Menneisyys: Given that the ancestor, the A5000, had a, noise-wise, *significantly* worse sensor than its big brother (the A6000), I wonder if the switch to using the current (A6000, D5300, D3300 etc.) stunningly excellent 24 Mpixel sensor means the possible A6100 is introduced with a, say, 30 Mpixel, absolutely excellent sensor.
The D5300 uses a Toshiba sensor and D3300 uses a Nikon sensor, neither being the Sony Exmor as found in the A6000 and now A5100.
Kirk Bruner: Alpha? Are they kidding? Did no one at Samsung think this name through? Did they even stop to think if someone might already be using the name Alpha for something? Lawsuit in 5, 4, 3, 2,
Except that outside the US, they won the overwhelming majority of them. Food for thought.
R Vaquero: Copying Apple as usual...
My Android hasn't slowed down in over 3 years of use. With the latest updates it actually feels smoother (not much faster). Bad apps, too many custom widgets (something iOS lacks to begin with), a full memory, Dalvik that hasn't been renewed in ages, it can contribute to a slower phone.
HaroldC3: When they say metal body do they mean metal trim? When I think metal body I think the whole thing.
It shoots 4k video yet it only has a 720p screen. I guess as long as it can output 4k then that's okay.
Even with the typical compression, the details resolved in 4k are still far greater than any smartphone in 1080p. Plenty of comparisons using existing 4k capable phones out there to support that notion. Or try it yourself...And you don't need a 4k screen to see the benefit, even downsampled on a 1080p screen it looks better, or on anything inbetween 1080p and 4k.
Boerseuntjie: Wow A7s is amazing at ISO 25600 JPEG's
There is no reason to assume on my behalf when:A) you often make vague claims contrdicted by available data and always fail to support those claims by sharing your files and test results.B) you just confirmed not using a perfectly controlled comparison setup ( no tripod but matched lighting on different days... sounds legit..)C) and most importantly the actual RAW files provided by Dpreview again prove your claims wrong, with a good margin too. See below.
Same for your A7s claims vs the Df that have been debunked countless times, even made easier by the fact that you don't believe in resizing (according to yourself).
kecajkerugo: it is interesting to see that Fuji X-T1 has less noise the Sony, even at ISO6400, unfortunately it is impossible to gather higher iso on the Fuji with the RAWs.
Despite slightly overstating its ISO's and lots of RAW smoothing, the X-T1 is still quite a bit noisier in low light. See here:http://i59.tinypic.com/2db12k5.png
Up to 2 stops difference in shadows.
Unless you shot those RAW files side by side in a controlled environment with the exact same physical exposure, which I am sure you didn't, any conclusion would be worthless.
aandeg: Why does the panasonic have so much noise? The A7s raw at iso 3200 looks like the Gh4 at iso 400.
As you move further up the ISO scale, the difference climbs well past three stops in shadow noise due to much lower read noise and better efficiency per unit sensor area too.
Sdaniella: so, the Sony A7S has video with Full M/M ISO (ExpSim LV) control like FF EOS LV Canons do, but Pany GH4 does not... ok, got it. Oly video is stuck in Auto-ev0/autogain LV (perfect, though for P&S videographers)
wait ... auto-iso isn't available ... hm, ok ESLV is available, but adjustable, but P&S autogain (auto-ev0) mode isn't available ... ok, now, I got it ... mimics most digicam autogain/auto-ev0 video modes, which lack full M/M ISO control ... how ironic.
btw, if you want on-the-go compensation adjustment of iso in M mode with auto-ISO, it's no different from making on-the-go adjustments with M ISO directly. in both cases you are bypassing auto-ev(pick your value) altogether anyway
PS: read out and reset speed are connected through the same clock and in this case it's specifically the reset scanning time that directly influences the usability (and thus availability) of EFC.
