 |
| From mmlacak |
Friday, December 23, 2011
Dreamed away
Have you got that tingling feeling nothing really matters any more? Your dreams are over? And there is nothing you can do about it? Just go away ...
Sunday, September 25, 2011
Nikon 1 system
As you're already aware, a few days ago Nikon introduced Nikon 1 system. And, as always with breakthrough news, web all too readily filled with wild speculations. Will it cannibalise DSLRs directly or just by eating R&D resources away? Will it fail? Will it be winner?
From examples I saw, Nikon 1's images are clean, but only just because Nikon decided to wipe out noise together with a lot of low-contrast details. So, it's not really that good as it appears at first. Having CMOS sensor doesn't help here either, but that was necessitated by the need to support HD videos.
From the technological point of view, Nikon 1 system isn't anything new, it's just rehash of existing bits and pieces, commodity really. Which doesn't bode well for Nikon, to be at whim of any other, more innovative, company.
I don't know for sure what Nikon managers thought and discussed. Based on what they decided, it seems they believe flickr-dwelling happy-snappy tech-unsavvy population of youngsters won't be in short supply any time soon. Which might as well be true.
But only for a short while. You see, in a dwindling economy like this, more and more people probably will think twice, or simply, Google it, before spending not-so-small lump of hard-earned cash. And suddenly they'll find out there already is similar small camera which delivers, Panasonic GF-3 + new 14-42 mm lens.
Trying to protect DSLR sales by designing inferior mirrorless system will just lead to diminishing company as a whole, similar to fate IBM once faced. All one can hope for is that Nikon managers would recall company does not need to commit itself to only one, inferior mirrorless system. And, most likely, I'm not the only one who'd like to see larger, DSLR comparable sensor in mirrorless offer from Nikon.
 |
| Nikon 1 system J1 camera |
From the technological point of view, Nikon 1 system isn't anything new, it's just rehash of existing bits and pieces, commodity really. Which doesn't bode well for Nikon, to be at whim of any other, more innovative, company.
 |
| Panasonic GF-3 + new 14-42 mm lens |
But only for a short while. You see, in a dwindling economy like this, more and more people probably will think twice, or simply, Google it, before spending not-so-small lump of hard-earned cash. And suddenly they'll find out there already is similar small camera which delivers, Panasonic GF-3 + new 14-42 mm lens.
Trying to protect DSLR sales by designing inferior mirrorless system will just lead to diminishing company as a whole, similar to fate IBM once faced. All one can hope for is that Nikon managers would recall company does not need to commit itself to only one, inferior mirrorless system. And, most likely, I'm not the only one who'd like to see larger, DSLR comparable sensor in mirrorless offer from Nikon.
Saturday, September 24, 2011
How to outfox Kolrami
Some time ago I have read How to Defeat Kolrami, which sounded to me like a very reasonable proposal. This is why I removed all ads from my blog recently, even though they were there since beginning. However, today struck me maybe even better proposal.
Why wouldn't I use Google AdSense to earn money from their affiliate program, and then give it away to people in need? I mean, I already have donated to this little girl, but I bet she won't be the last one in need. So, whatever comes to me through AdSense I'll gladly donate to people in need. Of course, I don't have copyright to this idea, if you have blog or something similar, feel free to apply it liberally, wherever it makes sense. And, please, do spread the idea.
Thank you!
Why wouldn't I use Google AdSense to earn money from their affiliate program, and then give it away to people in need? I mean, I already have donated to this little girl, but I bet she won't be the last one in need. So, whatever comes to me through AdSense I'll gladly donate to people in need. Of course, I don't have copyright to this idea, if you have blog or something similar, feel free to apply it liberally, wherever it makes sense. And, please, do spread the idea.
Thank you!
Tuesday, September 20, 2011
Monday, September 19, 2011
Canon Powershots Gs, than and now
In light of Canon's choice of 12 Mpixel BSI CMOS sensor for S100, I planned to write how I just hope Canon won't repeat itself
with disaster known as G9 (search for "Real world examples"), which also had 12 Mpixels, and not very good
ones, at that. Then I got interested in performance of CCD sensors found in G-series, so I compared output from old G7, which had 10 Mpixels, with newest G12,
also with 10 Mpixels CCD sensor of the same 1/1.7" size. Of course, on G12 page you have to manually lower ISO rating back to 800, to compare images at equal levels. While not exactly the same
shot, quite a few objects are in both images, so you can compare
Martini's bottle neck, blue robot, watch face, ...
