Jacket-pocketable, APS-C, up to 75mm eq. focal length?

[ControlD]

^^^

Dang. Thanks for the update though. I was really hoping to hear the opposite of what you have found (would LOVE to ditch my heavy DSLR stuff) but appreciate your findings.

 

[Syborg1211]

Small update:
Been comparing some more night shots between the RX100 II and the A6000. In all honesty... it doesn't look good. I fail to see much of a consistent advantage in terms of low light performance from the larger sensor at all. Under equivalent conditions the RX100 II will be able to shoot at a lower ISO and/or higher shutter speed. The aperture advantage of the RX100 is noticeable, the sensor size difference is not.

It's unacceptable to me. There is hardly any advantage to lugging the A6000 and its lenses around. The only pro is the ability to use lenses with shorter or longer focal lengths than the RX100 can manage at acceptable F-stops, but you've got to change the lenses constantly which is a major pain.

I swapped the gimped pancake zoom lens for a Sigma 19mm f/2.8 prime lens that gets great reviews, but again the difference is marginal if noticeable at all. The pancake zoom was plainly worse than the RX100 II at night.

Another swap I've made is to trade in the 55-210mm for an 18-200mm Tamron lens. Still gotta test that one, but so far the sharpness seems slightly inferior. Whether its worth it for the extra versatility is to be seen.

I've had some good results from my daylight shooting in the weekend, but the RX100 performs great there as well.

This thing probably requires Carl Zeiss lenses to properly perform at its potential. Don't bother if you just plan to use budgety ones.

As it stands, I'm ready to throw in the towel and return the A6000 on Saturday. I'm pretty sure what I'm suffering from is mainly a luxury problem; that RX100 II is a force to be reckoned with. But I just can't recommend a camera like this to anyone when there are so much more convenient options available.

Yea, I looked into the sony line for a friend last year, and my conclusion was that the lenses just aren't there.

However, were you comparing jpegs out of camera or what were the conditions of your comparisons? Remember to note that since the sensor is smaller, equal ISO numbers doesn't mean equal noise. Compare the raw files and perform your own noise cancellation on both night shots to see which has more potential.

 

[blastingcap]

Yea, I looked into the sony line for a friend last year, and my conclusion was that the lenses just aren't there.

However, were you comparing jpegs out of camera or what were the conditions of your comparisons? Remember to note that since the sensor is smaller, equal ISO numbers doesn't mean equal noise. Compare the raw files and perform your own noise cancellation on both night shots to see which has more potential.

He's comparing slow lenses probably stopped down what do you expect?

Try taking a photo wider than 28mm (or 24mm for the RX100III). Oh wait you can't. Try the a6000 + 50/1.8 OSS ($249 on sale, $299 regular) vs RX100 and see which gives more subject separation. Try the $399 Rokinon 12mm f/2.0 and compare to the RX100.

To OP: Don't buy the a6000 if you are happy with RX100. Because frankly you aren't using it to its full potential anyway and there's nothing wrong with just buying what you need in the smallest form factor possible.

The a6000 is meant for people who care about things like interchangeable lenses, long exposures with remote control, EVF, more buttons and dials to customize, much faster autofocus (especially AF-Continuous tracking), faster processing of in-body HDR or bracketing, etc. If you don't care, then an RX100 might be a better fit for you.

 

[Rdmkr]

Welp, the deadline has passed I haven't returned the cam. I don't know what it is about this thing that messes with my mind but I just can't bring myself to return it. No idea what else I'd spend the money on and I do want a better camera, even if this one's a bit of a disappointment relative to my expectations.

The 18-200mm Tamron lens has been fun to use. At least in daylight this removes the need to constantly switch lenses. At the telephoto end it's hard to see a difference in quality with the 210mm Sony one. The sharpness is noticeably inferior to that of my prime lenses but it's all far from unacceptable. It's been a good purchase.

I've got my sights set on two Samyan Rokinon prime lenses. The first is a 12mm ultra wide f/2.0 lens, the second an f/1.4 !! 24mm. Their price is beyond competitive. The catch? They're manual focus only. I'll have to see if that's something tolerable or a deal breaker. Anyone have experience with lenses like that?

I'll probably sell my other prime lenses if these turn out to be good.

edit: oh, another catch: the f/1.4 lens weighs almost 600 grams; think I'll stick with the wide angle one, cause that's just pushing it.

