Guest wls.shanghai Posted September 11, 2009 Share #1 Posted September 11, 2009 Advertisement (gone after registration) anybody know the dynamic range from the M9 ? tks. wls Link to post Share on other sites More sharing options...
Advertisement Posted September 11, 2009 Posted September 11, 2009 Hi Guest wls.shanghai, Take a look here Dynamic range M9 ???. I'm sure you'll find what you were looking for!
adan Posted September 11, 2009 Share #2 Posted September 11, 2009 You can look at some bright sun/shade pictures here - looks pretty strong to me, at least in .dng http://www.l-camera-forum.com/leica-forum/leica-m9-forum/98128-m9-report-salt-lake-city.html Link to post Share on other sites More sharing options...
Guest wls.shanghai Posted September 12, 2009 Share #3 Posted September 12, 2009 Andy, thank you, I still don't know the dynamic range => f-aperture stops. from the M9 wls Link to post Share on other sites More sharing options...
reven Posted September 12, 2009 Share #4 Posted September 12, 2009 Or just for the fun of it... the M8 had around 11f stops with ISO 160, 10.5 at ISO 360 and 9 at 640 So my guess for the m9 is, 160 being between 11.5 and 12, which puts it around 1 stop behind the best current CMOS sensors, which makes sense. the M9 sensor is just a slightly modified m8 version. ISO 360 at 11 stops and 640 around 9.5. Link to post Share on other sites More sharing options...
Steve Ash Posted September 12, 2009 Share #5 Posted September 12, 2009 Or just for the fun of it... the M8 had around 11f stops with ISO 160, 10.5 at ISO 360 and 9 at 640 So my guess for the m9 is, 160 being between 11.5 and 12, which puts it around 1 stop behind the best current CMOS sensors, which makes sense. the M9 sensor is just a slightly modified m8 version. ISO 360 at 11 stops and 640 around 9.5. I am sorry but I do feel that a "guess" of anybody is of no value. It is only good to start internet rumors. Regards Steve Link to post Share on other sites More sharing options...
scott kirkpatrick Posted September 12, 2009 Share #6 Posted September 12, 2009 The standard way to determine dynamic range is to use manual exposure at the standard ISO and shoot pictures of a textured surface like a paper tower under slightly angled illumination. The difference in stops between the lightest image that still shows some texture and the darkest image that still shows some texture is the usable dynamic range. I don't have an M9 yet, but I think I'll try it when I get mine. scott Link to post Share on other sites More sharing options...
georg Posted September 12, 2009 Share #7 Posted September 12, 2009 Advertisement (gone after registration) As far as I know, there is no standard method to measure DR. There are also certain tricks possible to "pretend" additional dynamic range - you need spatial content (details). ARRI developed a new DR-measurent-system: Sensitivity | ARRI Digital Maybe this becomes the standard in the future? But "measurements" like dxomark who claim up to 2 stops difference with the same sensor!? I don't know if we should rely on that - we should trust our eyes. CCD-based cameras with high-full rate and 14/16bit-files like the P40+/P65+ offer the highest DR I've ever seen in files - additional DR is very difficult to achieve because you operate agains physicial limitations. Link to post Share on other sites More sharing options...
reven Posted September 12, 2009 Share #8 Posted September 12, 2009 Well if you take the specs, images posted, than you can guess pretty well. The sensor is nothing new, so all the real changes could only come from AD conversion. That means there won't be huge DR changes from the m8 just small upgrades. Link to post Share on other sites More sharing options...
