Re: On Sharpness (Confusion)
Tom C wrote: Reread again... this is clearly not a valid test... you messed with the image by resizing it, you messed with ithe mage by by saving it as a .jpg, you messed with image by resizing again, you NEVER had the original image out of the camera to start with, and you don't know how the photo was manipulated prior to it being on the Pentax website... Because it was supposedly taken with that camera. Of course, you can do the same by yourself with any other good picture taken with the *ist D and see what happens. Makes me wonder about the validity of the 70-210 tests. As I already wrote, I used two different lenses prior to publish the test. As I already wrote, I'm going to repeat that test (with maximum care) as soon as I'll get the Sigma 2.8/70-200 for comparison. In case I was wrong, I'll have no problem in admitting that, as I already did on some occasions in the past. I know very veeery few persons doing that. Dario Bonazza
Re: On Sharpness (Confusion)
Tom C wrote: Sorry, you're just not making sense to me. First you refer to and complain explicitly about the *istD compared to other cameras and now you're saying you're referring to the image and the raw converter. Because it was supposedly taken with that camera. Of course, you can do the same with any other good picture taken with the *ist D. And unless I'm missing something here, you're doing it with an image that you retrieved from the web, not took yourself, Because it was supposedly taken with that camera. Of course, you can do the same with any other good picture taken with the *ist D. and have manipulated after that to prove some point about pixels worth of data... am I wrong? If so, please let me know. I just resized it down and then resized it up, to see a possible data loss (which didn't happen, and that's the interesting point). No other manipulation. I'm not interested in making tricks. I'm only interested in understanding something more. Of course, I can do something wrong and I always welcome useful suggestions. Dario Bonazza
Re: On Sharpness (Confusion)
Yes, in absolute terms, it is not a valid experiment. My assumption was that two different medium speed normal lenses are probably the closest thing to using the same lens on both bodies. As it is not possible to do it the right way, what would be the best other way to do it? Andre It's like trying to compare two audio amplifiers, each with a different set of speakers. Since, IMO, all other things equal, image 'quality' is largely a factor of the lens, using different lenses only says something about the lens or the lens/camera system, not the camera. Tom C. If you change more than one parameter, you invalidate the test. William Robb From: Andre Langevin You'd need to compare the exact same scene, same time, same everything including lens, which is impossible... Tom C. Isn't it quite possible if you you compare raw data from the cameras at test, previously fit with (almost identical) medium speed normal lenses?
Re: On Sharpness (Confusion)
On 8 Jun 2004 at 11:30, Boris Liberman wrote: Hi! It would appear from the most recent talk that Sharpness is more or less __the__ most important characteristic of the lens. However, from previous talks it has become my understanding that Pentax do not optimize their lenses specially for sharpness (except may be macro lenses and such). Pentax, AFAIU, optimize their lenses for plasticity, or overall smoothness of the picture. OTOH, Nikon are known for having their lenses tack sharp all over the frame. At least this is what I have accumulated in my small knowledge bag so far. Hi Boris, I'm not sure if I'm about to confuse you or not but I suspect that these theories that you mention are really only relevant to older prime lenses and film use. Virtually all late primes are sharp enough as to be sharper than the sensors can resolve at all but their extreme aperture and some are even good at the extremes. The biggest problems regarding lens compatibility on DSLRs is mainly chromatic aberration. Zooms are another story, many particularly the older ones are quite poor. The links that Dario recently put up show these differences very graphically. Please unconfuse me - why all this talk about sharpness? I do realize that for digital lenses should be very sharp. Or at least, it would be a reasonable thing - to want one's lenses to be sharp. But sharpness is not all, right? Part of the recent talk of sharpness was referring to in-camera sharpening, this is an independent issue to lens sharpness. The talk of the F series zoom sharpness was another story. I had the SMC F 70-210 f4.0-5.6 and I'm sorry to report that it was quite poor in all respects relative to my primes. Cheers, Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Re: On Sharpness (Confusion)
