Ad
Ad
Ad
Pages: « 1 [2] 3 4 »   Bottom of Page
Print
Author Topic: You've Got To Be Kidding!  (Read 19607 times)
jjj
Sr. Member
****
Offline Offline

Posts: 3541



WWW
« Reply #20 on: October 23, 2008, 06:53:45 PM »
ReplyReply

I had to buy a new pocket camera after my Canon S70 recently became unreliable. I ended up with a Ricoh GX200 as the G9 simply was not the tool for me only having a 35mm wide lens compared to the Ricoh's ideal for me 24mm. Also I really like the Ricoh camera design as it has some very nice touches to it that are truly excellent.
Sadly the image quality isn't as good as the design quality, I daren't use it over 100ISO due to the awful noise. Not good for an $800 camera! So whilst the Canon may be better despite the MP increase over the G9, it seems the GX200 is markedly inferior to the GX100 in that department. I tested the older version a short while back and it was certainly usuable at the higher ISO, but appalingly slow at RAW capture, so no good for me.

I'm now wishing I'd known about the G10 as I may have well have bought that instead. Maybe I could buy one for the other pocket!!


Logged

Tradition is the Backbone of the Spineless.   Futt Futt Futt Photography
BernardLanguillier
Sr. Member
****
Offline Offline

Posts: 8210



WWW
« Reply #21 on: October 23, 2008, 07:25:05 PM »
ReplyReply

Quote from: mikelouw
Very interesting article. What I'm looking forward to is a review of the G10's low-light performance. I have a G9 which I use for "social" pics, which are usually indoors in less than ideal light. The performance under such conditions is less than stellar (high noise) in comparison to DSLR's, and I would be really impressed if the G10 has made significant improvements in low-light imaging.

It is clearly not the point of the article, but yes, that is indeed a good question.

The P6000 that I have been using is capable of very decent results at base ISO, but things get worse quickly beyond that.

I am sure that just about 10MP+ compact used at low ISO would have delivered results similar to that of the G10. Now do the test again at ISO 400 and the story will probably be pretty different. Still, it would seem that the G10 is among the best at medium ISO values though.

Cheers,
Bernard
« Last Edit: October 23, 2008, 07:43:40 PM by BernardLanguillier » Logged

A few images online here!
Ken Tanaka
Full Member
***
Offline Offline

Posts: 129



WWW
« Reply #22 on: October 23, 2008, 07:30:25 PM »
ReplyReply

My admittedly brief impression of the G10 is very different from Michael's.  Since the G10 uses the same 1/1.7 CCD as the G9, and 17% more pixels crammed into that tiny chip, how could we expect its performance to be better?  And, of course, it's not.  At ISO's above 100 the image begins to get flatter and coarser.  By 400 the images, to my eyes, were much noisier in the darks than the G9, which produces a usable dim light 400.  The ISO 80 that Michael used is lovely on the G9, and apparently also on the G10.  But it's not a realistic practical benchmark.

If I did not already own a G9 I'd probably grab a G10.  But the G10 is just not a $500 step forward from the G9.

Apparently I'm not alone in my opinion, either.  The G10 price is already beginning to erode, while the remaining stocks of new G9s are selling at their original $500 or, in many cases, far ABOVE that mark.

Addendum:  The (also) new Powershot SD990IS is very much worth a look if you just want a good occasional p&s.  No, it doesn't shoot raw.  But in nearly all other respects it has nearly equivalent specs to the G10 for $150 less.  Same sensor.  Same pixellage.  Same Digic processor.
« Last Edit: October 23, 2008, 08:00:57 PM by Ken Tanaka » Logged

Best Regards,
- Ken Tanaka -

www.KenTanaka.com
DaFu
Newbie
*
Offline Offline

Posts: 23


WWW
« Reply #23 on: October 23, 2008, 08:40:27 PM »
ReplyReply

I just came back from two weeks in France with a G10 in my pocket every day. I bought it because I was getting tired of lugging around a Rebel with a zoom. I wanted something with good if not excellent picture quality and an unobtrusive presence. I didn't take a laptop with me so I had to wait until coming home to see what things looked like. ISO 80 and 100 were literally stunninga substantial improvement over the Rebel and demonstrating the quality that Michael describes. After that . . . well, things get noisy pretty fast. 200's OK; 400 is time for Noise Ninja; and 800 & 1600 are grim.

