 |
 | 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 / Digital Photo / April 2006Benefits of more megapixels?
Suppose I have two cameras. No.1 has 3 megapixels No.2 has 6 megapixels Suppose that both are set to the same resolution and I take 2 identical shots, one with each camera. I then send these 2 shots off to a Photo processors for printing and I order 2 prints of each shot at the following sizes: 10" x 8" and 4" x 6" The 10" x 8" print from camera No.2 will be better quality than the 10" x 8" print from camera No.1. Agreed? My question is this: Will the 4" x 6" prints be the same quality or will the 4" x 6" print from camera No.2 be better? >The 10" x 8" print from camera No.2 will be better quality than the 10" >x 8" print from camera No.1. Agreed? Correct. 3 mpix theorically looks good at 4x6" and smaller but will show some artifact (barely noticeable) at 10x8" >My question is this: >Will the 4" x 6" prints be the same quality or will the 4" x 6" print >from camera No.2 be better? That may depend on the equipment they use but human eye won't see the difference between 3 mpix or 6 mpix on a 4x6" prints.  SignatureWhen you hear the toilet flush, and hear the words "uh oh", it's already too late. - by anonymous Mother in Austin, TX Spam block in place, no emil reply is expected at all. The difference between a lower and higher pixel count camera becomes more obvious when the picture is cropped, and then expanded to fit a standard picture size. > >The 10" x 8" print from camera No.2 will be better quality than the 10" > >x 8" print from camera No.1. Agreed? [quoted text clipped - 8 lines] > That may depend on the equipment they use but human eye won't see the > difference between 3 mpix or 6 mpix on a 4x6" prints. So, assuming the two cameras and settings are exactly the same except for the amount of pixels and there's no cropping involved or anything else then there *will* be a difference in the quality of the respective 6 x 4 prints but it will be un-noticeable by the human eye? If that's the case then it presumably would make no difference as far as a smallish print goes between a 4 mega pixel camera and say a 30 mega pixel camera. The major difference being only the size of the print you could get without artifacts. So assuming you don't want extra large prints then a 5-6 mega pixel camera should be be good enough. After that I suppose it's down to lens quality Is there anything else to consider? Thanks > The difference between a lower and higher pixel count camera becomes more > obvious when the picture is cropped, and then expanded to fit a standard [quoted text clipped - 11 lines] >> That may depend on the equipment they use but human eye won't see the >> difference between 3 mpix or 6 mpix on a 4x6" prints. Today Ben Hardy commented courteously on the subject at hand > So, assuming the two cameras and settings are exactly the > same except for the amount of pixels and there's no [quoted text clipped - 10 lines] > Is there anything else to consider? > Thanks See the replies I just posted, Ben, then come back with some more data for me and the others to work from. Theoreticals are just that, theoretical. You need to help us help you by saying what cameras you're testing and how you're using them. Besides lens, sensors, and storage format, consider also whether the comparos are daylight, indoors with flash, indoors available light at high ISO, what have you. There is far, far more to this biz than immediately meets the eye. Take exposure as one simple to understand example. A properly exposed image will /always/ produce a better print than an overexposed image lacking in contrast with all the highlights blown out, or a severely underexposed image with no shadows but black ink and lots of noise in what's left of the shadow detail and the mid-tones from the camera overamplifying the sensor signal trying its damndest to give you an image. >> The difference between a lower and higher pixel count >> camera becomes more obvious when the picture is cropped, [quoted text clipped - 18 lines] >>> Austin, TX Spam block in place, no emil reply is expected >>> at all. SignatureATM, aka Jerry >A properly >exposed image will /always/ produce a better print than an [quoted text clipped - 3 lines] >detail and the mid-tones from the camera overamplifying the >sensor signal trying its damndest to give you an image. At the risk of sounding pedantic, I must say that the issue is not really the camera's amplification. If an image is under-exposed, the camera doesn't know or care about this when amplifying the sensor data. It gets amplified the same, at any given ISO, regardless of exposure. If you're talking about shooting at a higher ISO instead of a lower one, then, again, the issue is not the amplification, but rather, the low sensor exposure. Multiplying a low exposure in software is more detrimental to image quality, generally speaking, than amplifying in the camera. