 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
| PhotoKB HomeDiscussion Groups Digital Photography Digital PhotoDSLR CamerasZLR CamerasPoint & Shoot CamerasFilm Photography 35 mmLarge FormatMedium formatDarkroomFilm and LabsOther EquipmentPhoto Technique Nature PhotographyPeople PhotographyTechnique GeneralGeneral Photo Topics General TopicsAustralian PhotographyUK PhotographyDirectoryPhoto Clubs |
Photo Forum / Digital Photography / DSLR Cameras / July 2005Bayer artifacts
http://www.geocities.com/angels2000photos/bayer.jpg The large text on that image is at 400% - you can see the pixels clearly. The smaller text is 100%. Desaturated crops for comparison of how it *should* have looked. Text was on a sign held by a group in a job I did a few days ago, and just printed. I didn't notice the coloration at normal viewing size on screen, and at 100% must have missed it. But it was quite noticeable on a 10 x 8 print - as it is in the center of an area that is otherwise totally black text on a white sign. Weird - I've never seen that before. Shot with a D70 / 18-70 lens. Testimony perhaps to the sharpness of the lens - able to resolve detail to the pixel level where bayer artifacts (if that's what they are called) can actually be very significant. I'm trading in the D70 on a Sigma dslr as soon as I can find one for sale ;-) >http://www.geocities.com/angels2000photos/bayer.jpg >The large text on that image is at 400% - you can see the pixels [quoted text clipped - 12 lines] >I'm trading in the D70 on a Sigma dslr as soon as I can find one for >sale ;-) That's the D70's weak AA filter.  Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>http://www.geocities.com/angels2000photos/bayer.jpg >>The large text on that image is at 400% - you can see the pixels [quoted text clipped - 14 lines] > > That's the D70's weak AA filter. Yes - I guess it is - but if more anti-aliasing was applied to eliminate this effect, would the text still be able to be read? Incidentally, this example was from a raw image - and no sharpening applied. I have no idea if in camera processing would have an effect. >>>http://www.geocities.com/angels2000photos/bayer.jpg >>>The large text on that image is at 400% - you can see the pixels clearly. [quoted text clipped - 12 lines] >>>I'm trading in the D70 on a Sigma dslr as soon as I can find one for sale >>>;-) Have you tried duplicating that test with film? Or with a different camera/lens perhaps you can give all the parameters of the test, so people can try a similar setup and see what happens. > Have you tried duplicating that test with film? Or with a different > camera/lens perhaps you can give all the parameters of the test, so people > can try a similar setup and see what happens. It won't occur with film. It wasn't a test - it was something that just happened. To reproduce it, you would need to shoot very small black text on a white background with a sharp enough lens to resolve the detail, with a digital camera that has a weak anti-aliasing filter. I guess I could back calculate the size of text by measuring the size in pixels of the text. I think that they are about 10 pixels high. That is 1/200th of the height of the frame, or about 1.25mm high on a print at 300dpi. >>>>http://www.geocities.com/angels2000photos/bayer.jpg >>>>The large text on that image is at 400% - you can see the pixels clearly. [quoted text clipped - 16 lines] > camera/lens perhaps you can give all the parameters of the test, so people > can try a similar setup and see what happens. I have added to the image. http://www.geocities.com/angels2000photos/bayer.jpg After digging around, I found a 10x8 Cibachrome that I 100% know for sure was shot using Fuji 50 iso film with a 50mm f1.4 AI-s Nikkor, using flash exposure indoors - so I know that there is no camera shake, and although I don't know the f-stop used, it must have been at least f5.6 or f8, based on the DOF. The text is actually slightly larger - closer to 1.75mm high. Much larger text alongside it is readable, the exposure is spot on. Try as hard as I could with USM, that is about the best I could do with it. It would be interesting to see how the original "technology" text resolved on state of the art lab equipment. >Yes - I guess it is - but if more anti-aliasing was applied to eliminate >this effect, would the text still be able to be read? >Incidentally, this example was from a raw image - and no sharpening >applied. I have no idea if in camera processing would have an effect. Try stopping down to the smallest aperture the lens has. That should work as a makeshift AA filter. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>Yes - I guess it is - but if more anti-aliasing was applied to eliminate >>this effect, would the text still be able to be read? [quoted text clipped - 3 lines] > Try stopping down to the smallest aperture the lens has. That should > work as a makeshift AA filter. lol - it probably would. Now let's see - the width of the text is almost exactly one pixel wide x 10 high. Unfortunately, the Airy circle at f22 is only then about twice the diameter of the sensor pixel. Methinks that might give a reduction, but not eliminate it totally. > http://www.geocities.com/angels2000photos/bayer.jpg > The large text on that image is at 400% - you can see the pixels [quoted text clipped - 6 lines] > - as it is in the center of an area that is otherwise totally black text > on a white sign. Just added at the bottom of that image a 1200dpi scan from the print. You can't read the text easily without a magnifying glass - but you can see the colours very clearly. > I'm trading in the D70 on a Sigma dslr as soon as I can find one for > sale ;-) That would be a huge step backwards. > http://www.geocities.com/angels2000photos/bayer.jpg > The large text on that image is at 400% - you can see the pixels > clearly. The smaller text is 100%. Desaturated crops for > comparison of how it *should* have looked. Is this a camera JPG original or is it converted from NEF? If the latter - what RAW conversion software did you use? Signature- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ] ------------------------------------------------------------------------ Kodak DCS460, Canon Powershot G5, Olympus 2020Z ------------------------------------------------------------------------ >>http://www.geocities.com/angels2000photos/bayer.jpg >>The large text on that image is at 400% - you can see the pixels [quoted text clipped - 3 lines] > Is this a camera JPG original or is it converted from NEF? > If the latter - what RAW conversion software did you use? Good thinking - thanks! Shot in raw - converted to JPEG using UFRaw(DCRaw)... Redone using Nikon RAW converter - no sharpening, just a tweak to get it close to the same brightness/contrast as the other. The result was amazing. I have uploaded it - please take a look. There is some problem with DCRaw/UFraw. One apparent thing is that the files are 3038 x 2012, not 3008 x 2000. I think that you can see this as the crops from what was the same NEF file are a little smaller. The way that the text from the nikon converter "snaps" seems like some sharpening must be applied - even though none was specifically applied. But the alignment of pixels to details in the text is also different, so something is not quite right with DCraw perhaps. >>>http://www.geocities.com/angels2000photos/bayer.jpg >>>The large text on that image is at 400% - you can see the pixels [quoted text clipped - 16 lines] >But the alignment of pixels to details in the text is also different, so >something is not quite right with DCraw perhaps. It's not so much something "wrong" with dcraw; you're just seeing some artifacts caused by the particular choice of reconstruction algorithm dcraw has chosen. The Nikon converter apparently uses a different reconstruction algorithm, so you don't see those artifacts (although it probably suffers from some different problems elsewhere). (This is one of my major complaints about raw file processing; there is no one-size-fits-all 'best' reconstruction algorithm for all uses, yet all raw converters lock you into a single algorithm, and with no publicly available information about which algorithm is being used) >(This is one of my major complaints about raw file processing; there >is no one-size-fits-all 'best' reconstruction algorithm for all uses, >yet all raw converters lock you into a single algorithm, and with no >publicly available information about which algorithm is being used) The adobe RAW converter uses algorithm 67-Delta-34, and DCRaw uses one known to a select few as 'Melissa'. I started working on an algorithm called 'Jane' after Melissa's rather cute cousin... for obvious reasons, but had to abandon the project due to a lack of beer. Does this help? ...I thought not. -- Owamanga! http://www.pbase.com/owamanga >>> http://www.geocities.com/angels2000photos/bayer.jpg >>> The large text on that image is at 400% - you can see the pixels >>> clearly. The smaller text is 100%. Desaturated crops for >>> comparison of how it *should* have looked. >> what RAW conversion software did you use? > Good thinking - thanks! > Shot in raw - converted to JPEG using UFRaw(DCRaw)... [quoted text clipped - 8 lines] > But the alignment of pixels to details in the text is also different, > so something is not quite right with DCraw perhaps. Very interesting. Would you mind sharing the NEF? I want to check out how ACR and C1 deals with this image. Signature- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ] ------------------------------------------------------------------------ Kodak DCS460, Canon Powershot G5, Olympus 2020Z ------------------------------------------------------------------------ >>>>http://www.geocities.com/angels2000photos/bayer.jpg >>>>The large text on that image is at 400% - you can see the pixels [quoted text