These are studio shots at full bright artificial Sun, so as I already explained here, these shots are high speed/sports high ISO samples, not night/low light high ISO ones. And before you ask, no, generally, you cannot assume image quality will linearly scale down, when light levels drop. You have to find high ISO images taken at slow enough shutter to have genuine insight just how camera performs when it gets really dark. However, even to untrained eye, it is, I hope, clearly visible just how much cleaner output Canon managed to achieve in latest G12.
I also viewed ISO 1600 images from both G7 and G12. Although not directly comparable because it's not the same scene photographed, these shots are taken in both cases at exposures longer then 1/30 seconds at wide end, enough to represent low light/night high ISO samples. Again, G12 sample looks much better.
I'm not a fan of Canon once again opening a can of Mpixel war worms with its 12 Mpixel S100 sensor, and I do hope it won't be contagious. I mean, if all would settle with 10 Mpixels for advanced compacts, which is plenty, given time we could enjoy compact P&Ss with usable ISO 6400 and beyond, which would be awesome for night street photography.
These are studio shots at full bright artificial Sun, so as I already explained here, these shots are high speed/sports high ISO samples, not night/low light high ISO ones. And before you ask, no, generally, you cannot assume image quality will linearly scale down, when light levels drop. You have to find high ISO images taken at slow enough shutter to have genuine insight just how camera performs when it gets really dark. However, even to untrained eye, it is, I hope, clearly visible just how much cleaner output Canon managed to achieve in latest G12.
I also viewed ISO 1600 images from both G7 and G12. Although not directly comparable because it's not the same scene photographed, these shots are taken in both cases at exposures longer then 1/30 seconds at wide end, enough to represent low light/night high ISO samples. Again, G12 sample looks much better.
I'm not a fan of Canon once again opening a can of Mpixel war worms with its 12 Mpixel S100 sensor, and I do hope it won't be contagious. I mean, if all would settle with 10 Mpixels for advanced compacts, which is plenty, given time we could enjoy compact P&Ss with usable ISO 6400 and beyond, which would be awesome for night street photography.
Sunday, September 18, 2011
Canon PowerShot S100
Now it's becoming really interesting. You see, competition mounts more and more pressure on Canon's top dog among point-and-shooters, that would be G12. Just look at the field, we now have Panasonic LX5, Samsung TL500/EX1, Olympus XZ-1, Nikon P7100 and, best of all, Fuji X10 is about to come really soon now. And yet, Canon manages to undermine most compact among them (Panasonic LX5) with just announced S100.
So S100 is mostly evolutionary stuff, like allegedly improved handling, added GPS, 12 Mpix sensor, ... nothing to write home about, only on blogs like this one ;) . The most significant change is lens, i.e. seemingly the same design and size, it now starts at 24 mm, instead of 28 mm, which is a really big deal for landscapes, while maintaining fast f/2.0 aperture, which is a really big deal for night shooting and such. Also it extends further to 120 mm, although ends at very slow f/5.9 aperture, which isn't that good.
Another thing why S100 retains its appeal is form factor inherited from its predecessors, S95 and S90. Soap box design, small overall dimensions, with just a thin ring around lens protruding from the body by about 2 mm. First bigger competitor, Panasonic LX5, isn't much bigger, but protruding (and much brighter!) lens make it so problematic when one tries to find unobtrusive carrying bag. Which you definitely want, because you already carry mobile phone, hopefully not in your pants.
I just compared small Canon Powershot SD4000 IS (a.k.a. IXUS 300 HS) to previous cameras in S-series (S95 and S90) and G12. SD4000 has 10 Mpixel CMOS BSI sensor, similar to but a little smaller then one in S100. So, pixel density would be about the same, which might mean one could approximate how good S100 might be when it finally arrives. SD4000 did acceptable in comparison, but visibly lagged behind both S90 and S95, at least at sports ISO 1600.
Update! I also had a look at SD4000 samples, for a change, those were proper low-light high ISO images. To put it bluntly, ISO 1600 images and up are appalling, more mobile-phone like, then output from dedicated camera. ISO 800 might be usable, for small images typically found in web galleries. No good. Lets just hope Canon could pull S100 image quality to levels obtainable by S95, though I'm sceptical myself.