I didn't find the weight a problem at all. Coat pockets were fine. What I did find irksome was the controls: they're too close together. Why they couldn't have stretched the camera a bit so you don't have to do finger yoga before using it is a real question. And I really do not like power zoom but I know at this size it's pretty much a requirement.

The one thing I notice is consistently burned out highlights in pictures where I simply wouldn't have expected them. There is an exposure compensation dial that will reduce this but I've found I always forget to set it and then get really irritated when I don't need it and forget I've set it.

%^)

All in all I'm very pleased with the G10 but still hanker for the grace of the Spotmatic that I used decades ago.

Dave
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8908


« Reply #24 on: October 23, 2008, 08:47:54 PM »
ReplyReply

Quote from: Ken Tanaka
My admittedly brief impression of the G10 is very different from Michael's.  Since the G10 uses the same 1/1.7 CCD as the G9, and 17% more pixels crammed into that tiny chip, how could we expect its performance to be better?  And, of course, it's not.  At ISO's above 100 the image begins to get flatter and coarser.  By 400 the images, to my eyes, were much noisier in the darks than the G9, which produces a usable dim light 400.  The ISO 80 that Michael used is lovely on the G9, and apparently also on the G10.  But it's not a realistic practical benchmark.

All the P&S cameras that I've owned have produced reasonably good images only at base ISO and about one stop above base ISO. I think we have to accept this limitation.

Nevertheless, there is a particular, narrow set of circumstances where the P&S camera can almost take on the much more expensive, heavier and bulkier camera. In fact, I sometimes wonder if the latest P&S models, such as the G10, can actually exceed the image quality of the more expensive DSLR or MFDB in certain conditions.

Specifically, supposing the photographer requires both an extensive DoF for a particular shot and a reasonably fast shutter speed to freeze movement in the scene, say a street scene at dusk. Flash doesn't have the range and a long exposure on a tripod would produced too much blurring of people moving.

The options might be a Canon G10 at F3.5, ISO 80 and 1/100th sec exposure, or a 1Ds3 at F16, ISO 1600 and 100th sec exposure, or an MFDB at F22, ISO 800 underexposed 2 stops (if that's the highest ISO setting) and 100th sec exposures.

If all images have the same DoF, the same shutter speed and the same FoV, which camera will produce the best image quality in the circumstances described? Any of you bright chaps know?


Logged
AGphoto
Newbie
*
Offline Offline

Posts: 5


WWW
« Reply #25 on: October 23, 2008, 09:10:36 PM »
ReplyReply

Love the article Michael,
I have been searching for a good p&s camera,..was going to buy the G9 last year but I got over ruled lol...so instead took my 30D and all three L lenses on holiday... great way to loose weight lol
15 years ago I backpacked through Europe, I had my Eos 35mm and two lenses, plus a p&s with a glass lens.... after my two year stint, i sometimes would look at an image and find it hard to tell which camera it was shot with, until I check the rest of the negs,... seems you got that same feeling with the G10

.. now its two cameras i want,.... the 5D2 and the G10 lol

its exiting times to be a photographer

Adrian
 
Logged

Adrian Gonzalez
A.G. Post Productions
www.agphotography.com.au
Don Libby
Sr. Member
****
Offline Offline

Posts: 729


Iron Creek Photography


WWW
« Reply #26 on: October 23, 2008, 09:28:47 PM »
ReplyReply

When we went to Alaska earlier this year I of course took my Mamiya/P30+ for landscape as well as a Canon 1Ds II for wildlife; we also took a Canon G9 for a walk-around.  We used the G9 almost everyday and at the end of the day as I was saving the daily images I kept being surprised at the image quality.  No, Im not getting rid of my medium format but then again Im not about to toss the G9 out either.  I was offered a G10 when they first came out however I think Ill wait till the G12 come out just think what the quality will be then  

Good article Michael as usual.

don
Logged

Wayne Fox
Sr. Member
****
Offline Offline

Posts: 2886



WWW
« Reply #27 on: October 23, 2008, 09:36:41 PM »
ReplyReply

Quote from: Ken Tanaka
My admittedly brief impression of the G10 is very different from Michael's.  Since the G10 uses the same 1/1.7 CCD as the G9, and 17% more pixels crammed into that tiny chip, how could we expect its performance to be better?  And, of course, it's not.  At ISO's above 100 the image begins to get flatter and coarser.  By 400 the images, to my eyes, were much noisier in the darks than the G9, which produces a usable dim light 400.  The ISO 80 that Michael used is lovely on the G9, and apparently also on the G10.  But it's not a realistic practical benchmark.

