PowerShot S100 reviewed and compared to the S95

[ecka]

Do you realize that Nikon 1 with the lens attached is twice bigger and heavier than S100?

yes, it is less compact. About 25% larger body and the lens sticks out = about twice as "thick" as the S100 with lens retracted = not ready to shoot.

But so what: I still get a V1 with 10-30 in my right jacket pocket and a 30-110 in my left pocket.
And more importantly: every millimeter and every ounce is worth it, many times over.

Whatever works for you. I'm going to keep my 5D2 as my main and only camera for now . Still waiting for Canon mirrorless.

 

[koolman]

The main drawback of the compact cameras to my mind - is the slow focus time and shutter lag as compared to DSLR's. The second drawback is the lack of an optical viewfinder which is critical to me for composition.

The S100 is once again - a small step up of ISO performance - but basically more of the same (as apposed to the Nikon V1 which breaks the speed barrier)

For home use then, which can be GREATLY expanded if you could shoot kids on the move, and any other moving objects, I would NOT look at the S100, but rather one of the newer fast focusing options, V1, GX1, ep3, etc. based on your budget and physical size.

 

[elflord]

but today I saw a sale on the new olympus pen mini...ohhh boy, it's beautiful and I love olympus.
I thought of Pen EP3 or even the 2 but it's waaay to expensive and too big for her tastes.
Ssssso, what you guys suggest me?

The Panasonic GF2 can be had used for about $250. It's close to the size of a compact. Pair it with a prime and it's quite small.

The proliferation of m43 bodies means that the older models are pretty cheap used (that includes the EP2)

 

[NotABunny]

The black doesn't ever look quite as black on the S95 through the ISO's which is probably why you're picking up the 'washed out' look. Must just have different contrasts in the lenses.

Exactly, it looks like there is more tonal room in S100. A simple black level can fix the black / color intensity on S100, but you can't get more tonal details on the S95.

 

[fifowarehouse]

The battery drains so FAST. Picture quality is NOT much difference compared with S95 under low light.

What I like about the S100 more than S95:
1. Shoot faster
2. Feel Solid, Hand Grip, and texture
3. GPS
4. Got little more zoom, but can't use under low light

 

[KyleSTL]

I'll report back on other issues raised.

One little feature I forgot to mention, and haven't seen mentioned elsewhere. The S100 has a built-in 3-stop ND filter. Nothing in the manual confirms it's an optical filter, although the click when it's turned on and subsequent lag to correct the exposure do strongly suggest that it is, and the image circle is small enough that it's quite feasible. The only lingering skepticism is that Canon does manipulate sensor sensitivity on the sly, e.g. when ultrafast lenses are used on a dSLR, where they compensate for an angle of incidence too shallow for the sensor by 'secretly' increasing sensitivity (i.e. ISO). But, I suspect in this case, it's a real ND filter. Handy for waterfalls and other desired motion-capture situations, and for overcoming the max 1/2000 s shutter speed when needed.

For what it's worth, even the SD700 (5 years old) had a built-in opitcal ND. I disassembled one a couple years ago to remove sand from the lens barrel. I would imagine Canon is still using optical ND.

 

[neuroanatomist]

One more note about the ISO performance - the main goal was to compare the S100 to the S95, not to examine 'real world' high-ISO performance of the S100. There is certainly less ISO noise with the S100 than the S95 - but they are still using 1/1.7" sensors - and since ISO noise is determined primarily by total light gathered, even the 'better' S100 will be beaten soundly by a dSLR.

It's worth noting that almost every ISO noise test out there on the internet fails to look at 'real world' high-ISO performance, at least in terms of when you usually think of using high ISOs - in low light. Granted, higher ISOs can come in handy to push up shutter speed in decent light, too. But the problem with ISO noise tests is that when you increase ISO during the test, you have to decrease something else to keep the exposure the same. Most commonly, what is changed is shutter speed, so your high ISO shots are short exposures, which reduces the impact of read noise, the main noise component in low-light shooting. Although I shot my S100/S95 comparisons in the 'traditional' way (increase ISO and decrease shutter speed to compensate), when I ran my 5DII/7D tests, I kept aperture and shutter speed constant as I increased ISO, instead decreasing the illumination with ND filters. In that test, the noise at high ISO on both cameras looked worse than other published noise tests, because high ISOs were used in low-light situations (like the 'real world' but unlike most tests).