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< Today commented courteously on the subject at hand >>A properly >>exposed image will /always/ produce a better print than an [quoted text clipped - 15 lines] > exposure in software is more detrimental to image quality, > generally speaking, than amplifying in the camera. That is as much a sweeping generalization based on unfounded assumptions as mine. I should have said "all factors being equal, but they seldom are." The reasons for noise at high ISO and noise resulting from underexposure, particularly when severe, are oft debated everywhere I go on Usenet where they talk about digital camera. Not everybody thinks as I do, but many do. When I complained about noise at ISO 400 and higher on my new Rebel XT on rec.photo.digital.slr-systems, I was roundly flamed for complaining about a problem I caused by incorrectly exposing with flash in museums (upping ISO theoretically extends flash range) and for complaining about unacceptable noise (to me) when shooting available light at 800 and 1600. I also know from personal experience about buying bleeding edge technology, and I know a few pros and "serious amateurs" with high-end Nikon and Canon DSLRs who tell me exactly the same thing wrt to the broad-based issue of noise related to ISO and image size for a given level of particular camera families. "The proof of the pudding is in the eating", and WYSIWYG come to most people's minds when talking about noise. If you see it, it is there, regardless of what "experts" may say. SignatureATM, aka Jerry Sorry folks, maybe I didn't explain properly. The question is one of theory/principle really. This is why I don't mention any particular models. What I'm trying to understand is how the available pixels are implemented in a digital camera. Let's take two imaginary cameras. In every respect they are identical except for the CCDs. Camera A has a CCD of 4 MPx. The CCD on camera B is 16 MPx. Now we take a picture with each camera. Let's say the two pictures are taken at the same time with the same settings of the same subject from the same position etc. Again everything remains equal except the CCDs. We then send these two images to same Print Processor where once again everything remains equal. He sends us the two printed images, both of the same size - say 6" x 4". My question is this: Will there be a noticeable difference (to the human eye) between the two prints? > Today commented courteously on the subject at hand > [quoted text clipped - 42 lines] > to most people's minds when talking about noise. If you see > it, it is there, regardless of what "experts" may say. Today Ben Hardy commented courteously on the subject at hand > Sorry folks, maybe I didn't explain properly. > The question is one of theory/principle really. This is why [quoted text clipped - 4 lines] > Camera A has a CCD of 4 MPx. > The CCD on camera B is 16 MPx. There's really no such thing. > Now we take a picture with each camera. Let's say the two > pictures are taken at the same time with the same settings [quoted text clipped - 4 lines] > images, both of the same size - say 6" x 4". > My question is this: There's also no such thing as this, unless you literally used the exact same camera and exact same settings for both images. As soon as you bring a 2nd camera in, you change from apples- to-apples to apples-to-pomegranites, it isn't even as close as apples-to-oranges. CCDs are only 1/4-1/3 of the equation. The lens is a part, the JPEG compression is a big part, and the "quality" of the math in the camera's firmware A-to-D conversion, image processing, and writing to memory is another. Again, the old saying "all things being equal, but they seldom are" is true here. > Will there be a noticeable difference (to the human eye) > between the two prints? I don't think your scheme is going to work the way you'd like, Ben. It just isn't possible for anyone to render a useful opinion if they don't know at least something about what you're looking at. It just isn't as simple as looking up the specs on the sensor size and type, nor is it enough to look at the contrived manufacturer's specs, nor even fully rely on reviews, such as dpreview.com. At the risk of being crude and insulting you, "opinions are like a.sholes, everybody has one." Including me, so how do you know if I'm knowledgeable or a bullshit artist? Did you look at