Boris, some remarks to add to your confusion ;-) o It seems that lenses from Canon/Nikon and the likes are often closer to what test procedures ask from a lens (contrast/sharpness), so Pentax lenses often get less good results - at least it looks like that in my part of the world and yes, it could well be that this is only so because Pentax advertises less in the magazines that publish the tests. We don't know what Pentax is thinking. If they really have a different take on what a good lens is, a more holistic approach instead of a 'scientific' one, based on measured sharpness and contrast, then they are certainly not good at 'selling' it. o We cannot be sceptical enough regarding our own perceptions and test procedures. I believe it is plain impossible to say anything about a lens based on a set of 4x6 prints. Even if you look at slides you need to always be aware that the loupe or the projector lens also is part of what you see. If you try to compare pictures taken of identical subjects you still have a whole bunch of issues that may make the results worthless. You just don't know whether somebody claiming a lens to be 'less sharp' in reality is talking about his shaky tripod... A person may also just not telling the truth - half of what is written on the internet is wrong. o We don't know if and how much sample variation there is. I personally have not been able to show much of this when comparing lenses of the same series but a) I do not have a testing laboratory and b) it is certainly possible the the particular lens that the owner claims to be less good is a lens that deviates fom the rest. I personally have concluded to stop 'testing' more or less. I have my personal lens favourites (my M 4/20, my FA 24-90, my M 2/35 my 3,5/35-105 stopped down) and I would go as far as recommending to give any of these a try, but no further. Sven Zitat von Boris Liberman [EMAIL PROTECTED]: Hi! It would appear from the most recent talk that Sharpness is more or less __the__ most important characteristic of the lens. However, from previous talks it has become my understanding that Pentax do not optimize their lenses specially for sharpness (except may be macro lenses and such). Pentax, AFAIU, optimize their lenses for plasticity, or overall smoothness of the picture. OTOH, Nikon are known for having their lenses tack sharp all over the frame. At least this is what I have accumulated in my small knowledge bag so far. Please unconfuse me - why all this talk about sharpness? I do realize that for digital lenses should be very sharp. Or at least, it would be a reasonable thing - to want one's lenses to be sharp. But sharpness is not all, right? Most recently I've sold my Soligor C/D 70-222/3.5 lens in favor of my Pentax SMC F 70-210/4.0-5.6. It could very well be that Soligor is sharper on some apertures. It is faster too. However it has rather ugly bokeh - at least to my taste, and general zoomish taste on the pictures I've been able to take with it. Pentax lens however is very smooth, very 3D, very pleasant. It is also lighter, has AF and is of course fully compatible with my MZ-6's electronics. It also allows for use of built-in flash at least for some of the focal distances. But this is already tech-talk. Still, why to be so aware of sharpness? I mean, if one wants sharpness, perhaps one should look for 3rd party lenses specially optimized for sharpness. Please unconfuse me. Thanks. Boris
Re: On Sharpness (Confusion)
We don't know what Pentax is thinking. If they really have a different take on what a good lens is, a more holistic approach instead of a 'scientific' one, based on measured sharpness and contrast, then they are certainly not good at 'selling' it. I am certain many Pentax users would be interesed to know what their engineers have been thinking as well. There was a pdf file which was talking about the FA43 FA77, but unfortunately, it was available in Japanese only. Is the international market so worthless to them they won't even translate something that might sell their products? A person may also just not telling the truth - half of what is written on the internet is wrong. And most people have the tendency to praise whatever they had. o We don't know if and how much sample variation there is. I personally have not been able to show much of this when comparing lenses of the same series but a) I do not have a testing laboratory and b) it is certainly possible the the particular lens that the owner claims to be less good is a lens that deviates fom the rest. I don't know much about optical sample variation, but their QC is bad, as far as I can tell. And I am talking about * Limited lenses too, and that frustrates me a lot. Regards, Alan Chan http://www.pbase.com/wlachan _ Add photos to your messages with MSN Premium. Get 2 months FREE* http://join.msn.com/?pgmarket=en-capage=byoa/premxAPID=1994DI=1034SU=http://hotmail.com/encaHL=Market_MSNIS_Taglines
Re: On Sharpness (Confusion)