clipped - 18 lines] > Very interesting. Would you mind sharing the NEF? > I want to check out how ACR and C1 deals with this image. No problems. I'll have to try uploading it later tonight (work to do now)- the nef is just over 5mb, so not sure if I might have to split it to get it uploaded to the geocities server (IIRC they have a 5mb limit per file on the free plan). >>>>> http://www.geocities.com/angels2000photos/bayer.jpg >>>>> The large text on that image is at 400% - you can see the pixels [quoted text clipped - 23 lines] > to get it uploaded to the geocities server (IIRC they have a 5mb limit > per file on the free plan). Couldn't upload it sorry. Anything over a mb or so, and the Geocities server was timing out on me or something. (Time for me to get adsl and some web storage) Do you still want it? Incidentally, Rawshooter Essentials seems to yield the same result as the Nikon RAW converter. > >>http://www.geocities.com/angels2000photos/bayer.jpg > >>The large text on that image is at 400% - you can see the pixels [quoted text clipped - 16 lines] > But the alignment of pixels to details in the text is also different, so > something is not quite right with DCraw perhaps. I am glad you got it working better, you really would not have been happy with a Sigma camera. Scott > I am glad you got it working better, you really would not have been > happy with a Sigma camera. > > Scott Why is that so? > > I am glad you got it working better, you really would not have been > > happy with a Sigma camera. > > > > Scott > > Why is that so? Because it costs a fortune for a 3.2MP kiddie toy that yields Homer Simpson skin tones? >> > I am glad you got it working better, you really would not have been >> > happy with a Sigma camera. [quoted text clipped - 5 lines] >Because it costs a fortune for a 3.2MP kiddie toy that yields Homer >Simpson skin tones? According to the "What Digital Camera" article on the various sensors out there, (they voted the Fuji S3's the best) the Sigma did deliver very good colour rendition once the white balance issue was addressed. However, noise was unacceptable past 100ISO. -Rich >>> > I am glad you got it working better, you really would not have been >>> > happy with a Sigma camera. [quoted text clipped - 11 lines] > However, noise was unacceptable past 100ISO. > -Rich Noise above ISO 100 is not good with the SD9 but is more than acceptable above ISO 200 with the SD10. >Noise above ISO 100 is not good with the SD9 but is more than acceptable >above ISO 200 with the SD10. You're talking about hue noise, no? I don't see much luminance noise in SD shots at all. The noise I see looks like the blue and green layers steal from each other in blotches. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>Noise above ISO 100 is not good with the SD9 but is more than acceptable >>above ISO 200 with the SD10. > > You're talking about hue noise, no? I don't see much luminance noise in > SD shots at all. The noise I see looks like the blue and green layers > steal from each other in blotches. Yes, hue noise. >> > I am glad you got it working better, you really would not have been >> > happy with a Sigma camera. [quoted text clipped - 5 lines] > Because it costs a fortune for a 3.2MP kiddie toy that yields Homer > Simpson skin tones? Back to your cave Randall, please don't come back until you have something of substance to add. >>> > I am glad you got it working better, you really would not have been >>> > happy with a Sigma camera. [quoted text clipped - 8 lines] > Back to your cave Randall, please don't come back until you have something > of substance to add. Forget substance, I'd just be happy for some variety... SignatureSkip Middleton http://www.shadowcatcherimagery.com > Forget substance, I'd just be happy for some variety... Then discuss something other than the alleged quality of Sigma products. >> Forget substance, I'd just be happy for some variety... > > Then discuss something other than the alleged quality of Sigma products. In case you hadn't noticed, RA, there is a plethora of discussions here not involving the quality, or lack thereof, of Sigma lenses. I haven't noticed much participation from you on any level in any of these, maybe I missed something... Sigma does give an alternative to OEM lenses to people who cannot afford top grade mfr. glass, like Canon's "L" stuff. The Sigma EX line is pretty good, better than a lot of Canon's midline stuff, and on a par, cost wise, with it. Some of their stuff is better than the equivalent Tamron and Tokina lenses, or at least on par with them. You say it's for people who are "too cheap" to buy the good mfr. lenses, well, there are a lot of people who can't afford a $1500 16-35 f2.8 L, or can't justify the cost vs. use, but a $500 17-35 f2.8-4 works for them. When I first bought my Canon AF gear, the 17-35 Sigma was tested by Photodo as pretty comparable to the then current Canon 17-35, but at 1/3 the price. I couldn't afford the Canon, nor could I justify waiting and saving to buying it, since WA isn't a critical part of my oeuvre. Yes, I've since bought the 16-35 f2.8, a much better lens than the earlier Canon, but at the time it was not an option. The Sigma 15mm f2.8 fisheye is another example, less in cost, and pretty much as good as the Canon version. And frankly, they're both much better built than my 28-135 f3.5-5.6 IS. But then, so's my 28-70 f2.8 Tokina... Now, if you want to slam Phoenix/Vivitar/Cosina, you'll not get much argument... SignatureSkip Middleton http://www.shadowcatcherimagery.com > Now, if you want to slam Phoenix/Vivitar/Cosina, you'll not get much > argument... I can't afford "L" glass either. My lenses are both consumer-grade Canon lenses, but the 16x20 & 20x24 prints I've had made look great. As for Vivitar, Cosina, etc. - yeah...they suck too. I don't think Vivitar makes anything, do they? They're a marketing company. >> Now, if you want to slam Phoenix/Vivitar/Cosina, you'll not get much >> argument... [quoted text clipped - 4 lines] > As for Vivitar, Cosina, etc. - yeah...they suck too. I don't think > Vivitar makes anything, do they? They're a marketing company. Actually, Cosina owns the Vivitar name, after years of just being a marketing company, or under the umbrella of Hanimex. SignatureSkip Middleton http://www.shadowcatcherimagery.com > Actually, Cosina owns the Vivitar name, after years of just being a > marketing company, or under the umbrella of Hanimex. I remember Hanimex. I was never much into 35mm, but did make the mistake of buying 3rd party lenses way back when. Want a real dog? Remember Lentar? >> Actually, Cosina owns the Vivitar name, after years of just being a >> marketing company, or under the umbrella of Hanimex. > > I remember Hanimex. I was never much into 35mm, but did make the > mistake of buying 3rd party lenses way back when. Want a real dog? > Remember Lentar? Oh, yeah, I remember that name. SignatureSkip Middleton http://www.shadowcatcherimagery.com cool replie! cool reply! > > I am glad you got it working better, you really would not have been > > happy with a Sigma camera. > > > > Scott > > Why is that so? Well others have already hit the high points, The Sigma has a lot of noise even at ISO 800, what's more the noise is an odd splotchy kind of noise. One of the reasons to buy DSLR is the low noise at high ISOs that you can get. With the Sigma you are pretty much stuck with their lenses. The raw only shooting is crazy. The camera is just not selling so it my well go out of production. Scott >>>I am glad you got it working better, you really would not have been >>>happy with a Sigma camera. [quoted text clipped - 11 lines] > > Scott The "odd splotchy" noise is very easily removed with simple Photoshop plugins and results in a sharper image than you'll get from a 20D's software noise suppression system. ISO 800 is not the ideal speed for any digital camera. They will all produce noise at that speed. As for Sigma lenses... The professional ones are every bit the equal of Canon lenses. http://www.tecphoto.com.au/pelican.htm is just one example. Sigma may not sell in numbers anywhere near Canon's or Nikon's but then... Jaguar cars don't sell in the Chevie or Ford numbers either. Should they also go out of production? > >>>I am glad you got it working better, you really would not have been > >>>happy with a Sigma camera. [quoted text clipped - 16 lines] > software noise suppression system. ISO 800 is not the ideal speed for > any digital camera. They will all produce noise at that speed. Compare the Sigma at ISO 800 http://www.imaging-resource.com/PRODS/SSD10/FULLRES/SD10LL8003us.HTM To the Canon 20D at ISO 800 http://www.imaging-resource.com/PRODS/E20D/FULLRES/E20LL0803.HTM > As for Sigma lenses... The professional ones are every bit the equal of > Canon lenses. http://www.tecphoto.com.au/pelican.htm is just one > example. Sigma may not sell in numbers anywhere near Canon's or Nikon's > but then... Jaguar cars don't sell in the Chevie or Ford numbers either. > Should they also go out of production? The Sigma is hardly a high end camera, they are getting just about $30 more for it then the Canon 20D, and who would buy the Sigma when you could get the 20D at the same price, it turns out hardly anyone. Could Jaguar stay in business if they sold their cars for the same price as a Ford but could only sell as many cars as the do? The digital camera market is extremely competitive and we are going to see any number of companies get out of the business. The Sigma is slow, does poor in low light, just about matches Canon older camera, the 10D, for resolution. If they hope to stay in the DSLR business they are going to have to come out with a new camera that is a lot more then the SD10 real soon. Scott > The Sigma is slow, does poor in low light, just about matches Canon > older camera, the 10D, for resolution. If they hope to stay in the DSLR > business they are going to have to come out with a new camera that is a > lot more then the SD10 real soon. I don't think so. A 3.42MP kiddie toy that yields Homer Simpson skin tones hardly compares to the 10D. >>The Sigma is slow, does poor in low light, just about matches Canon >>older camera, the 10D, for resolution. If they hope to stay in the DSLR [quoted text clipped - 3 lines] > I don't think so. A 3.42MP kiddie toy that yields Homer Simpson skin > tones hardly compares to the 10D. Words from the wannabe specialist who absolutely knows the facts but has never owned any of the gear he lambastes... Brilliant! Douglas > > I don't think so. A 3.42MP kiddie toy that yields Homer Simpson skin > > tones hardly compares to the 10D. > > Words from the wannabe specialist who absolutely knows the facts but has > never owned any of the gear he lambastes... Brilliant! Wanna be specialist? Don't get that one. I don't wanna be anything. And no, I've never owned anything by Sigma. I've been too smart to waste my money on any of it. The truth is tha I've never been much of a 35mm kinda guy. For most of my photographic career I've been a Hasselblad freak. > > I don't think so. A 3.42MP kiddie toy that yields Homer Simpson skin > > tones hardly compares to the 10D. > > Words from the wannabe specialist who absolutely knows the facts but has > never owned any of the gear he lambastes... Brilliant! So you're telling me that the Sigma digital camera is not 3.42MP and does not give yellow skin tones? > > > I don't think so. A 3.42MP kiddie toy that yields Homer Simpson skin > > > tones hardly compares to the 10D. [quoted text clipped - 4 lines] > So you're telling me that the Sigma digital camera is not 3.42MP and > does not give yellow skin tones? Have you even looked at one review of the SD10, I am not a fan of it but at least if have read the reviews and looked at the test photos. Compare the photos from the SD10 and the 10D and you will see very little difference. I got to tell you that you are giving those of us that bash the SD10 a bad name. BTW by the time I am done I will have probably pissed off both the people who like the SD10 and those who hate it. Scott > > The Sigma is slow, does poor in low light, just about matches Canon > > older camera, the 10D, for resolution. If they hope to stay in the DSLR [quoted text clipped - 3 lines] > I don't think so. A 3.42MP kiddie toy that yields Homer Simpson skin > tones hardly compares to the 10D. Well first off the really bad color probelms were with the SD9, the SD10 does much better. But also note that I did not say the SD10 was as good as a 10D, it is clearly not, what I said was they its resolution was close to the same, and it is. Calling the SD10 a kiddie toy is being rather silly, the camera mostly takes very good photos, but it is not as good as others that are out there, well in fact just about any of the DSLRs that are out there. Scott Scott >Calling the SD10 a kiddie toy is being rather silly, the camera mostly >takes very good photos, but it is not as good as others that are out >there, well in fact just about any of the DSLRs that are out there. It does a great 3.43MP greyscale, if you don't mind the aliasing. I suppose also, that the color would be a lot better if you had a filter that cut the middle of the green/blue range of the spectrum, or used a flash or a light that was very weak in that range. You might lose some saturated colors in that range, but overall, the image will look better with normal response curves for the blue and green channels. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< J...@no.komm wrote: > >Calling the SD10 a kiddie toy is being rather silly, the camera mostly > >takes very good photos, but it is not as good as others that are out [quoted text clipped - 8 lines] > with normal response curves for the blue and green channels. > -- Here is a comparison of the SD10 to the Canon 10D. http://www.dpreview.com/reviews/sigmasd10/page16.asp To my eye the images from the 10D look just a bit better, but look at the photos, they are not crap they are in fact pretty good, not good enough that anyone should buy the camera but in true the photos are not bad. The SD9 was utter crap, but they have fixed a lot of the problem from that one. Also a note to Gorge P. if you are reading any of this, the SD10 does not have more resolution then a Canon camera, it has less. Scott Randall Ainsworth 1 jul 03:46 opties weergeven Nieuwsgroepen: rec.photo.digital.slr-systems Van: Randall Ainsworth <r...@nospam.techline.com> - Berichten van deze auteur zoeken Datum: Thu, 30 Jun 2005 18:46:16 -0700 Lokaal: vr 1 jul 2005 03:46 Onderwerp: Re: Bayer artifacts Beantwoorden | Auteur beantwoorden | Doorsturen | Afdrukken | Afzonderlijk bericht | Origineel weergeven | Misbruik melden In article <1120139546.414313.156...@o13g2000cwo.googlegroups.com>, Scott W <biph...@hotmail.com> wrote: > The Sigma is slow, does poor in low light, just about matches Canon > older camera, the 10D, for resolution. If they hope to stay in the DSLR > business they are going to have to come out with a new camera that is a > lot more then the SD10 real soon. >I don't think so. A 3.42MP kiddie toy that yields Homer Simpson skin >tones hardly compares to the 10D. cool again! >> > I am glad you got it working better, you really would not have been >> > happy with a Sigma camera. [quoted text clipped - 11 lines] > > Scott The SD10 is certainly going out of production to make way for the next SD camera. > http://www.geocities.com/angels2000photos/bayer.jpg > The large text on that image is at 400% - you can see the pixels clearly. [quoted text clipped - 12 lines] > I'm trading in the D70 on a Sigma dslr as soon as I can find one for sale > ;-) Those text 'artefacts' are pretty much par for the course with a D70. I found them very annoying. However, don't give up on Bayer just because of the D70 - Pentax exhibits no such artefacting (nor, for that matter, does the Canon 350) Personally I found that the D70 was vastly overrated - a big lightweight (without the kit lens) plastic camera that produced over sharpened images with the sort of problems you highlighted. Try a Pentax *ist D - good quality, good images, and a knockdown price as it gets phased out. >Personally I found that the D70 was vastly overrated - a big lightweight >(without the kit lens) plastic camera that produced over sharpened images >with the sort of problems you highlighted. You can of course adjust the amount of sharpening in custom settings, or use raw for best results. SignatureKen Tough >>http://www.geocities.com/angels2000photos/bayer.jpg >>The large text on that image is at 400% - you can see the pixels clearly. [quoted text clipped - 25 lines] > Try a Pentax *ist D - good quality, good images, and a knockdown price as it > gets phased out. Thanks for your suggestions. If you follow the thread to the conclusion, and look at the images I posted again, then you can see that the D70 with kit lens can clearly resolve detail down to the pixel level, and with the manufacturer's raw converter, it does that with effectively no artifacts, and with remarkable preservation of detail. If the pentax can do the same, then it is also a good camera. The D70 may have some faults - like all cameras, but construction quality, handling and general ergonomics are excellent. >>>http://www.geocities.com/angels2000photos/bayer.jpg >>>The large text on that image is at 400% - you can see the pixels clearly. [quoted text clipped - 34 lines] > The D70 may have some faults - like all cameras, but construction quality, > handling and general ergonomics are excellent.< I never had the chance to try the Nikon raw converter - it wouldn't work on an AMD 64 bit machine. Nikon were, frankly, less than interested, but I did use RAW and opened with ACR & (best of all) Raw Shooter Essentials. I don't remember whether there were artefacts on the RAW files or not - but some of the Jpeg's were affected in the way you illustrated. I've been quite pleased with the Pentax - not least because of the good cheap lenses that I've picked up on ebay. However, my lad has just bought a Casio E100 for work - and the images are pretty good from the tiniest, thinnest camera I've ever seen!. so much so that I feel pretty miffed at lugging bag full of DSLR + lenses + filters + Flash + tripod + kitchen sink...... all to obtain images that are better - but not earth shatteringly better than the ones he can get out of a cheap camera small enough to lose in a shirt pocket! I'm beginning to seriously wonder whether the whole 'entry level' DSLR craze isn't simply a stupendous fraud, fuelled by vanity and techno-pretension. I think that it's been said in here before that a reasonable digican is more than adequate for most peoples needs about 85% of the time - and, boy, don't we pay through the nose for the remaining 15% of high quality?! >>>>http://www.geocities.com/angels2000photos/bayer.jpg >>>>The large text on that image is at 400% - you can see the pixels clearly. [quoted text clipped - 58 lines] > than adequate for most peoples needs about 85% of the time - and, boy, don't > we pay through the nose for the remaining 15% of high quality?! Trapezium, You said that those artifacts were "pretty much par for the course with a D70. I found them very annoying" I can assure you that those artifacts are not par for the course for a D70. They were clearly the result of the method used for post-processing of RAW files, and they are something that I've never seen on an ex-camera jpeg. Now you tell me that you use a Pentax *ist D, then state your concern "whether the whole 'entry level' DSLR craze isn't simply a stupendous fraud, fuelled by vanity and techno-pretension". To trade a D70 for an *ist D, as you imply you have done (or even vice versa), is just