So S100 is mostly evolutionary stuff, like allegedly improved handling, added GPS, 12 Mpix sensor, ... nothing to write home about, only on blogs like this one ;) . The most significant change is lens, i.e. seemingly the same design and size, it now starts at 24 mm, instead of 28 mm, which is a really big deal for landscapes, while maintaining fast f/2.0 aperture, which is a really big deal for night shooting and such. Also it extends further to 120 mm, although ends at very slow f/5.9 aperture, which isn't that good.
Another thing why S100 retains its appeal is form factor inherited from its predecessors, S95 and S90. Soap box design, small overall dimensions, with just a thin ring around lens protruding from the body by about 2 mm. First bigger competitor, Panasonic LX5, isn't much bigger, but protruding (and much brighter!) lens make it so problematic when one tries to find unobtrusive carrying bag. Which you definitely want, because you already carry mobile phone, hopefully not in your pants.
I just compared small Canon Powershot SD4000 IS (a.k.a. IXUS 300 HS) to previous cameras in S-series (S95 and S90) and G12. SD4000 has 10 Mpixel CMOS BSI sensor, similar to but a little smaller then one in S100. So, pixel density would be about the same, which might mean one could approximate how good S100 might be when it finally arrives. SD4000 did acceptable in comparison, but visibly lagged behind both S90 and S95, at least at sports ISO 1600.
Update! I also had a look at SD4000 samples, for a change, those were proper low-light high ISO images. To put it bluntly, ISO 1600 images and up are appalling, more mobile-phone like, then output from dedicated camera. ISO 800 might be usable, for small images typically found in web galleries. No good. Lets just hope Canon could pull S100 image quality to levels obtainable by S95, though I'm sceptical myself.
Tuesday, August 30, 2011
High ISO performance
Ok, so I saw this time and time again and, as the saying is going, if you can't change the world, make another blog post.
Sony A77 samples contain a series of images of colorful shirts at various speeds, including ISO 12.800. However, you can't use this series to assess image quality at night. This is because imaging sensors generally produces more noise if used at high speed in pitch-dark night, compared to the same high ISO when used at day or well-lit event. You can use these samples to predict just how Sony A77 will behave when used for sports photography, where high ISO is used just to achieve really short exposures, not because it was very dark.
To assess realistic high ISO performance of camera, first you have to find images made at night using camera you're interested in. Subject, exposure compensation and aperture values are irrelevant. Then make sure exposure length is inverse of focal length in 35 mm equivalent, or longer. If not, you're looking at another sports picture, and again, it'll look better then what you can realistically expect from camera.
In link I posted above, shirts are pictured at ISO 12.800, but look at exposure, it's ludicrous 1/6.400 s, while focal length is 45 mm equivalent, which all means it was done at high noon, and high ISO performance is skewed. Let's try to figure out what would be equivalent night ISO performance. Since it's completely static subject, one could use 1/50 s or longer to make that picture, which yields ISO of just 100. Not very nightish high ISO picture there, isn't it?
If you ever bought camera which was trumpeted for its alleged high ISO performance, while you find it otherwise in actual use, well, now you know why.
Sony A77 samples contain a series of images of colorful shirts at various speeds, including ISO 12.800. However, you can't use this series to assess image quality at night. This is because imaging sensors generally produces more noise if used at high speed in pitch-dark night, compared to the same high ISO when used at day or well-lit event. You can use these samples to predict just how Sony A77 will behave when used for sports photography, where high ISO is used just to achieve really short exposures, not because it was very dark.
To assess realistic high ISO performance of camera, first you have to find images made at night using camera you're interested in. Subject, exposure compensation and aperture values are irrelevant. Then make sure exposure length is inverse of focal length in 35 mm equivalent, or longer. If not, you're looking at another sports picture, and again, it'll look better then what you can realistically expect from camera.
In link I posted above, shirts are pictured at ISO 12.800, but look at exposure, it's ludicrous 1/6.400 s, while focal length is 45 mm equivalent, which all means it was done at high noon, and high ISO performance is skewed. Let's try to figure out what would be equivalent night ISO performance. Since it's completely static subject, one could use 1/50 s or longer to make that picture, which yields ISO of just 100. Not very nightish high ISO picture there, isn't it?
If you ever bought camera which was trumpeted for its alleged high ISO performance, while you find it otherwise in actual use, well, now you know why.
Subscribe to:
Posts (Atom)