 

[blastingcap]

Welp, the deadline has passed I haven't returned the cam. I don't know what it is about this thing that messes with my mind but I just can't bring myself to return it. No idea what else I'd spend the money on and I do want a better camera, even if this one's a bit of a disappointment relative to my expectations.

The 18-200mm Tamron lens has been fun to use. At least in daylight this removes the need to constantly switch lenses. At the telephoto end it's hard to see a difference in quality with the 210mm Sony one. The sharpness is noticeably inferior to that of my prime lenses but it's all far from unacceptable. It's been a good purchase.

I've got my sights set on two Samyan Rokinon prime lenses. The first is a 12mm ultra wide f/2.0 lens, the second an f/1.4 !! 24mm. Their price is beyond competitive. The catch? They're manual focus only. I'll have to see if that's something tolerable or a deal breaker. Anyone have experience with lenses like that?

I'll probably sell my other prime lenses if these turn out to be good.

edit: oh, another catch: the f/1.4 lens weighs almost 600 grams; think I'll stick with the wide angle one, cause that's just pushing it.

12/2 is a great lens. Who cares if it's MF, it's an ultrawide. If you want to use it brain-dead just set it to the relevant hyperfocal distance depending on what aperture you most use (mark it with a piece of tape or whiteout or whatever you want, on the distance scale if you want). (NOTE: The distance scale is non-linear so you may want to do some trial-and-error, taking some test shots, before marking the hyperfocal distance on the lens barrel.)

http://www.dofmaster.com/dofjs.html

For an a6000 + 12/2 the hyperfocal distance is about 12 feet @ f/2.

At f/8 the hyperfocal is about 3 feet.

 

[Rdmkr]

Cool, sounds like a lens I can't go wrong with then. I'll be ordering it later today.

By the way, I did a few more "empirical" comparisons of the quality between the RX100 and the A6000 and it looks like my subjective estimates were a bit too harsh on the A6000. The difference may be subtle to the naked eye, but it looks like it all matches up quite well with the expectation that an APS-C sensor has a two stop advantage on a 1 inch sensor based on their metric area. There just two things that made it less obvious:

- the A6000's image quality is usually about as good at any specific ISO as the RX100 is (at least) one ISO setting down.
- the RX100 seems to have a default exposure compensation that is roughly one stop down from the A6000's default level.
- the RX100 wins about 1 ISO level when used at maximum aperture compared to an f/2.8 lens on the A6000

It obscured the difference in quality between the two from me at first. Superficially it sometimes (at night mostly) looks like the RX100 can make do with much lower ISO's and and higher shutter speeds, but the quality of the result is not the same when carefully compared one on one. Apologies for the false alarm.

I can post some pictures for comparison later.

edit: also one thing I had to learn is that ISO 6400 is a weak setting on the A6000 (confirmed by Techradar's test results). Using ISO 3200 or 12800 instead tends to yield better results.

 

[Rdmkr]

Another update. The 12mm Samyang lens does pretty much as expected. It's good as an ultra wide angle lens but for general low light purposes it's not really that useful. The angle is just too wide make anything at a typical distance look very detailed. It's probably not very sharp at F/2.0 to begin with. Manual focus isn't really an issue with this lens because you reach the point where everything is at the hyperfocal distance very quickly, even at F/2.0.

I stand by what I wrote in my last post, but I can't seem to manage to reproduce the positive results in conditions of actual night-time with street lights. In such conditions, the RX100 II is better by a hair than the A6000 with Sigma f/2.8 lens. That's right. The APS-C system camera with well reviewed prime lens is beaten by the pocket sized point and shoot...

My feeling of disappointment has been intensifying and I'm considering trading in the camera and its lenses for something more serious, even if I have to accept a slight loss on the trade. Luckily the shop I bought it at allows such late trade ins. Right now I'm considering getting the Nikon D7100, an APS-C DSLR on a similar sensor quality level as the A6000, but with much better lenses available for it. I'd stick to getting two lenses, a Nikon 18-300 f3.5-6.3 superzoom and a Nikon 30mm f1.4 prime lens. Reviews for both of these look very good. I'm thinking this should get me the quality boost I want both in low light and broad daylight conditions, even if I have to part with some portability and some extra cash to get there.

I'm left with the impression that the A6000 is an awkward model. It ticks a lot of the right boxes when it comes to sensor quality, but it's stuck with a lens line-up designed for cameras that just aren't in its league, on top of being held back by the design goal of keeping things within the limits of "barely portable". There is something weird about the inconsistency between its JPEG and RAW quality, too. I'm speculating it relies on a lot of internal image processing to squeeze out the JPEG quality it manages to, and when it comes to actual optical quality it's way too big for its britches. The supposed quality in the JPEG files is probably also something that fools benchmarks better than it fools the human eye.