alt.binaries.pictures.autos at my latest Canon Rebel XT prints, as I think I suggested? I'm not holding these up as super pictures, as flash of cars in a museum is a long series of compromises, of which ultimate sharpness gets the biggent ding. So, image quality and print quality is highly subjective, as I've tried to illustrate. As to your latest question, you're still not providing nearly enough information to answer it intelligently. What software was used to create the print? What kind of printer was used? What is the resolution of the printer? What are the comparo subjects, lighting, exposure, et al like? At what distance are you viewing these prints, from across the room or at 2" away using a jeweler's loupe? "by the human" eye isn't enough to go on, I'm afraid. Again, "beauty is in the eyes of the beholder" and "the proof of the pudding is in the eating," so in the final analysis, only you would know what is and what is not a "good" print. Let me tell you a story. The "experts" will vehemently disagree, but that's OK. I have been scanning and printing for 15 years and only went digital 5 years ago. But, I've printed to 8x10, 11x17 and 13x19 with my HP1220C using only a 1280 x 960 image. At the big size, that is only 67 PPI, less than 1/3 of the presumed minimum. But, I put those big prints made on glossy paper on a wall of my overall office where people couldn't get closer than about 3-4 feet. So, at that distance, they looked just fine. But, at the distance I stood putting them up with putty, the pixelation and posterization was simply outrageous. The 8x10s (really 8x10.3 printable area on 8.5x11 paper) were on the back wall of my 12x12 office, where people sat a couple feet away, or stood outside my office and looked. More times than you might imagine, people would come rushing in gushing on me about the quality of my /Kodak/ enlargements! No, I ain't bragging, I'm just saying that a decent image /can/ be printed to even large sizes if you're skillful, and if you keep people from putting their noses up to your prints. Ask more specific, pointed questions, and I'll give you the benefit of my experience. But, it is still up to you to evaluate everything you've read and learned in light of your own requirements and defintion of "good". SignatureATM, aka Jerry > Today Ben Hardy commented courteously on the subject at hand > [quoted text clipped - 8 lines] > > There's really no such thing. We're at cross purposes here. I'm talking theoretical. Obviously the scenario I describe is impossible in reality. I'm just attempting to understand what difference the number of MPx makes. >> Now we take a picture with each camera. Let's say the two >> pictures are taken at the same time with the same settings [quoted text clipped - 7 lines] > There's also no such thing as this, unless you literally used > the exact same camera and exact same settings for both images. Exactly! That's what I'm saying. The exact same cameras in *every* aspect except for the number of MPx. > As soon as you bring a 2nd camera in, you change from apples- > to-apples to apples-to-pomegranites, it isn't even as close as > apples-to-oranges. Forget every other aspect of the camera apart from the MPx. I just want to understand what difference the number of MPx makes. > CCDs are only 1/4-1/3 of the equation. The lens is a part, the > JPEG compression is a big part, and the "quality" of the math > in the camera's firmware A-to-D conversion, image processing, > and writing to memory is another. Again, the old saying "all > things being equal, but they seldom are" is true here. As above. OK, lets go further into my imaginary scenario. Suppose that you could take a picture with just the CCD and nothing else - no lens, no firmware, no image processing, no writing to memory, no nothing, just two absolutely (albeit imaginary) identical CCDs that take the very same picture at the very same moment in time from the exact same position. The one difference is that one CCD has the capacity for 4 MPx and the other has the capacity for 16 MPx. >> Will there be a noticeable difference (to the human eye) >> between the two prints? [quoted text clipped - 25 lines] > using a jeweler's loupe? "by the human" eye isn't enough to go > on, I'm afraid. You misunderstand my question. The software was the same in each case. I'm not using a printer - I'm sending the pictures off to a photo processor who will provide me with prints just like he did when I used to send off my 35mm films to him. The lighting, exposure, subjects are the same in each case. I will view each print at exactly the same distance in the same light with my eyes. > Again, "beauty is in the eyes of the