Why not cover sharpness and distortions?

Lack of time, mostly. :-\ I might test those, at some point. But, while relative sensor performance was an unknown, about differential sharpness and distortion performance we can make some pretty educated guesses.

Sharpness will be slightly higher with the S100 than the S95, based on it's 12 MP sensor vs. the 10 MP sensor. Lots of comparative data to support that assertion - take any lens you like, and use DxOMark to compare resolution on different cameras with sensors of the same size (5DII/5D, or 7D/50D/40D, etc.). Resolution goes up with increasing MP.

Distortion will be higher on the S100 - it's got a 5x zoom range with a 5.2mm wide end, vs. a 3.8x range with a 6.0mm wide end. That's almost certain to mean more barrel distortion at the wide end, and probably more pincushion at the long end.

The battery drains so FAST. Picture quality is NOT much difference compared with S95 under low light.

I'll try to do some bettery life testing over the weekend. For me, this isn't normally an issue - I have two batteries (for all my cameras, actually), and I swap them out either at the end of the day or before I take the camera out the next time. That's really the best method, IMO. Li-based batteries perform best and last longest when they are used frequently and lightly - that's true for cell phones, laptops, and cameras. I find that having a pair of batteries and swapping them is the best approach, because I always have a spare battery ready.

Over the coming weekend, I'll try taking some 'real world' shots to compare the S95 and S100 in low light. Perhaps I'll even shoot the same scene with my dSLRs, for comparison.

For home use then, which can be GREATLY expanded if you could shoot kids on the move, and any other moving objects, I would NOT look at the S100...

Ahhh...the irony. Actually, the very first place I took my S100 to shoot something other than a test setup was my daughter's ballet and tap class. I wanted to see how it would do in exactly that situation. It performed pretty well, getting a few decent shots from the short class. No, it's not a dSLR. I shot a class with one, and of course the results were better (and no, I didn't mind the looks from other parents when I pulled out a gripped body with a 70-200mm f/2.8L IS II, but the S100 didn't get any looks...). But it does ok, given that the reason I have one is for times when I simply cannot bring a dSLR. In those situations, compactness is key - for me, there's a big difference between the S100/S95 and a G12-m4/3, and a much smaller difference between the G12-m4/3 and my 5DII/7D. For others, that might not be the case, but my requirement is for the smallest possible camera giving usable results in RAW format. Anyway, here's a sample, 13mm (60mm FF equivalent), f/4, 1/100 s, ISO 800.

 

[donthink]

thanks for your review. the red gone pink is especially helpful to know

 

[distant.star]

All the technical jumbo mumbo aside, I love the composition of that picture, Big Brain. It's a real delight to look at, and for me that's always the bottom line in photography. Someone may have a million dollars of the finest equipment ever engineered by Nobel prize winning geniuses, but if the picture they give me is not pleasing to look at, I don't care.

I share your thoughts about the usefulness and purpose of the S-95/S-100 line. For me, it's the camera I have with me. Compared to the DSLR it has drawbacks in every situation, but it's an order of magnitude better than anything else in the real world of the genuinely pocketable P&S. Also, given the privacy/terrorism/paranoia issues we have today (especially noticeable to me since moving back to the east coast) my S-95 is hardly noticed amid groups of people taking pictures with cell phones. And I've seen few people who aren't instantly intimidated when they see someone pointing a big, two-handed machine with a long white tube at them.

For those folks considering the S-100, I'm always quick to point out this is simply one test from one person. Big Brain has a lot of credibility with me, and he's doing the photography world a great service, but there are a hundred other sources for tests, evaluations and opinions. I'm guessing he would also suggest you check many of those you trust before making a purchase decision.