 

[CuriousMike]

I don't know where you live, but if possible, look into renting the D7100 + the 18-300 for a few days.

I think you're onto the right idea: Find the lens you want, then buy the body that goes along with it.

I don't own an 18-300, but my feeling is it's a step-up lens - one level above the quality of kit lenses. I think you should keep your expectations in check with it.

 

[jtvang125]

According to dxomark the sensors on those two cameras are very close in performance. If you're expecting a huge difference I think you're going to be disappointed once again. Indeed you will have a much larger selection of lenses with Nikon but good glass is going to be expensive no matter what system you go with.

 

[Rdmkr]

Dxomark also has measurements for the lenses and that's where the difference is massive. The 18-300mm lens gets almost twice the amount of detail at 300mm as the 18-200mm e-mount tamron gets at 200mm. The 30mm lens has f/1.4 aperture and is sharp at that setting. I really don't think I can go wrong with these. But I'll look into renting one, that is a great idea.

Really, I'd be able to tolerate the A6000 if there were good zoom lenses available for it, but if weak 200mm ones are the best I can get, no thanks.

 

[blastingcap]

Dxomark also has measurements for the lenses and that's where the difference is massive. The 18-300mm lens gets almost twice the amount of detail at 300mm as the 18-200mm e-mount tamron gets at 200mm. The 30mm lens has f/1.4 aperture and is sharp at that setting. I really don't think I can go wrong with these. But I'll look into renting one, that is a great idea.

Really, I'd be able to tolerate the A6000 if there were good zoom lenses available for it, but if weak 200mm ones are the best I can get, no thanks.

You can't compare overall numbers, which are meaningless, first off. Look at individual field maps.

Secondly, you can't compare cross-system. DXOMark even has a disclaimer about this. There are too many variables like AA filter strength, pixel density, etc. to make blanket statements about how one lens for system X is twice as sharp as another for system Y just because of one number.

Also the Samyang 12/2 is pretty damned sharp for an ultrawide, even at f/2.

 

[Rdmkr]

Got the swap planned on Thursday. I'm losing 160 euros on the deal, which afaic is a relief. Could have been a lot worse.

Current plans for what I'll be getting:
- Nikon D5300 (just as good as D7100 sensor wise but comes with swiveling screen, updated processor and a lighter body; don't know how these price tags got swapped but I'm taking advantage of it)

- Sigma 18-35 f/1.8; the reviews on this are insanely good. Sharper than good prime lenses across its zoom range and the aperture width is incredible. Zoom range is useful with an aperture like this too.

- Nikon 18-105 kit lens; really can't overlook this one because its sooo cheap and very sharp for a zoom lens according to dxomark.

- Nikon 70-300; this one's faster than the f/6.3 Nikon 18-300mm and considerably sharper according to some benchmarks I've seen. Should be good because dxomark's scores for the 18-300 were already pretty good. This also avoids a risk because I've only seen precise scores for the bulky f/5.6 18-300 and have been presuming from half-reliable comparisons only that the smaller f/6.3 is similar.

 

[Rdmkr]

I've got the Nikon D5300 now. The lens I got with it is the lightweight 18-300 f3.5 to f6.3 because it's convenience trumps that of everything else and I think any lens that goes up to 300mm will already be overkill in terms of telephoto detail so I don't need the ideal option.

I'll pick up a lightly used 18-35mm f/1.8 sigma lens on Saturday.

I've only tested the camera a bit at night with street lights, which obviously isn't ideal with an f/3.5 maximum lens. That being said though the difference with the A6000's f/2.8 isn't all that big, and this lens has pretty advanced image stabilization. Sharpness of the lens at wide-angle is probably just OK but that's not likely to be a bottleneck at night. Most pictures got made around ISO 6400 at a 1/20 shutter speed with 0 exposure compensation. I'm not expecting groundbreaking quality at such settings but something that approximates the A6000 at similar ones shouldn't be too much to ask.

Maybe this isn't something I should be surprised by, but the results have been pretty bad. I'm looking at the jpeg files because I don't have the time or expertise to post-process much. But the difference between these and the A6000 is not just a nuance.

I probably need to pick up on some post-processing skills to make a fair comparison, because it's clear this camera leaves a lot more of the work to the user than the A6000 did. DXOMark assures me the Raw quality is on a comparable level, so lets hope there's more I can squeeze out of these pictures.