on 08.06.04 10:49, Dario Bonazza at [EMAIL PROTECTED] wrote: AFAIK, the *ist D allows excellent blow-ups up to 20x30cm, and just acceptable ones up to 30x40cm. Not so bad, you could say, but can anybody explain me why the hell any good 4-5MP digital PS (Canon, Konica, Leica, Minolta, Nikon, Olympus, Panasonic, Pentax...) can go beyond that? I've had several proofs of that, since a friend of mine (owner of a pre-press service company and very skilled Photoshop user) shows me such large format prints again and again. The only 5-6 MP camera he was unable to bring beyond 30x40cm is (guess which?)... the *ist D! Unfortunately you are probably right Dario :-( While *istD is great at closer distances it looses many details at longer ones. I compared my photos to my friend's, made with C 10D (people photos, with flash) and the difference was clearly visible on hair - if it was taken from greater distance *istD loosed details and it looked slightly unnatural - it remained a kind of picture I am used to in video cameras... 10D pictures looked much better in this condition. And yes, my friend's Sony F707 could resolve more detail in people's hair and skin when in certain distance. I don't know what is responsible for this? I quess it is faulty bayer interpolation software, because it seems that Nikon D100 using the same CCD doesn't have this problems from what I've seen so far. That said, 20x30 cm prints from *istD are great, 30x40 cm made on inkjet (HP Designjet 10ps) are very nice too. -- Best Regards Sylwek
Re: On Sharpness (Confusion)
Sylwester Pietrzyk wrote: on 08.06.04 10:49, Dario Bonazza at [EMAIL PROTECTED] wrote: AFAIK, the *ist D allows excellent blow-ups up to 20x30cm, and just acceptable ones up to 30x40cm. Not so bad, you could say, but can anybody explain me why the hell any good 4-5MP digital PS (Canon, Konica, Leica, Minolta, Nikon, Olympus, Panasonic, Pentax...) can go beyond that? I've had several proofs of that, since a friend of mine (owner of a pre-press service company and very skilled Photoshop user) shows me such large format prints again and again. The only 5-6 MP camera he was unable to bring beyond 30x40cm is (guess which?)... the *ist D! Unfortunately you are probably right Dario :-( While *istD is great at closer distances it looses many details at longer ones. I compared my photos to my friend's, made with C 10D (people photos, with flash) and the difference was clearly visible on hair - if it was taken from greater distance *istD loosed details and it looked slightly unnatural - it remained a kind of picture I am used to in video cameras... 10D pictures looked much better in this condition. And yes, my friend's Sony F707 could resolve more detail in people's hair and skin when in certain distance. I don't know what is responsible for this? I quess it is faulty bayer interpolation software, because it seems that Nikon D100 using the same CCD doesn't have this problems from what I've seen so far. That said, 20x30 cm prints from *istD are great, 30x40 cm made on inkjet (HP Designjet 10ps) are very nice too. -- Best Regards Sylwek So maybe I'm not as hallucinated as other messages should suggest. Thanks Sylwek for reassuring me a bit. OK, let's try to get the most out of our excellent hardware with 4MP equivalent performance. Dario Bonazza
Re: On Sharpness (Confusion)
On Tue, 8 Jun 2004, Dario Bonazza wrote: So maybe I'm not as hallucinated as other messages should suggest. Hi Dario, Your posts always make sense and your photographs even more so. I fear you read too much into my looks like it's out-of-focus comment; I never thought you would be furnishing us with such a picture. I was criticising the lens performance. Keep writing and keep showing us your pictures. Kostas
Re: On Sharpness (Confusion)
on 08.06.04 15:15, Rob Studdert at [EMAIL PROTECTED] wrote: Well of course Bayer interpolation algorithms are the difference if the sensor is the precisely the same type. I just don't understand why there seems to be so much debate about something that isn't a problem. I'm happy with the sharpness of the *ist D, but I'm sure with the appropriate sharpening factors I could make its image like the D100. See the following set of samples and see which one appears least distorted: http://www.pbase.com/image/27208228/original Well, it seems that least distorted are samples from 10D and D1X, with *istD next. But what would happen if you cramp up sharpness and contrast of *istD image to the levels similar to let's say 10D??? -- Best Regards Sylwek
Re: On Sharpness (Confusion)
On 8 Jun 2004 at 15:33, Sylwester Pietrzyk wrote: Well, it seems that least distorted are samples from 10D and D1X, with *istD next. But what would happen if you cramp up sharpness and contrast of *istD image to the levels similar to let's say 10D??? The D1X is pretty good but it still suffers from some pretty hideous aliasing errors mostly due to over sharpening I assume. I think that most of the cameras but the *ist D are being pushed past their usable limits in order to achieve good results in largely irrelevant tests. All I can say is that when used well the *ist D can produce printed photographic results that don't have that identifiable digital look. All that I can put this down to is that the designers wanted the images it produced to look like photos and as such didn't try to extract information from the sensor that was beyond its capabilities and I'm pretty glad about that. Cheers, Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Re: On Sharpness (Confusion)
