Ocolon: I don’t need an action cam but it’s interesting to see the return of a 1/2.3" sensor with 8 megapixels due to the demand for 4K. Maybe this will find its way into more products, making people looking for lower resolution point and shoot cameras happy again?
I highly doubt the JPG resolution on a higher MP camera can not be adjusted down to ~8mp if that file size is noticeably more convenient in your data workflow.
microlens elements on the top of the sensors are very good and lose very little light. Pixel wells when not over-exposed capture photons at near-perfect rates. There are some technology gains left to be had (especially at high ISO) but ISO 100 performance is mostly determined by the size of the sensor, not it's pixel pitch. The lens performance is huge but the aperture is still a physical limit the lens cannot circumvent. Sure this is not a diffraction-only limited f2.0 lens but it's still significant to point out what a diffraction-only limited lens would be able to do.
You can use equivalent apertures if you want. It would be easier to do that if the camera manufacturer referred to them as well. Instead of f2.0 if they said f11 equiv, it would be easier to note that f11 equiv gives a 28mp diffraction limit in front of any size sensor. But since they like saying f2.0, I translated them for you.
Noise level per pixel is a completely irrelevant comparison if the two things being compared are different resolutions. Diffraction is also dependent on aperture size, not pixel pitch alone. An f2.0 lens can resolve a lot of detail before diffraction in front of a 1/2.3" sensor. Not 100mp, but more than 8.
Diffraction in front of 1/2.33" gives these approximate resolution limits:f2.0 = 28mpf2.8 = 14 mpf4.0 = 7 mp
Why won't this myth die? Lower resolution does not mean lower noise! Downsample the larger resolution sensor before comparison.
For example, the A7Rm2 has less noise than the A7s
1) is it a fisheye? I suppose it's yes but I don't understand the obsession with this. There are non-fisheye wide angles at similar viewing angles. If it's not, that's a real positive!2) what's the bitrates?
lesnapanda: Or you might try to grab A7 with kit lens. Not necessarily better than the other cameras here, but it is an option.
Sony replaced the RX100 a long time ago and at this point has 3!! other near-identical models on the market with it at the same time. I don't think it's fair to discount the A7. It's available from many places.
They should include that in their test. It's price absolutely means it needs to get compared against an GX8 despite the drastic sensor size difference.
bobbarber: I have an earlier model of this camera, sx230 hs, which I have found incredibly useful. I carried it with me in South America, and when I fell and crushed the camera in my pocket, straightened out the crumpled metal in front of the lens, bought a new rear screen off ebay for $15, and she's still ticking.
The pros are small size, incredible array of zooms, ability to take movies, and CHDK, which works on my camera at least, and probably will on this one soon, giving me raw, 1/100,000 shutter speed, intervalometer, etc. etc. etc.
I'm going through an important life event right now which I need to document, and the trusty SX230 HS was the first camera I reached for. It does the job and gives me the versatility to get a lot of different kinds of shots. I'd glad I have enough common sense not to sneer at cameras like this. Travel zooms are quite cool.
SOME 1/2.33" sensors are lucky enough to be paired with lenses a lot brighter than this and they can indeed make use of 16 or even 20mp of resolution. There should be negligible diffraction (read: no optical barrier) on a 20mp 1/2.33" sensor at an f-stop of about f2.0. Sony is well aware of this, which is why they pair their latest 1/2.33" phone sensor with an f2.0 lens.
This lens is f3.2 at it's fastest. There's no advantage to 20mp over 16mp even at that end of the zoom let alone at telephoto where diffraction should limit things to about 3.2mp worth of information from the light. That said, a more modern 20mp sensor is probably a better and possibly even a cheaper (more sales volume) choice.
SO we have f6.6 1/2.33" lens. That's good for all of about 3.2 mp worth of diffraction free resolution (give or take depending on color, etc).