The above is of course very inconclusive since I've only had one evening to do the testing and I've not had time to master the camera's controls. Still, first impressions matter.

I'll be testing the 18-300mm in daylight on Saturday. I do expect it to shine there. This is by no means a lens designed to handle low light conditions well. Testing of the 18-35mm sigma should happen on Saturday as well.

 

[Rdmkr]

http://www.dpreview.com/reviews/nikon-d5300/17

The graphs on this page (click Graphs to the right of Cameras Compared) tell you the whole story. With Noise reduction turned on, the A6000 gets insanely impressive SNR results. But with it turned off, it underperforms my D5300 slightly.

I've definitely noticed the huge difference in noise performance between the two with NR turned on.

Guess it's time to figure out if the A6000's NR voodoo can be reproduced with post-processing. If not, I may have made a mistake trading it back in.

RX100 II has incredibly good noise reduction too:
http://www.dpreview.com/reviews/sony-cybershot-dsc-rx100-m2/11

Another one of those graphs that I really should have seen sooner.

 

[Berliner]

I'll be testing the 18-300mm in daylight on Saturday. I do expect it to shine there. This is by no means a lens designed to handle low light conditions well. Testing of the 18-35mm sigma should happen on Saturday as well.

Sorry, but that lens won't shine under any conditions, unless you mean shine from CAs.

The 18-35 will be much better, except it does not have VR and you will have very limited DOF at f/1.8.

I'm not sure what type of photography you do under the lighting conditions described, but a tripod would probably help you a lot, if what you want is great image quality. If you are happy with the 18-35 and don't want to lug around a tripod and a big lens, look at Nikon's 35mm f/1.8.

Also the 5300 is among the best what APS-C has to offer, in terms of sensor performance. Do RAW and try different, more serious, denoising programs than the internal engine, if you HAVE to shoot in that light constantly. (The A6000 has pretty much the same sensor though, I guess.)

The question is: What kind of pictures do you expect in such dismal light?

 

[CuriousMike]

I don't know if it's been mentioned, Rdmkr, but I think your expectations are out of sorts.

The A6000 and D5300 are both praised as state of the art for DX sensors when it comes to noise.

My guess is you're not going to find the RAW files giving you the jump you're looking for.

I'm not impressed with about any NR software I see - it just muddies the image.

I think you'd be served by renting a Nikon D4 with one of the holy-trinity lenses and shooting that at ISO 6400.

 

[Rdmkr]

All I want is results that are noticeably better than what a pocket camera with a 1 inch sensor produces. Is it really too much to ask? If I carry around 1KG of extra weight and pay almost $2000 why would I demand anything less?

The RX100 II doesn't even have a particularly great lens according to DXOMark, apart from its wide maximum aperture. It's Carl Zeiss in name only. I really don't think this sets the bar all that high for an APS-C camera.

I'm testing at night because low light is one of the few areas where quality differences between cameras matter a lot. My RX100 already does daylight photography at short focal lengths fine. I'm not under pressure to find a replacement there.

Anyway, I shot at ISO 3200 at -1 exposure compensation today and got quite a bit better results. I've got to remember that this camera is a lot less forgiving about shooting at ISO's above 3200 than the RX100 II and the A6000 were. There's none of that aggressive noise reduction to fall back on. I find that night pictures look a bit unnatural at 0 exposure compensation anyway. It's like everything becomes light emitting then. There's nothing wrong with a night scene that looks dark in the picture like it is in reality.

Thankfully high ISO performance shouldn't be a factor with the Sigma. I've got pretty good hopes about what it will do. With the 1.7 stop aperture difference, I should easily be able to shoot at ISO 1600 in that light. And its sharpness should be incredibly high.

Sorry, but that lens won't shine under any conditions, unless you mean shine from CAs.

I'm happy to get any pictures at all at 300mm. Like I said, even if the detail is reduced to 3MP like on tamron lenses, 300mm is overkill. I have the sigma for quality images and the 18-300mm for convenience and telephoto reach. They should complement each other pretty well.

By the way, I've done my research and you're just wrong about the CA. I don't doubt this lens has its set of flaws, but CA is not among them. Here's the DXOMark review that just got published (switch it to the D5300 and click the measurements tab):

http://www.dxomark.com/Lenses/Nikon...0mm-F35-63G-ED-VR-mounted-on-Nikon-D5300__919

On a D5300, it gets up to 6 megapixels of detail at 300mm which is great for a telephoto lens under $1000; you gotta have realistic standards when it comes to these focal lengths. And the CA is very minimal.