Rob Studdert wrote: Well of course Bayer interpolation algorithms are the difference if the sensor is the precisely the same type. I just don't understand why there seems to be so much debate about something that isn't a problem. I'm happy with the sharpness of the *ist D, but I'm sure with the appropriate sharpening factors I could make its image like the D100. See the following set of samples and see which one appears least distorted: http://www.pbase.com/image/27208228/original Those targets support well my idea that the problem with the *ist D can be the low-pass filter in front of the sensor, causing both lower moire (color artifacts) and lower resolution. When you shoot natural subjects (not optical targets), the moire is seldom a problem, while lower resolution is always visible. I believe you answered the question about the *ist D performance. Dario Bonazza
Re: On Sharpness (Confusion)
Hi, to my eye the *ist-D is the one with the least distortion. Only a slight color moire far down. Am I wrong? Or is even that subjective? Regards, Hans. --- Sylwester Pietrzyk [EMAIL PROTECTED] wrote: on 08.06.04 15:15, Rob Studdert at [EMAIL PROTECTED] wrote: Well of course Bayer interpolation algorithms are the difference if the sensor is the precisely the same type. I just don't understand why there seems to be so much debate about something that isn't a problem. I'm happy with the sharpness of the *ist D, but I'm sure with the appropriate sharpening factors I could make its image like the D100. See the following set of samples and see which one appears least distorted: http://www.pbase.com/image/27208228/original Well, it seems that least distorted are samples from 10D and D1X, with *istD next. But what would happen if you cramp up sharpness and contrast of *istD image to the levels similar to let's say 10D??? -- Best Regards Sylwek _ 23a mail
Re: On Sharpness (Confusion)
On 8 Jun 2004 at 7:13, Hans Imglueck wrote: Hi, to my eye the *ist-D is the one with the least distortion. Only a slight color moire far down. This is my conclusion also, it is effectively the closest to being naturally band-width limited, ie it's just like film. Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Re: On Sharpness (Confusion)
Hans Imglueck wrote: to my eye the *ist-D is the one with the least distortion. Only a slight color moire far down. Am I wrong? Or is even that subjective? And the *ist D is the one where the BW lines get already confused when other competitors can still resolve them, hence same comment as to Rob's message: Those targets support well my idea that the problem with the *ist D can be the low-pass filter in front of the sensor, causing both lower moire (color artifacts) and lower resolution. When shooting natural subjects (not optical targets), the moire is seldom a problem, while lower resolution is always visible. Dario Bonazza
Re: On Sharpness (Confusion)
Hi, from that test I cannot distinguish which one has the better resolution. To my opinion all of them are losing real resolution at the middle of the graphs. The bottom part is faked on the most ones. Real resolution would mean: Seeing 9 lines. Regards, Hans. --- Dario Bonazza [EMAIL PROTECTED] wrote: Rob Studdert wrote: Well of course Bayer interpolation algorithms are the difference if the sensor is the precisely the same type. I just don't understand why there seems to be so much debate about something that isn't a problem. I'm happy with the sharpness of the *ist D, but I'm sure with the appropriate sharpening factors I could make its image like the D100. See the following set of samples and see which one appears least distorted: http://www.pbase.com/image/27208228/original Those targets support well my idea that the problem with the *ist D can be the low-pass filter in front of the sensor, causing both lower moire (color artifacts) and lower resolution. When you shoot natural subjects (not optical targets), the moire is seldom a problem, while lower resolution is always visible. I believe you answered the question about the *ist D performance. Dario Bonazza _ 23a mail
Re: On Sharpness (Confusion)
Only film is just like film. The primary difference between film and digital images is that whereas digital pixels are, of necessity, regularly ordered, each is of identical effective size and their placement is uniform. Grain on film is randomly organized in position, size and density. For me, film has an ineffable quality that I will always appreciate. Regards, Bob... --- No man's life, liberty or property is safe while the legislature is in session. -- Mark Twain From: Rob Studdert [EMAIL PROTECTED] On 8 Jun 2004 at 7:13, Hans Imglueck wrote: Hi, to my eye the *ist-D is the one with the least distortion. Only a slight color moire far down. This is my conclusion also, it is effectively the closest to being naturally band-width limited, ie it's just like film.