The TZ100 features a much larger 1" sensor that for a 3.2mp resolution crop (2189x1459 = 3.2mp = 2.5x crop factor) of, wait for it, exactly 625mm (250x2.5)
The SX620HS deosn't offer any reach advantage over the TZ100. It's cheaper indeed but lacks a whole plethora of other features. I just think we should be more clear that if reach alone is the driver, there's little to no difference between the two. 250mm vs 625mm is just a crop away.
Nice to see this camera getting some excitement. IBIS driven sensor shift seems like a real wonder in slower shutter situations. People may not find the AF Pentax lens they need but there are plenty of cheap and juicy MF primes to make good use out of that sensor shift detail...
mosc: what's the time like for 4 captures? Are we talking 1 second or 1/10th or 1/100th for a maximum speed shutter 4-some.
Dynamic range in this era is terrible. Film dynamic range, CRT dynamic range, the kind of quality you see in a movie theater is far in excess of what passes today. We didn't use to describe it in terms of shadow depth and some HDR multi-exposure type edited mess. We used to just call this contrast and color. It's not SUPPOSED to be black. It's SUPPOSED to be a dim color with some contrast to the stuff around it. That's how shadows used to look and how we can get back to them looking through pulling in post. Why? Because an 8-bit display can't possibly fit the shades we're talking about.
what's the time like for 4 captures? Are we talking 1 second or 1/10th or 1/100th for a maximum speed shutter 4-some.
Arizona Sunset: This camera looks wonderful.
Perhaps paired with a faster f1.2 Olympus or Panaleica lens at 35mm FOV, to compliment the Nocticron, even better.
A smallish cheapish 12mm f1.2 m43 lens would make me re-consider the entire system.
Gibraltar01: I'm considering pairing this camera with Panasonic's 12-35mm F2.8 Lens (H-HS12035) as a travel camera. Thinking this would be a good step up from my current Lx5 while still much easier to travel with than an SLR.
Will be spending time in Europe this summer and will want quality snaps of the Family (mostly outdoor) as well as indoor pics of churches and museums...
Yeah, buy an LX100 instead.
forpetessake: I have a feeling, Samyang is not an outlier and the other manufacturers will follow the suit. Sony fe-mount is going to dominate the market, they only need a cheaper FF model for masses under $1000. Affordable body and lenses will quickly capture the market. Canon and Nikon are risking to become third party lens manufacturers. The crop sensor cameras are also going the way of Dodo, so developing new APS-C lenses makes little sense. It makes sense for Sigma, Tamron, Tokina to switch their efforts to fe-mount before the field becomes too crowded,
"MFT F4 is exactly the same as FF f4 when it comes to exposure"
ONLY if you stupidly set two entirely different cameras with entirely different sensors to the exact same ISO simply because you like the sound of that number.
If you use a corresponding ISO in terms of noise, then the exposure of the 300mm f4 m43 (say ISO 200) lens is most similar to a 600mm f8 ff lens (Say ISO 800).
dpr4bb: Compromised lenses galore...
Well these lenses aren't perfect when downsized to "print" resolution either, though I think other than a couple of the Canon's I'd say they're good enough.
I've made that point quiquae on MF lenses too. They actually suck if you wanted a 20mp crop from the middle 1" of them (with I suppose a pair of adapters and an NX mini). Their advantage is over the type of pixel pitches they get used they're phenomenal. A lot of that is due to the size.
photominion: Does the equivalency-comparison account for the fact that the LX100 uses a crop of the 4/3s sensor and not the entire surface at any given time?
1) the LX100's sensor is not smaller than m43 SIMPLY because it is multi-aspect ratio. That would indeed reduce the area but it reduces MORE than that. 2) It still annoyingly doesn't output it's entire light sensitive area in any mode. No raw with circular corners to make your own ratio from, for example. And 1:1 could go higher up utilizing pixels on the sensor no other mode could (but it doesn't).3) So it's clearly a cropped 4:3 sensor no matter how you slice it, but it's still much bigger than 1".4) the lens is much better than you're making it out to be. It's regularly compared to the panasonic 12-35 f2.8.
and not of one of them thicker than 70mm turned off?