If I did not already own a G9 I'd probably grab a G10.  But the G10 is just not a $500 step forward from the G9.

Apparently I'm not alone in my opinion, either.  The G10 price is already beginning to erode, while the remaining stocks of new G9s are selling at their original $500 or, in many cases, far ABOVE that mark.

Addendum:  The (also) new Powershot SD990IS is very much worth a look if you just want a good occasional p&s.  No, it doesn't shoot raw.  But in nearly all other respects it has nearly equivalent specs to the G10 for $150 less.  Same sensor.  Same pixellage.  Same Digic processor.

How can it be the same sensor, if contains more photo sites? Canon states it is a new sensor, why couldn't it be?  Why assume that a higher density sensor cannot deliver a lower noise image?  Technology improvements have been doing this for a long time, and there are several of those improvements in the G10. If smaller pixels always led to more noise how could we have ever gotten to where we are today - that is what engineers have been doing for a couple of decades now.  Where's the limit?  Don't ask me ... according to many the 1DsMark2 was pretty much the limit, yet my 1DsMark3 easily out performs it.  Who knows - the only absolute limit would be a sensor with no noise .. so until we get to that point I can't assume ways won't be invented to improve on what has already been done.

I have shot both cameras side by side specifically to see how the cameras handle noise.  I even posted the images for anyone interested, and those that responded agreed that the G10 appears to have better noise performance.  Neither camera performs well at 400 and above - snapshot performance at best.  Not much reason to discuss this, but it clearly indicates that despite the increase in pixel count, engineers have indeed managed to lower noise.  If I capture the same physical area with higher resolution, but the total noise remains the same, have I not lowered the overall noise?  Same data spread over more pixels (signal), same amount of pixels with noise.  Sounds like an improvement in the signal to noise ratio.

Your assessment of whether there is a reason to upgrade is a different discussion, and you might be right.  Well exposed G9 images would have performed very similarly in a side by side test of 13"x19" prints - the G9 performs very well.  However, I do believe the G10 may offer some improvements in some cases, such as more aggressive cropping or larger sizes.

The need for a camera to record and resolve detail has always been dependent on the amount of desired cropping latitude and overall desired print size.  There really is no right or wrong here ... just each persons goals and desires, as well as their budget.

Logged

jjj
Sr. Member
****
Offline Offline

Posts: 3541



WWW
« Reply #28 on: October 23, 2008, 10:28:36 PM »
ReplyReply

Quote from: Ken Tanaka
Addendum:  The (also) new Powershot SD990IS is very much worth a look if you just want a good occasional p&s.  No, it doesn't shoot raw.  But in nearly all other respects it has nearly equivalent specs to the G10 for $150 less.  Same sensor.  Same pixellage.  Same Digic processor.
You can hack Canon Firmware on the P+S cameras and I think, you can enable RAW as well as lots of other features. Even more handy now an Ixus has manual controls, shame it isn't the one with the 28mm lens.

Hacking Canon Article

The hacks
Logged

Tradition is the Backbone of the Spineless.   Futt Futt Futt Photography
jjj
Sr. Member
****
Offline Offline

Posts: 3541



WWW
« Reply #29 on: October 23, 2008, 10:32:05 PM »
ReplyReply

Quote from: AGphoto
15 years ago I backpacked through Europe, I had my Eos 35mm and two lenses, plus a p&s with a glass lens.... after my two year stint, i sometimes would look at an image and find it hard to tell which camera it was shot with, until I check the rest of the negs,... seems you got that same feeling with the G10
As both had the same 'sensor' i.e. film, there shouldn't  be the big differences that you see with digital P+Ss compared to FF cameras.
Logged

Tradition is the Backbone of the Spineless.   Futt Futt Futt Photography
BernardLanguillier
Sr. Member
****
Offline Offline

Posts: 8210



WWW
« Reply #30 on: October 23, 2008, 11:13:00 PM »
ReplyReply

Quote from: DaFu
The one thing I notice is consistently burned out highlights in pictures where I simply wouldn't have expected them. There is an exposure compensation dial that will reduce this but I've found I always forget to set it and then get really irritated when I don't need it and forget I've set it.