Re: On Sharpness (Confusion)
You can't screw with the picture in Photoshop and call it a test of the camera! Tom C. From: Dario Bonazza [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: On Sharpness (Confusion) Date: Tue, 8 Jun 2004 17:39:40 +0200 Kostas Kavoussanakis wrote: Hi Dario, Your posts always make sense and your photographs even more so. I fear you read too much into my looks like it's out-of-focus comment; I never thought you would be furnishing us with such a picture. I was criticising the lens performance. Hi Kostas, No, no. My comments are not based on your comment on the F 70-210 lens. They come from 6-months experience with the *ist D and being disappointed in seeing 4-5MP digital compacts allowing larger prints. *ist D folks, do you want me to shock you all? I resized that DA 14mm picture down to 4MP and saved it as best jpeg (Photoshop quality=12), then I closed the file for being sure not to retain 6MP info in Photoshop memory. Then I opened the 4MP file again and I resized it up to 6MP. You can find the result here: http://www.dariobonazza.com/tests/istD_4to6MP.jpg Compare any detail between the original picture and this one (both images side by side on your monitor at 400% or so). You'll see that there are few if any details lost, while the bad outline interpolation of the Pentax RAW converter has been fixed, so that the image looks more natural (slanted and curved lines are smoother, showing less pixelation). Conclusion? The original *ist D picture featured more or less 4MP information in it. For that reason, I consider the *ist D to be a 4MP equivalent camera, at least in RAW/Pentax Lab converter workflow. Hopefully, a decent RAW converter could do something better. Keep writing and keep showing us your pictures. Of course! Dario Bonazza
Re: On Sharpness (Confusion)
You can't take two pictures with different cameras, different lenses, maybe different apertures and therefore DOF, and who knows what else is different post-camera, compare them, and say anything with certainty about the camera itself! Tom C. From: Sylwester Pietrzyk [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: [EMAIL PROTECTED] [EMAIL PROTECTED] Subject: Re: On Sharpness (Confusion) Date: Tue, 08 Jun 2004 13:23:38 +0200 on 08.06.04 10:49, Dario Bonazza at [EMAIL PROTECTED] wrote: AFAIK, the *ist D allows excellent blow-ups up to 20x30cm, and just acceptable ones up to 30x40cm. Not so bad, you could say, but can anybody explain me why the hell any good 4-5MP digital PS (Canon, Konica, Leica, Minolta, Nikon, Olympus, Panasonic, Pentax...) can go beyond that? I've had several proofs of that, since a friend of mine (owner of a pre-press service company and very skilled Photoshop user) shows me such large format prints again and again. The only 5-6 MP camera he was unable to bring beyond 30x40cm is (guess which?)... the *ist D! Unfortunately you are probably right Dario :-( While *istD is great at closer distances it looses many details at longer ones. I compared my photos to my friend's, made with C 10D (people photos, with flash) and the difference was clearly visible on hair - if it was taken from greater distance *istD loosed details and it looked slightly unnatural - it remained a kind of picture I am used to in video cameras... 10D pictures looked much better in this condition. And yes, my friend's Sony F707 could resolve more detail in people's hair and skin when in certain distance. I don't know what is responsible for this? I quess it is faulty bayer interpolation software, because it seems that Nikon D100 using the same CCD doesn't have this problems from what I've seen so far. That said, 20x30 cm prints from *istD are great, 30x40 cm made on inkjet (HP Designjet 10ps) are very nice too. -- Best Regards Sylwek