Big Brain, you may have a real future as the antithesis of that well known loony Web log character who names a new greatest camera/lens/radiation detector that was ever made in the history of humanity -- every day.

Thanks again, and keep taking good pictures!

 

[sjprg]

Here is a side by side comparsion of the LX5 and the S100. I would like an S100 for the size but the LX5 is hard to replace so far.

http://www.digitalversus.com/duels.php?ty=1&ma1=60&mo1=1256&p1=9802&ma2=1&mo2=1543&p2=12094&ph=6

 

[PaulRivers]

Appreciate the posting of the shots and the review!

ISO6400 on the S100 looks less noisy than ISO1600 and almost as good as ISO800 on the S95.

That is because the s100 uses much, much stronger noise reduction than the s95 did. It nearly completely eliminates noise, but at the expense of a little detail, and gives pics a slightly "plasticy" look to them (some people I've asked like it, others don't).

Using a test chart the noise reduction works perfectly, it handles solid lines and the like very easily and produces natural look results. It's not as clear if you take pics of people, or stuff with texture to it. I would not that the s100 lets you set the noise reduction to low, and if you shoot raw you can remove this completely.

If you want to see what I mean, or are interested in more comparison shots, check out these threads I created -

s95 vs s100 - low light w/people comparison pics
http://forums.dpreview.com/forums/readflat.asp?forum=1010&thread=39700614

s95 vs s100 - outdoor shots
http://forums.dpreview.com/forums/readflat.asp?forum=1010&thread=39709784

s100 - daylight - dpp conversion vs jpg in-camera
http://forums.dpreview.com/forums/readflat.asp?forum=1010&thread=39721598

s95 vs s100 auto mode improvements and not - "auto flash" fixed
http://forums.dpreview.com/forums/readflat.asp?forum=1010&thread=39624054

s95 vs s100 - new non-image-quality features / new annoyances
http://forums.dpreview.com/forums/readflat.asp?forum=1010&thread=39707426

EDIT: I would also note that the "s95 vs s100 - outdoor shots" suggests that the increase from 10 to 12mp has not made any improvement in detail captured by the camera in good light. Since the sensor changed from CCD to CMOS, one can't jump to any huge conclusion about megapixels, but in this case at least I do not believe increasing megapixels has increased resolution at all.

 

[7enderbender]

I'd love to buy a new camera for my lady.
She has a suuuuper old Pana/leica (dont recall the name) and she loves to take pictures but the DSLR route is not in thoughts because of weight and size. She wants quality yet pockability and versatility.
She had her eyes on a G12 some months ago but I kinda pushed her on the s95 way since it's a marvellous compact camera.

Are you sure you're doing her a favor here? Which Panasonic-Leica is she looking to replace? A little P&S may not be quite in the same league. I understand that actual image quality is going to be somewhat meh in any case given the limitations of the tiny little sensors. But a G12 may just be nicer to handle depending on which camera she used to know and how often she tweaks settings.

Thanks to neuroanatomist for posting this, especially the comparison shots. Makes me kind of go an find a bargain on the S95 honestly. I personally don't care much for high ISO and low noise. Not a big fan of the plasticky look and I find myself adding grain often in post processing. As far as colors go the S95 is the clear winner here also. So maybe it would make a good replacement for my old SD500 (which I never really liked honestly).

 

[Meh]

Appreciate the posting of the shots and the review!

ISO6400 on the S100 looks less noisy than ISO1600 and almost as good as ISO800 on the S95.

That is because the s100 uses much, much stronger noise reduction than the s95 did. It nearly completely eliminates noise, but at the expense of a little detail, and gives pics a slightly "plasticy" look to them (some people I've asked like it, others don't).

We'll have to wait for neuro to confirm but I think he shot RAW and used the same software to process the RAW files so any noise reduction, and he may not have applied any, would be the same I think?

