Superka: Hey, Sony sensors lovers! Sony sensors are little better in terms of noise, but much worse in terms of colors. So stop please! Canon is the best!
Depends on the toppings. Sony cameras usually have the best color separation. Canon struggles the most with reds, a sacrifice made to improve their low light performance. To get an idea, go to the full sensor reviews (not comparison pages) at DXOmark and look at the "colour response" tab.
facedodge: which of these can achieve the most compression and shallow depth of field for good subject isolation/background blur, Sony RX100 or Canon G1X?
Up to about 50mm (35mm equiv.) they are within 1/3 of a stop, above that the G1X has an advantage of up to 2/3 of a stop in terms of shallow(er) DOF, plus a slightly longer FL at the long end.
cgarrard: No finder or finder choice, no hotshoe, no grip, no image stabilization, flat buttons, first things I think. I can't get around those much. Sure it has a bigger sensor and decent looking lens, but holding and using a camera makes more of an impact to me. Had it a front grip and I.S. I might give it a pass, but not in today's market and the year is still young. Surely Sony aren't the only ones that are going to make big sensor compacts. Just watch and wait.
If Cgarrard doesn't like it and predicts it will fail, be sure to buy one before they are sold out. He has a taste for oddball market failures.
jonikon: This Sony RX100 is a BIG improvement over all the other truly pocketable cameras and should be well received by many. Cameras like this should have been made years ago, as the demand has been there for some time now. Although the RX100 is definitely a Canon G1X killer, it is not perfect however. I would like to see Sony add:1. an EVF2. Phase detection auto focus for acceptable continuous AF of moving subjects.3. Less megapixels. 10MP is enough, but 20MP is unnecessary and results some IQ issues (like color accuracy and diffraction limiting, noise reduction smearing), that could have been avoided with a 10 or 12 MP sensor.4. A way to remove the lens for sensor cleaning.5. Lower price.
That said, I think the RX100 is good enough to take away a lot of sales from their NEX line of cameras that are definitely NOT pocketable with a zoom lens attached, and offer little more than lens interchangeability over the RX100.
Plenty of documentation. Even the ancient Sony R1 had thermal noise already very well controlled back in 2005, see the Dpreview test that specifically tested this.Sensorgen.info will tell you how read noise as a total (thus including thermal noise) contributes to the noise figures from many sensors.
And the differences between the D4 and D800 at ISO 12800 are minimal for a given output size (different sensor makers and technologies too btw), about a third of a stop. Plenty of documentation about that from Bclaff's tests, DXOmark and normalized RAW samples from Imaging Resource.
Cold electricity would be a straw man and more pixels per square mm "confusing things" is a very confused statement to begin with.
HowaboutRAW: Higher ISO capacity and a lower pixel count would be an improvement. Might quiet those clamoring for an hot shoe too. (PS: the Panasonic LX5 has a hot shoe.)
Sony made the same kinds of mistakes with the A77 and the Nex 7--at least wide open this RX100 lens is fast and of course likely to be much better than a G series lens.
Imaging Resource has RAW files up, DCRAW ( Rawtherapee in my case) opens them.
Actually, you just demonstrated you have no idea what you're talking about. Thermal noise is a tiny factor these days. It's part of many factors contributing to read noise and the latteras a TOTAL is already a tiny factor compared to shot noise.
And no, by my logic the D4 would NOT have a 30MP sensor, because:
A) there wouldn't be a space for the D800 (2 different cameras targeting 2 audiences with an overlap is more sales than one camera)B) 30MP @ 11 FPS. Think about the buffer you just mentioned, even assuming the sensor could sustain these read out speeds to begin with. Long sustained shooting at these speeds is much more important for pro sports shooters than people buying cameras like the RX100.
Erik Neu: Time lapse !Great video capabilities !No lens cap !210g !
Perfect.Just the features I am looking for.
Timelapse recording: yes
It doesn't seem any worse at high ISO than a J1/V1 and the latter 2 have pixels twice the size and an efficiency per area similar to the best sensors available (D3s/D4). Think twice before you're expecting a suggested 12MP version of the RX100 would have done that much better (and thus beat the efficiency per area of a D3S and D4). I'll happily take the extra resolution.
For those that shoot RAW and provided the camera is fast enough (which it seems to be) to handle these RAWs and your computer is too (welcome to 2012), more MP has no perceptible downsides, with or without diffraction. Only benefits.
Sdaniella: I have hit the limits on my 5DMkII for extreme low light shooting (moonlit or not at night) to which ONLY the 5DMkIII is up to the task.
All other rivals, fall short, either in resolution in such conditions (D4), or poor high ISO noise in such conditions (D800); so neither could meet my requirements of 'improvement' over my 5DMkII as the 5DMkIII solely has.
There is no other, and the closest to exceed for less noise in extreme low light at high ISOs is Canon's own 1Dx, but it has less resolution.
Canon has clearly hit the 'max' technical development for sensor 'sweetspots'.
Are you trying to make people believe that you actually tested the MkIII directly against the D800 and D4 under such conditions?Because most people have figured out by now that you haven't.
Dave Oddie: Sony A77 review:
"However, the somewhat mushy JPEG image quality, high noise levels and slightly laggy operation stop the A77 from getting a gold award. "
EOS 5 D MkIII conclusion:
"Destructive noise reduction results in mushy JPEGs, even at base ISO"
Show some consistency DPR or lose credibility.
@Parappaman:High shadow noise levels at base ISO, see page 23. Which leads to almost a stop and a half less DR than say... an APS-C A77 WITH fixed mirror (extra half a stop light handicap).
Tim Ashton: So what is DP review hiding???Normally on he page showing dynamic range (page 19 in this case) there are comparative wedges below the chart which highlight the performance of the camera under review with its immediate competitorswith the 5D3 those comparative wedges are missing. everyone knows that the D800/exmore sensor wins here, but to hide the results only suggests the difference is worse than it actually is in real life; or is it?It is a pity that DP review no longer has the courage to be truly objective (See comments by Taikonaut)Did Canon/Nikon get in Amazon/DPreviews ear??We al lose:(
Correct. It's one of the biggest omissions in their reviews, although as said page 23 highlights some of the DR handicap at low ISO. What baffles me more is that none of those findings returns in the conclusions, especially knowing that all crop cameras carrying Sony sensors already perform much better in this area.
The DR comparisons by Dpreview on that page typically reflect jpeg contrast/gamma curves and do not show sensor DR, which tends to be much larger than any jpeg can display. Page 23 gives a small hint of the sensor DR by comparing shadow noise.
Dpreview skipped the findings on page 23 in their conclusions? The Issues with RAW lattitude as displayed there would clearly classify as a con vs competition.
Raist3d: Not sure what the point is of this ISO 12800 shot (and the next one), which are shot at slow apertures and rather high shutter speeds… ?
The same issue with the EM-5 gallery (ISO 6400 with 1/4000th shutter on two shots. What is that supposed to show? )
Noise for a given visual exposure. Contrary to popular believe, the amount of light in a scene doesn't have much influence on noise levels. So the same scene with much less light, same ISO and slower shutterspeed plus larger F stop, would have roughly given the same amount of noise.
Ashley Pomeroy: Has to be said that the image quality, on a per-pixel level, doesn't seem any better than the old Olympus C8080 wide zoom from 2004, or the Sony F828 from the same era. The HX200V produces a large file, but it doesn't look as good. The only real advance seems to be the longer lens and image stabilisation; I'm surprised that the technology has stayed static for eight years. Perhaps there's a fundamental limit to what you can do with small sensors, and it was reached a long time ago.
@peevee:especially considering the fact that the sensor is many times smaller than the one in the C8080.
TrojMacReady: Either Dpreview forgot to turn off default chroma NR in ACR, or Pentax now officially cooks its RAW files at all ISO's and more than ever, rather than the previous smoothing that kicked in above ISO 1600.If the latter, it seems to have fooled atleast a few people already, into thinking it's producing better RAW files.
Read this post from Dpreview:http://forums.dpreview.com/forums/read.asp?forum=1036&message=41407104
And for the record, while Pentax uses pretty effective NR, I can achieve better results with Lightroom and a little Topaz when using the output from other cameras carrying this sensor.
Edmond Leung: Image quality of OM-D is much better than this K-01.
Read this post from Andy Westlake:http://forums.dpreview.com/forums/read.asp?forum=1036&message=41407104
This is about testing consistency. In general, the output from all cameras looks better with chroma NR left to default in ACR/Lightroom, rather than off, which is what Dpreview --> normally<-- uses. Apples and oranges applies here.
You're talking about a Sony camera from many years ago.If cameras were to remove all chroma NR by default, you'd miss out on a lot of red channel detail (which this test scene sadly lacks, see Imaging Resource).
Anyway, the RAW files as posted here show that Dpreview forgot to turn off the default NR in ACR, which they did with all other cameras.