Re: On Sharpness (Confusion)
I don't call it a test of the camera. I call it a visible demonstration of the information stored in original image file and how poor and unnatural the outlines are rendered by the Pentax RAW conversion software (this was already known). Should you want to keep your eyes closed, you're free to do that, of course. Somebody else could be interested in knowing and understanding something better. No offense intended, just plain straightforward words. Dario Bonazza - Original Message - From: Tom C [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Tuesday, June 08, 2004 8:55 PM Subject: Re: On Sharpness (Confusion) You can't screw with the picture in Photoshop and call it a test of the camera! Tom C. From: Dario Bonazza [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: On Sharpness (Confusion) Date: Tue, 8 Jun 2004 17:39:40 +0200 Kostas Kavoussanakis wrote: Hi Dario, Your posts always make sense and your photographs even more so. I fear you read too much into my looks like it's out-of-focus comment; I never thought you would be furnishing us with such a picture. I was criticising the lens performance. Hi Kostas, No, no. My comments are not based on your comment on the F 70-210 lens. They come from 6-months experience with the *ist D and being disappointed in seeing 4-5MP digital compacts allowing larger prints. *ist D folks, do you want me to shock you all? I resized that DA 14mm picture down to 4MP and saved it as best jpeg (Photoshop quality=12), then I closed the file for being sure not to retain 6MP info in Photoshop memory. Then I opened the 4MP file again and I resized it up to 6MP. You can find the result here: http://www.dariobonazza.com/tests/istD_4to6MP.jpg Compare any detail between the original picture and this one (both images side by side on your monitor at 400% or so). You'll see that there are few if any details lost, while the bad outline interpolation of the Pentax RAW converter has been fixed, so that the image looks more natural (slanted and curved lines are smoother, showing less pixelation). Conclusion? The original *ist D picture featured more or less 4MP information in it. For that reason, I consider the *ist D to be a 4MP equivalent camera, at least in RAW/Pentax Lab converter workflow. Hopefully, a decent RAW converter could do something better. Keep writing and keep showing us your pictures. Of course! Dario Bonazza
Re: On Sharpness (Confusion)
Sorry, you're just not making sense to me. First you refer to and complain explicitly about the *istD compared to other cameras and now you're saying you're referring to the image and the raw converter. And unless I'm missing something here, you're doing it with an image that you retrieved from the web, not took yourself, and have manipulated after that to prove some point about pixels worth of data... am I wrong? If so, please let me know. Tom C. From: Dario Bonazza [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: On Sharpness (Confusion) Date: Tue, 8 Jun 2004 23:23:39 +0200 I don't call it a test of the camera. I call it a visible demonstration of the information stored in original image file and how poor and unnatural the outlines are rendered by the Pentax RAW conversion software (this was already known). Should you want to keep your eyes closed, you're free to do that, of course. Somebody else could be interested in knowing and understanding something better. No offense intended, just plain straightforward words. Dario Bonazza - Original Message - From: Tom C [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Tuesday, June 08, 2004 8:55 PM Subject: Re: On Sharpness (Confusion) You can't screw with the picture in Photoshop and call it a test of the camera! Tom C. From: Dario Bonazza [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: On Sharpness (Confusion) Date: Tue, 8 Jun 2004 17:39:40 +0200 Kostas Kavoussanakis wrote: Hi Dario, Your posts always make sense and your photographs even more so. I fear you read too much into my looks like it's out-of-focus comment; I never thought you would be furnishing us with such a picture. I was criticising the lens performance. Hi Kostas, No, no. My comments are not based on your comment on the F 70-210 lens. They come from 6-months experience with the *ist D and being disappointed in seeing 4-5MP digital compacts allowing larger prints. *ist D folks, do you want me to shock you all? I resized that DA 14mm picture down to 4MP and saved it as best jpeg (Photoshop quality=12), then I closed the file for being sure not to retain 6MP info in Photoshop memory. Then I opened the 4MP file again and I resized it up to 6MP. You can find the result here: http://www.dariobonazza.com/tests/istD_4to6MP.jpg Compare any detail between the original picture and this one (both images side by side on your monitor at 400% or so). You'll see that there are few if any details lost, while the bad outline interpolation of the Pentax RAW converter has been fixed, so that the image looks more natural (slanted and curved lines are smoother, showing less pixelation). Conclusion? The original *ist D picture featured more or less 4MP information in it. For that reason, I consider the *ist D to be a 4MP equivalent camera, at least in RAW/Pentax Lab converter workflow. Hopefully, a decent RAW converter could do something better. Keep writing and keep showing us your pictures. Of course! Dario Bonazza
Re: On Sharpness (Confusion)
You'd need to compare the exact same scene, same time, same everything including lens, which is impossible... Tom C. Isn't it quite possible if you you compare raw data from the cameras at test, previously fit with (almost identical) medium speed normal lenses? Andre
Re: On Sharpness (Confusion) - Bill?
- Original Message - From: Tom C Subject: Re: On Sharpness (Confusion) - Bill? I wish William Robb was a round to give us a refresher on sharpness vs. resolution. GAAAK. William Robb
Re: On Sharpness (Confusion)
- Original Message - From: Andre Langevin Subject: Re: On Sharpness (Confusion) You'd need to compare the exact same scene, same time, same everything including lens, which is impossible... Tom C. Isn't it quite possible if you you compare raw data from the cameras at test, previously fit with (almost identical) medium speed normal lenses? No. If you change more than one parameter, you invalidate the test. You want to compare naked resolution of the sensors on various camera makes and models, then you need to start with a lens that will resolve more than the sensor will, and that same lens must be used on every camera. Otherwise, you are testing lenses as much as sensors. William Robb
Re: On Sharpness (Confusion)
Well, not to me. It's like trying to compare two audio amplifiers, each with a different set of speakers. Since, IMO, all other things equal, image 'quality' is largely a factor of the lens, using different lenses only says something about the lens or the lens/camera system, not the camera. I understand your point, it's just that we're talking about minisucle details and about something that is largely subjective to begin with. Tom C. From: Andre Langevin [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: On Sharpness (Confusion) Date: Tue, 8 Jun 2004 16:19:03 -0400 You'd need to compare the exact same scene, same time, same everything including lens, which is impossible... Tom C. Isn't it quite possible if you you compare raw data from the cameras at test, previously fit with (almost identical) medium speed normal lenses? Andre
Re: On Sharpness (Confusion) - Bill?