Dave,

That's the main issue with these compacts, DR typically sucks.

Cheers,
Bernard
Logged

A few images online here!
barryfitzgerald
Sr. Member
****
Offline Offline

Posts: 607


« Reply #31 on: October 24, 2008, 04:42:27 AM »
ReplyReply

I do wish these "Moore's Law" vibes would stop. It has been changed over the years in an attempt to make it fit.

http://www.firstmonday.org/issues/issue7_11/tuomi/

Nobody really cares about it, and those in the computer industry, mostly ignore it.

Back to cameras, dynamic range is again the thorny issue here, and I am not convinced technology is driving things in the right direction. If it were, we would have smaller sensor with much better DR than we currently do.

So yes things get better. but not at the rate many suggest, nor in the areas that some want.
Another point here, you can only megapixel these small sensor cameras so much, already diffraction is a problem, and it's going to get worse. Companies are going to have to come up with something a bit more interesting in years to come.
Lastly the difference between the G9 and G10 is a mere 12 to almost 15mp, pretty much invisible in giant sized prints. The most interesting move is the lens which now starts at 28mm, not a few extra lousy megapixels. Yes sensors get better, but not at the rate people would like, or suggest
Logged
NikoJorj
Sr. Member
****
Offline Offline

Posts: 1063


WWW
« Reply #32 on: October 24, 2008, 06:40:26 AM »
ReplyReply

Quote from: BernardLanguillier
That's the main issue with these compacts, DR typically sucks.
Yes, and that's the usual caveat in such comparisons too : resolution itself doesn't depend on sensor size alone (until you hit the wall of light wavelength), but DR does.

On a french forum there also has been a 450D (the latest Rebel, XSi I think) vs. 5D comparison : the two images really held the same details, though there were discriminant ones (ie evident things near pixel size), but the scene was a street with a portion of sky, and the street being exposed the same, in one shot the sky was burnt, in the other not. Guess which is which!

Micheal article is great, in that it can draw photographers' attention to what really counts : photography itself, and not the toys we use for it. But it does also mention that large pixels do retain some avantages in some situations, yes of course.

So, I'd have a request for the G10 vs P6000 review : Michael, if it's not too late, would it be possible to include some DR-specific shots?
I'm thinking to a 9 or 10-stop scene (eg an interior shot with view outside the window), exposed to the right and then processed to pull some detail out of the shadow. Of course high ISO shots tell a part of the tale, but some real-world situation like that might be useful too.
Anyway, maaaaany thanks for all the informations you provide!
« Last Edit: October 24, 2008, 07:27:51 AM by NikoJorj » Logged

Nicolas from Grenoble
A small gallery
michael
Administrator
Sr. Member
*****
Offline Offline

Posts: 4896



« Reply #33 on: October 24, 2008, 06:56:38 AM »
ReplyReply

I am currently working on the review, and yes, I will include some DR examples and comments.

Michael
Logged
Moynihan
Full Member
***
Offline Offline

Posts: 118


WWW
« Reply #34 on: October 24, 2008, 07:01:12 AM »
ReplyReply

Interesting article.
I have experienced something similar, in a way.
I use a Fuji s6000fd as a "car seat" camera.
Meaning, when i can shoot from the road side, and do not want to get out and set up.
(i know, lazy
I use a little beanbag on the partially rolled up window as a rest for the lense.
Landscape speaking, an am still surprised by its IQ (at 100 ISO, and it is 6.3 mp, sensor is a 1.7 SCCD, similar size to the G10) at the size I normally print (8x10ish) in comparison to my DSLR's (two 10 mp ones).