+/- 12MP at wide angle is nothing to sneeze at either. That's a sharpness level worthy of a (cheap) prime lens.

 

[Berliner]

All I want is results that are noticeably better than what a pocket camera with a 1 inch sensor produces.

You will get that. As you would get even better results on a current gen FX, but remember that you asked for jacket-pocket size.

As I said, try RAW and different converters to control noise and detail to your taste.

By the way, I've done my research and you're just wrong about the CA. I don't doubt this lens has its set of flaws, but CA is not among them.

That was just to go along with the "shine" metaphor, I did not actually research the lens for CA values and if it has low CA thats good. But it hardly matters with digital anyway.

I mean, you can be happy with a lens like that. But I would never consider a super zoom if I'm trying to get image quality and shoot in the dark.

Unfortunately you have not answered what type of shots you plan on doing.

 

[Throckmorton]

12/2 is a great lens. Who cares if it's MF, it's an ultrawide. If you want to use it brain-dead just set it to the relevant hyperfocal distance depending on what aperture you most use (mark it with a piece of tape or whiteout or whatever you want, on the distance scale if you want). (NOTE: The distance scale is non-linear so you may want to do some trial-and-error, taking some test shots, before marking the hyperfocal distance on the lens barrel.)

http://www.dofmaster.com/dofjs.html

For an a6000 + 12/2 the hyperfocal distance is about 12 feet @ f/2.

At f/8 the hyperfocal is about 3 feet.

You can also use focus peaking which makes manual focus really easy

 

[IAteYourMother]

rx1

 

[Rdmkr]

Finally some good news. Now that I've tested the Nikon 18-300mm in daylight and the Sigma 18-35 at night, I'm getting good results across all conditions.

I'm really glad I didn't stick with the Tamron 18-200mm on the A6000; there is such a world of difference between that and this Nikon. The pictures are more detailed, less distorted and at a greater maximum focal length. It's a complete paradigm change.

It's not even that much more expensive or bulky. And the optical viewfinder is a great luxury to have, not to mention the kickass battery life it enables.

The Nikon lens exceeds my expectations on a lot of fronts. What I'm really surprised by is that its wide angle sharpness is so good that it hardly seems worth it to switch to the Sigma for extra detail. This makes it possible to leave the big and heavy sigma at home entirely and walk around with a single all-purpose lens without feeling like there's a huge quality compromise. That goes for days with bright sunlight only, of course.

The autofocus of the D5300 is noticeably less advanced than the A6000's, but so far it's never ruined a shot of mine.

With the sigma, shooting at ISO 1600, I get good night time pictures... Probably better than both the A6000 and the RX100 and certainly without pulling as many noise reduction tricks out of its hat. The aggressive noise reduction of the Sony cameras seems to kick in around ISO 3200 so if I stay below that, it isn't a big factor in the comparison.

Carrying the D5300 hasn't been a problem either. It is easily concealed under a jacket when hanging from a shoulder strap. I'm carrying a lot fewer lenses so the total bulk is comparable. If I carry only one lens with me, the weight is actually lesser. And with this camera, doing so comes at much less of a compromise than it would have with the A6000.

I ended up spending more on this kit than I had hoped, but I'm also happier with the results than I'd hoped.

 

[Berliner]

Great to hear that you are happy with your gear now.

I just wonder what makes you think the AF is "less advanced". Did you read what the manual has to say about it? There are a lot of AF settings.

 

[CuriousMike]

I just wonder what makes you think the AF is "less advanced". Did you read what the manual has to say about it? There are a lot of AF settings.

That comment stuck out like a pink unicorn to me.

 

[Rdmkr]

It's widely known that the Sony A6000 has an extremely advanced autofocus system for a camera at its price point. It has 179 AF points versus the D5300's 39. Even the D7100 pales in comparison to it with its 51 AF points. The A6000 also uses much more cross-discipline sensors capable of phase detection. On that note I don't think that comment needs much clarification.

 

[Berliner]

So you are saying more claimed AF points make an AF system necessarily better?

There are a lot of other factors going into autofocus and just because Sony claims theirs is better, you should not believe it as well. Just look at the Nikon 1s for example, which are faster. And don't get me started on AF point selection and motion tracking.

As I said, set your system according to your needs and you will have nothing to complain about.