Jamie Roberts Posted September 13, 2009 Share #9 Posted September 13, 2009 Well if you take the specs, images posted, than you can guess pretty well. The sensor is nothing new, so all the real changes could only come from AD conversion. That means there won't be huge DR changes from the m8 just small upgrades. Um, what makes you think the sensor is nothing new? LOL!! Just because the pixel pitch is the same doesn't mean it's not a new design. The color filters are new. The microlenses are new. The overall character of the sensor may be similar to other Kodak pro sensors, including the M8 and DMR, but I'm betting there have been a few advances in the last 3 or 4 years. So how do you know the sensor isn't a new design (within the boundaries of a bayer-pattern CCD, of course)? Link to post Share on other sites More sharing options...
reven Posted September 13, 2009 Share #10 Posted September 13, 2009 As far as I can see from the specs it is a further developed M8 sensor. Nothing new at all. They just managed to put a stronger IR Filter over it and optimized it. First of all if it would be a new design, Leica would have made that point a lot more clear. The Second point is that I'm quite sure that Kodak did not design a new sensor with the old specs... That makes no sense whatsoever if it would be a brand new sensor it would be 6.0 and more like the S2 sensor. I'm not saying it is a bad thing, I'm just pretty sure that the M9 sensor is NO new design. Link to post Share on other sites More sharing options...
scott kirkpatrick Posted September 13, 2009 Share #11 Posted September 13, 2009 Kodak published a full technical spec, describing the pixel, the overlayers and the whole chip circuitry, for the KAF-10500 (the M8's CCD) and for the chip in the DMR. They said in their press release that the chips in the S2 and the M8 are the KAF-37500 and the KAF-18500, but their website doesn't give technical specs for either chip. The press release does say that the pixel and the Bayer Filter were redesigned. The promotional literature talks about microlens placement, but the number that most directly affects dynamic range is the number of electrons that each pixel can capture and hold before it overflows. In the KAF-10500, that is 60,000 or just under 2^16.. But this is not all signal, since there are leakage, shot noise and side effects of the readout process to be dealt with. Read error is characterized as 16 electrons. Maybe newer designs have deeper wells, holding more electrons when full, or have reduced the sources of noise. If you take the ideal dynamic range to be the ratio in stops (powers of 2) of the smallest and the largest believable signal, that is 12 stops, but keeping more bits of signal and noise may allow use of averaging techniques to reduce the noise. *Usable * dynamic range will depend on the rest of the system, lens, raw file development, printing technique, which is why I mentioned the traditional paper towel experiment. scott. Link to post Share on other sites More sharing options...
BerndReini Posted September 13, 2009 Share #12 Posted September 13, 2009 which puts it around 1 stop behind the best current CMOS sensors, which makes sense. the M9 sensor is just a slightly modified m8 version. Where do you get that CMOS sensors have better dynamic range than CCDs. My firend openly admits that as far as dynamic range is concerned, my M8 blows his 5DII out of the water. I'd like to see that one stop of extra dynamic range from a CMOS sensor, in your dreams. Link to post Share on other sites More sharing options...
t024484 Posted September 13, 2009 Share #13 Posted September 13, 2009 Well if you take the specs, images posted, than you can guess pretty well. The sensor is nothing new, so all the real changes could only come from AD conversion. That means there won't be huge DR changes from the m8 just small upgrades. The noise from A/D conversion of the M8 is top notch, better than any Canono or Nikon DSLR, so there is not much to be expected from this device. The point is that at a higher ISO values, the A/D noise is getting less and less important and roughly as from ISO 640 the Read Noise from the sensor is the dominant noise. Since the Read Noise in the M8 is quite high, higher than the specs of the KAF10500 make expect, this is were improvements can be found by having a less noise susceptible design. Looking at the Nikon etc. the read noise could be lowered by a factor 2 to 3, giving a real bonus at higher ISO values, but hardly any advantage at ISO 160. So where the highest DR is to found, at ISO 160, the DR of the M8 and the M9 will be comparable. It could be that the M9 is using a different black current system that takes away some from the DR at the bottom end of the range from 0 to 16.383, by not starting at 0 but at a higher value like 500, but if so there is still enough left to cover a DR of 12 stops. As a matter of fact, the Shot Noise being the major noise source at higher ISO values is caused by physical reasons and cannot be influenced. If the M9 has a better and lower noise susceptible design to bring the Sensor signal to the A/D converter, than the physical limits will not allow any further noise reduction, other than by doing this in the digital domain which will most likely reduce resolution. Hans Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.