BTW: One little nit to pick at. The Moore's law thing is limited to central processor chips (the CPU in your desktop/laptop etc), not technology, generally.
« Last Edit: October 24, 2008, 07:04:33 AM by Moynihan » Logged
Neil Hunt
Jr. Member
**
Offline Offline

Posts: 59


WWW
« Reply #35 on: October 24, 2008, 11:05:29 AM »
ReplyReply

Well done for having the nerve to publish this, despite the criticism it will ineviatably attract from some. A very interesting and valuable insight, as were the things you've been putting out about the convergence of still and video.

Neil.
Logged

Gordon Buck
Sr. Member
****
Offline Offline

Posts: 409



WWW
« Reply #36 on: October 24, 2008, 11:15:44 AM »
ReplyReply

Quote from: Ray
Nevertheless, there is a particular, narrow set of circumstances where the P&S camera can almost take on the much more expensive, heavier and bulkier camera.

Ray, I know this isn't what you were thinking about but there is an interesting little flash photography trick that can be done with the Canon G9 and, I assume, the G10, that is not available on (any?) DSLRs.  The G9 will sync to flash at up to 1/2000 sec (depending on connections) provided it does not think the flash is a Canon flash.  By sync, I don't mean the high speed pulsing mode of Canon flash -- just a basic flash.  I learned this from the Strobist (who also has a G9) and it a handy little trick to turn day into night except for the effects of flash -- not to mention fun to tinker with.
Logged

lonna.tucker
Guest
« Reply #37 on: October 24, 2008, 02:28:13 PM »
ReplyReply

Quote from: T-1000
Hi Michael,
The G10 is a great camera, but I feel the P45 JPEG that I downloaded showed that you didn't exactly "NAIL" the focus, or perhaps you needed to stop down more.  I'm not telling you how to take photographs, but I've worked with P45 files, and this is one I would throw away due to nothing being in decent focus IMO.

I totally agree with this poster. The G10 files in this low contrast light seem nice, BUT Michael, this is a lousy example of a MFDB file. If this is what you have been getting/expecting for quality out of your Hasselblad/Phase, you don't know what you have been missing. I don't shoot Hasselblad or Phase, but I'm sure the quality should be close to what I achieve my Contax/Aptus. Maybe your equipment has suffered from recent travels and needs a checkup.
Logged
michael
Administrator
Sr. Member
*****
Offline Offline

Posts: 4896



« Reply #38 on: October 24, 2008, 03:15:25 PM »
ReplyReply

Oh come on.

This was one file, and both cameras had the same advantages and disadvantages under the situation.

Lots of other files were shot and my point about experts not being able to see differences on 13X19" prints stands.

One can nit pick, but it doesn't change the results.

And yes, trust me, after some tens of thousands of medium format shots with a dozen different cameras and at least half that many backs I do indeed know what a technically good shot looks like.

If I'd posted another pair someone would likely have found some other nit to pick.

Just download some of my sample G10 files, make a print or two, and then judge for yourself what this little sucker can do.

Michael

Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8908


« Reply #39 on: October 24, 2008, 07:30:43 PM »
ReplyReply

Quote from: lonna.tucker
I totally agree with this poster. The G10 files in this low contrast light seem nice, BUT Michael, this is a lousy example of a MFDB file. If this is what you have been getting/expecting for quality out of your Hasselblad/Phase, you don't know what you have been missing. I don't shoot Hasselblad or Phase, but I'm sure the quality should be close to what I achieve my Contax/Aptus. Maybe your equipment has suffered from recent travels and needs a checkup.

The point has probably been missed by some of you, that F11 on a DB produces a shallower DoF than F3.5 on a P&S, so there are inevitably some parts of that MFDB image that are less sharp than the G10 image. This is why I say that the circumstances where the P&S can take on the much heavier and more expensive camera are narrow.

Michael appears to have started with the MFDB shot, choosing an aperture that is moderately sharp and which provides a reasonable DoF, then as an after thought decided to shoot the same scene with the G10. To get the same DoF with the G10, he would have needed to use F1.8 which the G10 obviously doesn't have. Alternatively, he could have retaken the DB shot using F22 instead of F11. Doing that might have produced an even more shocking result because then the DB image might have been noticeably less sharp than the G10 in all parts of the image.
Logged
Pages: « 1 [2] 3 4 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad