http://www.popco.net/zboard/view.php?id=dica_review&no=810
Added on August 14, 2012, 6:21 pmHow I estimate LX7 sensor performance?
We will use dxomark extensively. It is the best reference while many argue the accuracy. But I have downloaded many sample raw pics from dpreview to analyst many cameras from 5d3 to XZ1 and I can confirm that dxomark is very accurate. We will be very interested in ISO score and we will talk a bit on dynamic range score. We are not interested in dxomark "overall score" which is meaningless.
The sensor size of LX7 is 1/1.7", same as Olympus XZ1, Canon G11/G12/S95/S100 and Nikon P7100 and slightly smaller than the 1/1.63" of LX5.
The biggest different this sensor with them is that it is a CMOS while all others are CCD (except S100/G12). CMOS tends to have better noise performance than CCD. That's why full frame Leica M9 CCD sensor has a poorer ISO performance than a smaller APS-C Nikon D7000.
And for the first time, LX7 is not using a panasonic sensor (strange for a company that made their own sensor). This is told in the launching interview in Taiwan. Their excuse is that they don't have such size.
Sony made sensors for many great camera brands such as Nikon and Pentax. The best APS-C sensor today are found in Pentax K5 and Nikon D7000, which is made by Sony.
Pentax Q has a sensor half the size of LX7. It is a CMOS and is believed to be made by Sony. The sensor, albeit smaller, outperform all 1/1.7" CCD sensor in ISO performance and dynamic range!
http://www.dxomark.com/index.php/Cameras/C...tabase/Pentax/Q
Pentax Q has a score of 189 for ISO performance.
That means if we use the same technology and put the same amount of megapixels (12) on a double sized sensor. It will 100% sure to be better!
How I estimate RX100 sensor performance?
Sony made the best sensor in the world. And for marketing reason for consumers (and 'pros' that think they know something but actually they don't), they like to stuck a lot of megapixels in their sensor, this will draw down the sensor performance but because they are ahead of the technology, it will still be better than other sensor makers.
The best reference is actually Nikon 1, which has the same sized 1" sensor but made by Aptina. However, it is 10mp only. Since RX100 jammed twice more pixels into it, the sensor performance will be around the same or slightly better than Nikon 1 only (by assuming Sony has an edge over Aptina).
http://www.dxomark.com/index.php/Cameras/C...base/Nikon/1-V1
It has a score of 346 for ISO performance.
LX7 vs RX100 ISO sensor performance
So my conclusion is very straightforward, if LX7 has a conservative score of 200 and RX100 has a conservative score of 400, RX100 will have at most 1 stop advantage over LX7 'only'. That means if you shoot f2 at ISO800 of RX100, you can shoot at f1.4 at ISO400 on LX7 using the same shutter speed. That's why the f1.4 bright lens play the biggest magic here.
Dynamic Range
Unlike the old CCD sensors, dynamic range for both CMOS sensor are good enough to stay above 11EV. So I assume they are good enough for a compact of such size.
Added on August 14, 2012, 6:36 pmHow do I gauge RX100 and LX7 lens performance?
Imaging Resource provides a lot of test pics for analysis for both RX100 and LX7 (and also LX5) such as this one. To get them, Google for the specific camera review.
http://www.imaging-resource.com/PRODS/pana...7/LX7hVFAWB.HTM LX7 24mm f1.4
http://www.imaging-resource.com/PRODS/sony...RX100hVFAWB.HTM RX100 28mm f1.8
I am interested in these cameras when they are shot WIDE open and at both WIDE end and TELE end because those are the "soft" spot.
Most lens are sharp at the center but the problem usually arise at the corners so I pay a lot of attention on them.
LX7 vs LX5
Both are Leica lenses (for ego purpose) and to my surprise, the f1.4 LX7 is obviously sharper than LX5 at the corners!!! Taking the same picture with LX7 at f1.4 will look better than LX5 at f2.0!!! That means LX7 is fully usable at f1.4 and not just a hype!!!
LX7 vs RX100
LX7 simply kill the RX100 at the borders and corners. This means while RX100 has a better sensor, it doesn't have a good enough lens to resolved it.
But RX100 is 20 megapixels and LX7 is only 10! If you downsized RX100 to 10mp and compare....
I hear you! Why downsized RX100 to 10mp? Why don't we make it more fun by up sizing 10mp to 20!!!
But corners are not important
You win!
Added on August 14, 2012, 6:52 pmDepth of Field (DOF) Comparison
I choked when both camera markets themselves as good at creating professional SLR quality bokeh...
LX7 has a crop factor of 5.1
RX100 has a crop factor of 2.7
LX7 aperture range f1.4-2.3
RX100 f1.8-4.9
In full frame DOF equivalent, we are like shooting with a 5d mark iii or D700 at the following aperture setting
LX7 f7.1-11.7
RX100 f4.9-13
Both sucks. Even though the f4.9 look "interesting" but it is on the wide end which is 28mm of FF so you don't really get much from it.
If you need bokeh, both cameras are not for you.
Added on August 14, 2012, 7:00 pmWhy LX7 will kill off Canon G11, G12, S95, S100, Nikon P7100 and Olympus XZ1?
G11 and G12 has largest aperture of f2.8. The sensor is slightly worst than LX7 due to CCD characteristic.
LX7 has largest aperture of f1.4 that is fully usable. It has 2 stops advantage on low light performance! When you need to reach ISO800 on G12, you can still use ISO200 on your LX7 and on sensor so small, it is SKY and EARTH on noise level!!!
Aperture means a lot. The S95/S100 has f2.0 but still means 1 stop poorer than LX7. The XZ1 is f1.8. [XZ1 has the worst 1/1.7" sensor made by Panasonic, lol]
This post has been edited by samsungfreak: Aug 15 2012, 04:51 PM
Aug 14 2012, 05:55 PM, updated 12y ago
Quote


0.0241sec
0.97
5 queries
GZIP Disabled