 

[PaulRivers]

Re: PowerShot S100 - not good enough ... will go for Nikon V1

Therefore, I will NOT buy the S100, but rather a Nikon V1, as soon as it becomes available at a reasonable price [i.e. approx. € 500 including kit lens].

Why? Because the little Nikon 1 is not that much bigger than a S100 and offers an entirely different performance level in terms of sensor performance and even more importantly, an up-to-speed AF system and way bettter operational speed plus a solid EVF on top (V1). Video not even mentioned, as it is of no interest to me.

The s100 is significantly smaller than a V1, the s100 is half the size of the xz-1 by volume of a box around each camera, and 50% by depth the most important measurement for pocketability - that the v1 will be at least as big as an xz-1 with a pancake lens, larger with any sort of zoom. However, let's just assume that size difference is irrelevant.

The problem, correct me if I'm wrong, is that the v1/j1 don't have any good f2.0-or-better low light prime lenses - http://www.nikonusa.com/Nikon-Products/Camera-Lenses/1-NIKKOR-Lenses/index.page

Best is a 10mm f2.8 pancake (that's like 28mm-equivalent).

dpreview has studio shots of the v1 up that you could compare to an s95 -
http://www.dpreview.com/reviews/q42010highendcompactgroup/page14.asp

At only one stop better performance, the v1 is still probably better in low light (with the s95 @ iso1600 and the v1 at iso3200, which is usually the kind of light I see in average indoor settings), but the low light performance is not...gigantic.

I am done for good with all those SLUGGISH, dwarfy-Sensor digi-compacts, including the Canon S and G line.

Can't argue with you there, that's for sure. The G and S cameras are speed demons compared to cameras from 5-10 years ago, but they're pretty relatively sluggish by todays standards, and it's annoying.

 

[PaulRivers]

We'll have to wait for neuro to confirm but I think he shot RAW and used the same software to process the RAW files so any noise reduction, and he may not have applied any, would be the same I think?

If you use Canon's included raw conversion software (dpp), it does things differently to different raw files based on the camera model number. s95 raw files clearly (if you've compared them side by side) get a very, very different kind of noise reduction applied to them than the s100 raw files do. (If you want to see examples, see the thread I linked to, there's a bunch of dpp converted raw files for both the s95 and s100 there, the difference in the kind of noise reduction being applied is very obvious).

 

[Meh]

We'll have to wait for neuro to confirm but I think he shot RAW and used the same software to process the RAW files so any noise reduction, and he may not have applied any, would be the same I think?

If you use Canon's included raw conversion software (dpp), it does things differently to different raw files based on the camera model number. s95 raw files clearly (if you've compared them side by side) get a very, very different kind of noise reduction applied to them than the s100 raw files do. (If you want to see examples, see the thread I linked to, there's a bunch of dpp converted raw files for both the s95 and s100 there, the difference in the kind of noise reduction being applied is very obvious).

Fair enough if that's the case. I tried to follow your links but the images don't display rather it comes up as "gallery image not found" so I'll try again later.

I did notice your statement at the outset that said "converted in dpp using whatever the defaults were". So dpp applies different default conversion settings specific to each camera. Presumably, those defaults are chosen to optimize the final image, but to what standard? Wouldn't it be a more apples-to-apples comparison to use the same conversion settings and possibly turn off any software NR when trying to compare the noise performance of one camera to another? The final image is what matters so it might be appropriate to apply the optimal NR algorithms and compare images particularly if for some reason the NR algorithms would work better on one file than the other but I'm not sure that's the case.