I suspected as much. :) Tom C. GAAAK. William Robb
Re: On Sharpness (Confusion)
I think you mean very film like which has been mentioned by the Japanese as well. Regards, Alan Chan http://www.pbase.com/wlachan All I can say is that when used well the *ist D can produce printed photographic results that don't have that identifiable digital look. All that I can put this down to is that the designers wanted the images it produced to look like photos and as such didn't try to extract information from the sensor that was beyond its capabilities and I'm pretty glad about that. _ MSN Premium with Virus Guard and Firewall* from McAfee® Security : 2 months FREE* http://join.msn.com/?pgmarket=en-capage=byoa/premxAPID=1994DI=1034SU=http://hotmail.com/encaHL=Market_MSNIS_Taglines
Re: On Sharpness (Confusion)
On 8 Jun 2004 at 17:39, Dario Bonazza wrote: No, no. My comments are not based on your comment on the F 70-210 lens. They come from 6-months experience with the *ist D and being disappointed in seeing 4-5MP digital compacts allowing larger prints. *ist D folks, do you want me to shock you all? I resized that DA 14mm picture down to 4MP and saved it as best jpeg (Photoshop quality=12), then I closed the file for being sure not to retain 6MP info in Photoshop memory. Then I opened the 4MP file again and I resized it up to 6MP. You can find the result here: Hi Dario, I'm not sure what you are trying to prove given your method and selection of test image and I'm not so sure why you are so surprised as to the results. Firstly in order to reduce a 6MP+ image to the pixel dimensions of a 4MP image the linear dimensions need to be reduced to about 83% so that equates to discarding less than one in five pixels in each dimension. This means that if the image is not very sharp and has no sharpening artifacts you're likely not see a difference in the before and after images. Have you done similar test with high quality 4MP images? If the results of such an experiment are similar to the previous one you made with the 6MP images where does that leave us :-) So in order to alleviate further misdirected energies towards non-existent sharpness problem of the *ist D I've made my own test which I believe will prove its self. I used a source image shot by myself using the 31mm LTD (the non-green image that I posted earlier) which was saved as best quality JPG in camera with minimum in-camera sharpening and contrast and average saturation. http://members.ozemail.com.au/~audiob/temp/EXIF.gif I simply reduced it's size to 83% then enlarged it again to it's original dimensions within PS (using best quality bicubic interpolation). I then saved the result as a loss-less file. I compared these two images side by side in my image browser (Thumbs Plus) at 3x magnification and made the following screen shot: http://members.ozemail.com.au/~audiob/temp/test.jpg Now have a guess at which side was the original? I'll give you a hint, it's the one that's lost resolution :-) I hope my point is made, if not then we obviously have a very different understandings and expectations regarding digital imaging. Cheers, Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Re: On Sharpness (Confusion)
On 9 Jun 2004 at 8:59, Rob Studdert wrote: Oh I wish I had an editor :-( This means that if the image is not very sharp and has no sharpening artifacts you're likely not see a difference in the before and after images. Should read: if the image is not very sharp and has sharpening artifacts Now have a guess at which side was the original? I'll give you a hint, it's the one that's lost resolution :-) Should read: Now have a guess at which side was the original? I'll give you a hint, it's the one that hasn't lost resolution Sorry for the confusion. Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Re: On Sharpness (Confusion)
We knew what you meant... Tom C. From: Rob Studdert [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: On Sharpness (Confusion) Date: Wed, 09 Jun 2004 09:13:44 +1000 On 9 Jun 2004 at 8:59, Rob Studdert wrote: Oh I wish I had an editor :-( This means that if the image is not very sharp and has no sharpening artifacts you're likely not see a difference in the before and after images. Should read: if the image is not very sharp and has sharpening artifacts Now have a guess at which side was the original? I'll give you a hint, it's the one that's lost resolution :-) Should read: Now have a guess at which side was the original? I'll give you a hint, it's the one that hasn't lost resolution Sorry for the confusion. Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Re: On Sharpness (Confusion)
Well of course Bayer interpolation algorithms are the difference if the sensor is the precisely the same type. I just don't understand why there seems to be so much debate about something that isn't a problem. I'm happy with the sharpness of the *ist D, but I'm sure with the appropriate sharpening factors I could make its image like the D100. See the following set of samples and see which one appears least distorted: http://www.pbase.com/image/27208228/original Ack, horrible conversion :) Someone's been using either dcraw, or the photoshop raw convertor :) Love, Light and Peace, - Peter Loveday Director of Development, eyeon Software
Re: On Sharpness (Confusion)
No. If you change more than one parameter, you invalidate the test. You want to compare naked resolution of the sensors on various camera makes and models, then you need to start with a lens that will resolve more than the sensor will, and that same lens must be used on every camera. Otherwise, you are testing lenses as much as sensors. Hmm, its possible to mount a screw mount on most (all?) major brand DSLRs, isn't it? Love, Light and Peace, - Peter Loveday Director of Development, eyeon Software
Re: On Sharpness (Confusion)
- Original Message - From: Peter Loveday Subject: Re: On Sharpness (Confusion) Hmm, its possible to mount a screw mount on most (all?) major brand DSLRs, isn't it? It would at least make sure the tester was paying more than lip service to the concept of valid testing procedure. William Robb