a little crazy in my opinion. My D70 cost me less than I paid for a new FM2. At that time, a cheap minolta 35mm zoom camera I also owned was capable of taking nice shots 85% of the time. What's changed? >>>>>http://www.geocities.com/angels2000photos/bayer.jpg >>>>>The large text on that image is at 400% - you can see the pixels [quoted text clipped - 80 lines] > minolta 35mm zoom camera I also owned was capable of taking nice shots 85% > of the time. What's changed? I'm not entirely sure what you mean.... I had a D70 for a couple of weeks and was generally disappointed with it for several reasons, some of which were aesthetic, and some related to image quality. In particular, I found the images to be over sharpened - initially pleasing, but ultimately 'harsh' looking. I also saw those artefacts (if that's the correct term for them) in Jpeg's that featured text (such as the lettering on the side of cans and bottles, etc) They were very pronounced and seemed (to me) to be related to the general tendency of the camera to over-sharpen. So I returned the D70 and went with the Pentax D - which seems (to me) to produce overall better images - perhaps not as superficially sharp, but I'd bet a Pentax RAW converted with RSE and subsequently sharpened in PS to any NEF. What is inarguable is that the Pentax has never produced the artefacts on lettering that the D70 did, It's all very subjective, though - and I'm certainly not rubbishing your choice of a D70, everyone has a different take on these things. As for the P&S versus DSLR debate - I'm really impressed by the images taken with my lads EX-S100, a camera so small it's hardly bigger than a credit card. No, of course they're not *as* good as DSLR images - but I really do have to question the validity of carrying thousands of pounds worth of kit around in a large bag, just to acquire images that are better - but, like I said, not earth shatteringly better. I know that, for myself, one of the reasons that I bought a DSLR was because I fancied the 'image' that it conveys - I could afford one, so I bought one, but I couldn't honestly say that I'm a great deal happier with my expensive toys that I would have been with an EX-S100 (and, when lugging the paraphernalia around on a hot summer day, I have to admit to a yearning for a credit card size camera that would still give me decent images. I suppose this means that I haven't got the soul of a true photographer - well, hey-ho, that's life. I suppose I'll keep the Pentax because I've got it - but I can also see myself borrowing mad lad's camera more and more frequently.... Enjoy your D70 - and congratulations on solving your original problem. >>>>>>http://www.geocities.com/angels2000photos/bayer.jpg >>>>>>The large text on that image is at 400% - you can see the pixels [quoted text clipped - 89 lines] > In particular, I found the images to be over sharpened - initially pleasing, > but ultimately 'harsh' looking. You did make a mistake buying any dslr if that was your conclusion. That is a crazy assessment of something that you apparently did not realise you have complete control over. > I also saw those artefacts (if that's the correct term for them) in Jpeg's > that featured text (such as the lettering on the side of cans and bottles, > etc) They were very pronounced and seemed (to me) to be related to the > general tendency of the camera to over-sharpen. I think you spent too much time looking at 400% on the computer screen. > So I returned the D70 and went with the Pentax D - which seems (to me) to > produce overall better images - perhaps not as superficially sharp, but I'd > bet a Pentax RAW converted with RSE and subsequently sharpened in PS to any > NEF. What is inarguable is that the Pentax has never produced the artefacts > on lettering that the D70 did, Look again at the text on the image I uploaded, and tell me that resolution is superficial, or the result of sharpening. The sample converted from raw with nikon raw converter would be about 1mm high on a print at 10x8 inches, and you could read it with a magnifying glass. > It's all very subjective, though - and I'm certainly not rubbishing your > choice of a D70, everyone has a different take on these things. [quoted text clipped - 15 lines] > I suppose this means that I haven't got the soul of a true photographer - > well, hey-ho, that's life. Did you ever own a 35mm slr? You don't show understanding of the technical skills required to make the most of what is available - you could learn if you were interested. If you don't want to do that, then you are certainly better to sell the dslr on ebay and buy something that fits in your pocket. > I suppose I'll keep the Pentax because I've got it - but I can also see > myself borrowing mad lad's camera more and more frequently.... > > Enjoy your D70 - and congratulations on solving your original problem. Thanks. |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.