 

[mp2011]

I bought this camera about a month ago (not sure exactly when). They were available at certain Best Buys and I got one that way. I did not have a S95, and I had not used a point and shoot for quite a while (I have been using a T1i for the last several years, and got a 5DMII a few weeks ago). After about a week with the S100, I returned the camera. Shooting stationary objects in good light produces good results. Otherwise, I found the camera to be pretty much unusable for any of my needs- my breaking point was taking it to a Halloween zoo event to take photos of my kids. The shutter lag was incredibly slow. None of the focus modes worked well for me- with a point and shoot type camera, I can really see how a touchscreen would be useful in picking a focus point. Shooting people in jpeg, there is so much in camera noise reduction applied, people look ridiculous. I honestly think that it has to be me, but I just do not get the following this camera has. I think if you have cooperative subjects (vs. small children), and shoot in RAW, your results are probably better. Awesome ergonomics, very pocketable, I loved the feel of the camera, but for me the everyday results were pretty awful. Some of my macro shots, of objects, were really, really good. But that is not the situation I want a point and shoot for. Since I don't really care about pocketability so much, I think another option (m43rds? X100? I have no idea) would be better for me. I also really forgot about shutter lag- like it was disorienting. I suppose that any camera that is not a DSLR probably has that though, to some degree.

 

[Doodah]

Maybe we can look at small cameras as a step up from camera phones?

 

[PaulRivers]

I bought this camera about a month ago (not sure exactly when). They were available at certain Best Buys and I got one that way. I did not have a S95, and I had not used a point and shoot for quite a while (I have been using a T1i for the last several years, and got a 5DMII a few weeks ago). After about a week with the S100, I returned the camera. Shooting stationary objects in good light produces good results. Otherwise, I found the camera to be pretty much unusable for any of my needs- my breaking point was taking it to a Halloween zoo event to take photos of my kids. The shutter lag was incredibly slow.

lol, it's kind of funny - I totally understand where you're coming from, it's dog slow compared to a snappy dslr (a good dslr with a decently fast autofocus lens). On the other hand, my dad recently tried shooting with his 5-10 year old Olympus compact, and remembered how much slower cameras used to be, lol.

None of the focus modes worked well for me- with a point and shoot type camera, I can really see how a touchscreen would be useful in picking a focus point.

One common mistake dslr users make when moving to a compact is getting really paranoid about the focus point. On a dslr in low light, the focus point is completely critical, if it's even a little off it ruins the pic. But a compact even at f2.0 has a *much* larger depth of field where the focus point is *far* less important. If your subject is anywhere near the middle of your shot, you're good.

If your subject isn't near the middle of your shot, the technique that's used is to put frame the camera so the subject is in the middle your shot, then either half-press the shutter or assign the shortcut button to focus lock and press it, then move the frame to how you want it and take the pic. With a smaller sensored camera the very slight difference in distance you create by moving the camera over is completely unnoticeable for focus (whereis with a dslr and f2.0 it can screw up your shot).

And fyi, that's the fastest way to do it. A new feature on the s100 is to use FlexiZone to move around your focus point on the screen.

Shooting people in jpeg, there is so much in camera noise reduction applied, people look ridiculous.

I kind of agree with you - it's the effect of the new noise reduction on the s100, the s95 used a totally different kind of noise reduction.

You can turn the noise reduction down in the camera to "low", but you still get some of the effect. If you shoot raw and convert with Canon's software you can turn detail noise reduction off, which is what I'm leaning towards doing. And I'm not sure what Lightroom will do, but it will probably give you more options.

It's funny, I didn't really like it's effect when people are in the pic. But I showed comparison pics with noise reduction on and off and they usually preferred how it looked with noise reduction on, so...

I do wish there was an in-camera option to turn noise reduction "off".

I honestly think that it has to be me, but I just do not get the following this camera has. I think if you have cooperative subjects (vs. small children), and shoot in RAW, your results are probably better. Awesome ergonomics, very pocketable, I loved the feel of the camera, but for me the everyday results were pretty awful. Some of my macro shots, of objects, were really, really good. But that is not the situation I want a point and shoot for. Since I don't really care about pocketability so much, I think another option (m43rds? X100? I have no idea) would be better for me. I also really forgot about shutter lag- like it was disorienting. I suppose that any camera that is not a DSLR probably has that though, to some degree.

The vast majority of compacts have similar amounts of shutter lag.