beholder" and "the proof > of the pudding is in the eating," so in the final analysis, > only you would know what is and what is not a "good" print. This is not about about subjective quality. This is about measurable quality. If I was to take a picture with a 4MPx camera and have two prints made, one 4ft by 6ft and the other 4inches by 6 inches it would be quite obvious to my eye (at say 3ft distance) that the smaller print would be sharper. > Let me tell you a story. The "experts" will vehemently > disagree, but that's OK. I have been scanning and printing for [quoted text clipped - 21 lines] > evaluate everything you've read and learned in light of your > own requirements and defintion of "good". OK, lets say that I only *ever* want to send my pictures off to my photo processor and have delivered 4" by 6" prints every time. Furthermore I only *ever* want to view my prints with my eyes at 3ft distance. Would it benefit me to have a 4MPx CCD or a 32 MPx CCD? By the way, thanks for your reply Ben Today Ben Hardy commented courteously on the subject at hand >>> Now we take a picture with each camera. Let's say the two >>> pictures are taken at the same time with the same [quoted text clipped - 10 lines] > Exactly! That's what I'm saying. The exact same cameras in > *every* aspect except for the number of MPx. If, and this is a very big IF, the camera's "natural" resolution is the highest MP, then /theoretically/ you will get more detail with more MP. But, this is hardly true in the real world, so I don't see the point. It's like how many angels can dance on the head of a pin or if a tree falls and there's nobody there, does it make a noise? The former has no answer, the latter does, see the diff? >> As soon as you bring a 2nd camera in, you change from >> apples- to-apples to apples-to-pomegranites, it isn't even >> as close as apples-to-oranges. > Forget every other aspect of the camera apart from the MPx. > I just want to understand what difference the number of MPx > makes. If you want to throw all the other stuff that matters, then you've almost answered your own question. The only thing left, /in theory/ is to leap to the conclusion that in a perfect world, all cameras are created equal wrt to mega pixels. Ain't so. >> CCDs are only 1/4-1/3 of the equation. The lens is a part, >> the JPEG compression is a big part, and the "quality" of [quoted text clipped - 11 lines] > capacity for 4 MPx and the other has the capacity for 16 > MPx. Are you writing a master's or PhD thesis or some sh.t, where you have to see if you can enticee idiots to play your game? If so, I quit, you win, I lose. If you're not playing a game, I see no point whatsoever in eliminating all the physical stuff that makes a digital camera work. There's mathematics and physics here, you cannot ignore them. > You misunderstand my question. > The software was the same in each case. [quoted text clipped - 4 lines] > I will view each print at exactly the same distance in the > same light with my eyes. What the hell are you doing, Ben? There is simply no such thing as D-to-A, so it is patently impossible to take any digital format and send it to Kodak. >> Again, "beauty is in the eyes of the beholder" and "the >> proof of the pudding is in the eating," so in the final [quoted text clipped - 6 lines] > eye (at say 3ft distance) that the smaller print would be > sharper. Then, get out an oscilloscope and examine the signals coming directly from the CCDs. That is phyisically the only way to bypass all the rest of the stuff in the camera, no matter how loud you should "theoretical." > By the way, thanks for your reply You are going in the troll dumpster. I'm happy to have made your day by playing some dumb a.s game. SignatureATM, aka Jerry I think you perhaps misunderstand English. > All Things Mopar wrote: > What the hell are you doing, Ben? There is simply no such > thing as D-to-A, so it is patently impossible to take any > digital format and send it to Kodak. Check this out: http://www.kodak.com/eknec/PageQuerier.jhtml?ncc=uk&lcc=&pq-path=11/3149/3144&pq -locale=en_GB > You are going in the troll dumpster. I'm happy to have made > your day by playing some dumb a.s game. I'm sorry you've taken offense at my questions. They arise as a result of discussions I've had where the argument has been made that: "Unless you want to print huge poster-size pictures then a 5MPx camera is sufficient" I take exception to (in plain English: I'm offended by) your unwarranted attitude. If you're incapable of civility then I suggest you get off this newsgroup and go play with your cars. > Sorry folks, maybe I didn't explain properly. > The question is one