much better than images from other bayer cameras, which is why its generally agreed it holds the same level of detail at a 9-12MP bayer camera (look at any SD-14 review, like the one in this months Popular Photography). It's not really a 4.7 MP camera, since it has 4.7 million red, green, and blue photosites each - a camera like the GX 100 has 5 million green photosites, but only 2.5 million red and blue photosites recording data. The difference is in the SD-14 (and the DP-1) the reg, green, and blue photosites are stacked.

To learn more about resolution differences between Foveon and Bayer sensors, you can read this:

http://www.ddisoftware.com/sd14-5d/

As for which compact camera will produce 10MP images that look good at 100% - well that's why we're waiting on the DP-1.

2:10 am - Thursday, June 14, 2007

#18 harold

I'm not a professional, or even a serious amateur. I've had a digital camera for a few years. It recently broke and I'm looking for a new one. Anyway, when I look at these photos they all seem "dead". It's like there's no colour. Were they all taken on a grey day or something?

4:26 am - Thursday, June 14, 2007

#19 Kendall Helmstetter Gelner

They were all taken under pretty overcast conditions it looked like, so that does totally wash out the sky. The colors are also somewhat flat as a result, but I think they look flatter than they should - flowers usually look pretty good on cloudy days.

5:07 am - Thursday, June 14, 2007

#20 Prognathous

Kendall, I'm well familiar with how Foveon sensors work. My point is that you should compare the SD14 and GX100 *final* output (post interpolation) after normalizing the files to the same pixel count - either both at 4.7 MP, or both at 10 MP. Alternatively, just print them.

Prog.

7:01 am - Thursday, June 14, 2007

#21 arvatoth

Yup. Printing is the ultimate test.

11:39 am - Thursday, June 14, 2007

#22 Stef

The lighting does make a big difference

If anyone is interested, I've just put a handful of unedited, full-size GX100 pictures that were taken at default settings in a small set here...

http://tinyurl.com/2wa89l

They're are not great as pictures but I picked them because they are a decent example of the resolution and colour I'm getting out of the GX100. All were taken at ISO80/100. ISO200 is noticeably fuzzier but still very usable. ISO400 really is as far as you'd want to go for largish prints. One problem I have already found in my first few days with the camera is its dynamic range, or lack of it. I'm having to deliberately underexpose a lot of shots to avoid blowing out the bright areas. Having such a wide angle lens that takes in so much doesn't help

@Harold above - this really is a 'serious' amateur's camera. If you don't fall into that category something like a Fuji F series or Canon A series offer much more consistent and pleasing results out of the box. They're a lot cheaper too

1:34 pm - Thursday, June 14, 2007

#23 Kendall Helmstetter Gelner

Prog - I agree you should compare images at equal sizes. But that equal size is not 4.7 MP. Upscaling SD-14 (or DP-1) images to 10MP would be fairer, or at least easier to judge... or as you say prints, of which I have had SD-14 images printed as large as 30x40" (for Sigma's booth at PMA), which is I think near a realistic limit for them.

Stef - I agree those look much better, especially the one with the reflection across the water. There still is some detail smoothing going on there though, and some noise (though greatly reduced over other shots). Thanks for posting those.

4:11 pm - Thursday, June 14, 2007

#24 Stef

@kendall - one thing I am sure of is that the GX100's lens is better than the sensor. The bottom line for me is that I kept the GX100 in spite of the fact that it was an expensive purchase. But this is really is a camera that requires some thought in its use and in post-processing. That's not a bad thing IMHO but coupled with the price there's no way this camera is anything other than a niche product. And that's always going to be a challenge for people writing reviews with the broadest possible audience in mind

4:34 pm - Thursday, June 14, 2007

#25 Adam

I have been mystified by a lot of gx100 "sample" photos I have seen on the web - most are just bad: bad lighting, bad choice of subject, bad composition, bad choice of camera settings. Here are some that I took:

http://www.themop.net/temp2/portrait_gx100.jpg
http://www.themop.net/temp2/cat_gx100.jpg

To me - these prove that good results are easily possible with this camera - at 1/4 or less the size of the SLR I usually carry. Yes - these are taken at ISO80 - so the noise issue is not present here. Yes - I think the noise in this camera at 400 or greater is not ideal - but it has lots of other merits that count for a lot in my book.

6:03 pm - Thursday, June 14, 2007

#26 Prognathous

Adam, any chance you could post these pictures (or similar) downsized to 4.7 MP and sharpened?

Prog.

6:24 pm - Thursday, June 14, 2007

#27 Adam

Sure - here they are:

http://www.themop.net/temp2/portrait_gx100_4_7mp.jpg
http://www.themop.net/temp2/cat_gx100_4_7mp.jpg

7:00 pm - Thursday, June 14, 2007

#28 Prognathous

Thanks Adam, these are excellent. Just like I though they would be.

Any Sigma enthusiast in the house to comment on those pictures?

Prog.

8:35 pm - Thursday, June 14, 2007

#29 Kendall Helmstetter Gelner

They still do not seem as sharp to me as a Foveon image of equivalent size. But again, you simply cannot compare by downsampling, you have to upsample the images. I would like to have seen the full size originals of those shots, not just the smaller versions.

11:45 pm - Thursday, June 14, 2007

#30 Adam

No problem, here are a couple of 100% crops from the original files:

http://www.themop.net/temp2/cat100_crop.jpg
http://www.themop.net/temp2/port100_crop.jpg

12:59 am - Friday, June 15, 2007

#31 Mark Goldstein

In answer to the point about the photos being taken in poor lighting (aka a typical UK day), I traveled all the way to Sunny Spain earlier this week, just to get some nice sunny pics for you all

There are now 10 extra photos in the GX100 gallery.

4:25 pm - Friday, June 15, 2007

#32 Stef

LOL

The sacrifices you are willing to make for your reading public

6:32 pm - Friday, June 15, 2007

#33 Prognathous

Mark, thanks for posting more samples, but unfortunately almost all them were shot with apertures that are too small for a small-sensor camera. The result is poor resolution compared to what the GX lens is capable of.

You can calculate the diffraction limit of the lens using the calculator at the bottom of this page:

http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm

Just select "digital compact with 1/1.8 sensor" and 10MP, then choose the aperture you wish to check. You'll see that f/8.0, f/9.5 and f/10 used in the gallery are way too small to get good quality. You really don't need to use such apertures, as small-sensor cameras have abundant DOF even in their wide-open aperture.

Prog.

6:44 pm - Friday, June 15, 2007

#34 Björn Utpott

I was willing to wait for the DP1 if I thought it might better suit my needs than the GX100. In the end, I decided for the GX100. Everywhere I go, I always have my camera with me. It needs to be small and unobtrusive. I decided to go with the added versatility that the GX100's zoom range offers. Under less ideal lighting, the GX100's bright lens and stabilisation will somewhat offset the image quality advantages that the DP1 will likely have.

For those interested in the GX100's image quality, I have posted unedited JPEG files with links to the corresponding RAW (DNG) files.

http://www.pbase.com/viztyger/gx100_test

7:44 pm - Sunday, June 17, 2007

#35 Prognathous

The review is up:

http://www.photographyblog.com/reviews_ricoh_caplio_gx100.php

Prog.

3:53 pm - Wednesday, June 20, 2007

#36 Zoltan

The last 10 shots - the ones taken in Spain - attest to the great dynamic range of the Ricoh's sensor. At least 4 of them appear to have been taken in very contrasty light, but no highlights are blown, while shadow detail is good and colours vibrant. A rare combo.

12:30 pm - Monday, July 16, 2007