Sony has at least 2 patents themselves describing EFC systems to cover their implementations. Second, read out speed has everything to do with electronic curtain (both first and second) and the larger the sensor and the more pixel lines/rows, the more crytical it becomes. Only the third iteration of the 36MP Exmor (Sony design directly based on the 16 MP APS-C sensor first found in the A580/A55) is capable of supplying an EFC. And it's also the one with the fastest read ouf of all 3 generations.
No, the A7R not having EFC has everything to do with read out speed, nothing with Nikon.
peevee1: DPR wrote: "Unlike the majority of cameras, the a7S offers full-sensor readout - its 1080p video is made up from all the pixels within a 16:9 crop of the sensor, then downsized to 1080 resolution."
I am pretty much sure it is not so - it is not 16:9 crop of the whole sensor (which would be 4240 x 2385), but 3840 x 2160 crop of the sensor, which then simply downscaled 4:1.
Also, can you check that no line skipping happens at 1080p30 and 1080p60? At least in the latter case (and MAYBE in 30p), I am pretty sure every second line is skipped.
Precision in the things like that matter - just parroting manufacturers' advertisements is not good enough.
It doesn't skip in 1080P30, as also specified in the official specs.
steelhead3: Not a Nikon 1 user, but I thought the review may be a little harsh...this camera has possibilities if it came with the 20 meg Sony sensor, which Nikon can do easily since Aptina is into automobile products now.
And btw, the NXMini uses the same Sony sensor, but applies horrible RAW NR (smearing) at higher ISO's.
And we can all compare the RAW files ourselves and notice the V3 performs worse, also at higher ISO's with a $900 lens. No amount of reliance on unknown or private sources will change the simple facts staring us in the eyes.
57even: Hmm, can't see many early adopters. Nikon, get a decent sensor, quick!
Better still, put all this excellent technology in an APSC body. You would probably sell millions.
The RAW files are not better than the Dpreview claims which they supported with samples shot in various light conditions (typical deep shadows in typical indoor light included, unlike Imaging Resource). Shot with a $900 lens too, so no excuses possible there either.
Relatively high shadow noise at all ISO's and an early colour shift in shadows at higher ISO's.
HowaboutRAW: Let's hope Google unhides the "limit background/crapware" feature that Google hid with the move from Android 4.1 to 4.2 and higher.
Plenty of free apps out there to tell you what's running in the background and what's loaded at startup. Including control over those apps. The feature you were referring to is for cached programs anyway, which has little relationship to CPU power being used. My cached apps rarely use more than 5% CPU, mostly well below 2%.
If 10 seconds of effort to unhide developer options is too much work....
And for recent phones high on (free) RAM, there are more downsides to limiting the standard amount of cached programs (such as much longer load times of frequently used apps and higher battery drain as a result) than benefits.
Stu 5: Why on earth compare it to the 5D MKIII when everyone knows the 6D is the better Canon camera at high iso of the two.
" So far, the A7s isn't an option to choose on the DPR Noise widget"
A7S in RAW vs the best sensors available, minus the 1DX.
igor_s: Simulation of higher ISOs by brightening in the post decreases the SNR, therefore, the comparison is not fair. However, the 7S has lower level of shadow noise, and therefore should do better at extra high ISOs (where the sensor is poorly illuminated). Perhaps natively by about 1 eV if in "boosted" ISO tests it wins by 2 eV.Unfortunately the above applies only to ISOs above 51200 (if you ever need it and satisfied by the quality). At lower and moderate ISOs the A7S loses to the a7R even in shadows. A specific ultra-low light camera.
Practically every camera does digital boosting of sensitivities at the highest ISO's. For just noise comparisons at extreme ISO's, there's little difference whether done in camera or in post processing.
Stephen123: There’s a pretty big Sony vs. Canon question in what color those arches bait up and left of center look to the human eye. Sony has them white and a bit blown out. Canon has them warm and detailed. If they are cream or yellow to the human eye then Sony is screwing up. If not, Canon is screwing up. It’s not like it’s a small difference.
You do realize that you're actually judging the Adobe calibration for preliminary support of the A7S?