There is one that I know of that does not - the Panasonic lx5. It's not "jeans pocketable", but it's still "could possibly cram it in a jeans pocket would just be really uncomfortable to walk around like that" sized.

Imaging resource's timing gave it's shutter lag timing almost the same as an entry level dslr - and more importantly I tried it out myself at my local camera store next to a Canon t2i and could not tell the difference in focus speed (unlike the s95, g12, and Nikon p7000 where I could tell the difference in focus speed). The lx5 has similar low light performance to the s100 (though not the s100's new noise reduction, which sounds like it would be a plus for you).

The other option is that the new Nikon Mirrorless system touts itself as happy instant autofocus (Nikon System 1 I think). It has a 60fps full resolution shot rate, so it's plausible. People hated it when it was announced because it's sensor, while noteably larger than a compact, is smaller than m43rds. The other big drawback is that it's best low light lens if f2.8. But if autofocus speed is your priority...

The latest Olympus m43rds cameras claim near-instant autofocus...but from what I've read that's only with the latest Olympus lenses, with the Panasonic f1.7 lens (the low light lens one would use for m43rds), it's not as fast, though it could certainly still be faster than the s100 (I don't know personally).

The newly announced Panasonic gx1 (I think that's the model number) uses Panasonic's new m43rds sensor which is better than the older one in the Olympus m43rds cameras. However, their m43rds aren't known for the snappy autofocus speeds, and they even only claim like a 10% improvement in the gx1.

The Sony NEX wasn't known for instant autofocus speeds, and it's best low light lens is unfortunately f2.8 as well.

The Samsung nx200 looks promising, it's so new there's not much info on it's autofocus speeds. Samsung claims 0.1 seconds (instant), I don't know if this pans out or not. They do have an f2.0 prime lens available right now for it, though.

 

[PaulRivers]

Fair enough if that's the case. I tried to follow your links but the images don't display rather it comes up as "gallery image not found" so I'll try again later.

Ugh, well they work normally, but occassionally dpreview get flaky on the links working. :-(

I did notice your statement at the outset that said "converted in dpp using whatever the defaults were". So dpp applies different default conversion settings specific to each camera. Presumably, those defaults are chosen to optimize the final image, but to what standard?

Right, there is absolutely no doubt that dpp applies different settings based on the camera model.

The standard is...whatever Canon thinks looks the best, lol.

Wouldn't it be a more apples-to-apples comparison to use the same conversion settings and possibly turn off any software NR when trying to compare the noise performance of one camera to another? The final image is what matters so it might be appropriate to apply the optimal NR algorithms and compare images particularly if for some reason the NR algorithms would work better on one file than the other but I'm not sure that's the case.

No, not in my opinion.

There's a couple of reasons, though to be fair I did actually do most of my comparison with detail noise reduction turned off (and color noise reduction left on), but that's because I prefer that look.

First, what most people are really interested in (including myself) is the actual results they will get out of the camera in my own day to day use. If Camera A is theoretically better than Camera B, but in reality Camera B produces better pics because of more advanced processing, and I usually use the default processing, who cares if Camera A is theoretically better?

In this case the only way to do a "fair" comparison is to use the exact same workflow and tools you would use...that is also a little impossible as different people use different tools, and new versions of those tools come out later, but I try to stick with what I will actually use.

Second, "turning off" noise reduction is fundamentally unfair for a couple of reasons -
1. There's no guarantee the software "really" turns off "all' noise reduction, giving an advantage to software that tells you it's off but applies a little anyways
2. There's no guarantee in-camera options "really" turn off "all" noise reduction either
3. It's an official fact that some sensor do on-chip noise reduction, which you cannot turn off and you cannot somehow turn off in the raw file

The bottom line is that there is no "processing independent" way to compare cameras - at least not a way for us regular users to do it. And what we really care about at the end of the day is how the camera performs when we use it, not how it might theoretically perform. Maybe that second one would be interesting, but the fact that it's kind of impossible to test it certainly adds weight to not spending a lot of effort on it.