of theory/principle really. This is why I don't [quoted text clipped - 13 lines] > Will there be a noticeable difference (to the human eye) between the two > prints? If you take the image at the maximum image size of each CCD for comparison. Very unlikely that you will be able to tell any difference with your eyes by looking at the 6" X 4" prints. In the first place, because of the limits of photographic paper, you will not have more than an apparent 300-400 DPI image on either print. The difference in the CCDs would be apparent if you make a very large print to compare. Like a 14" X 11" print from both CCDs. SignatureCSM1 http://www.carlmcmillan.com -- Thank you very much for that. That's just the sort of reply I was hoping for. It confirms what I thought. The counter argument to this was that the more MPx a camera has then the more detail will be apparent, even in relatively small prints. Do you know of any site that gives more detail about the limits of photographic paper and the processes used by such as Kodak to transfer digital images to paper? Many thanks Ben >> Sorry folks, maybe I didn't explain properly. >> The question is one of theory/principle really. This is why I don't [quoted text clipped - 26 lines] > > Like a 14" X 11" print from both CCDs. The page is about the Fuji Frontier Printer which is used in a lot of photo processing labs such as Walmart and Walgreens. http://bermangraphics.com/press/frontier.htm SignatureCSM1 http://www.carlmcmillan.com Reply address is false. -- > Thank you very much for that. That's just the sort of reply I was hoping > for. It confirms what I thought. The counter argument to this was that [quoted text clipped - 36 lines] > > > > Like a 14" X 11" print from both CCDs. >Is there anything else to consider? Zoom: always look at optical zoon. digital zoom is nothing more than enlargement and cropping which can make the end result look ugly. I've had digital zoom fucntion turned off on every camera I've owned. Memory card slot: make sure there's one as some cheap camera may not have any slot for extra pictures. You got a choice of CF, MS, xD, and SD. MMC and SM are both pretty much obsolete and limited to 128 max anyway. SignatureWhen you hear the toilet flush, and hear the words "uh oh", it's already too late. - by anonymous Mother in Austin, TX Spam block in place, no emil reply is expected at all. Today Impmon commented courteously on the subject at hand >>Is there anything else to consider? > [quoted text clipped - 7 lines] > choice of CF, MS, xD, and SD. MMC and SM are both pretty > much obsolete and limited to 128 max anyway. In most/all cameras, digital zoom is a simple pixel resize, guaranteed to introduce nasty pixelation. If the camera has enough MP and you need more than the optical zoom can give you, set the size to the max and crop the subject out of the middle, giving you an effective "digital zoom". I do this at car museums when I can't get close enough to a car, engine or sign. Other things to consider, depending on budget and need, are a hot shoe for an external flash. Built-in flashes are pretty puny, they seldom have a range more than 10, maybe 12 feet. For a moderate amount of money, you can buy an external that'll get you to 25 feet or more. Also, evaluate the camera's noise at ISO above 200. You'll need high ISO if you intend to do night shots or indoor available light. Depending on your budget, size requirements and purpose, take a look at the advanced photometric options available in each camera's set-up menu. Do your own research and/or pick them up in the store and try them out. The store sales people in camera stores will be knowledgeable but folks in the discount stores don't have a clue. Pick each camera you're considering up and handle it some. Does it fit the size of your hands? Is it too large, or too small? Is it too heavy, or so light it'll wobble? Are the controls well-placed? Can you change settings on-the-fly without diddling for 5 minutes? Is the LCD on the back large enough and bright enough to evaluate the correctness of the image, etc. You can do all these things before you buy it. And, most important of all, buy only from a store that will allow you to return it for refund if you don't like it for any reason. Take it for a test drive and shoot a few hundred pictures representative of the subjects and lighting conditions you expect to shoot in real life. Look at image quality at various resolutions and JPEG compressions and evaluate overall quality factors such as correct exposure, color balance, etc. SignatureATM, aka Jerry >Zoom: always look at optical zoon. digital zoom is nothing more than >enlargement and cropping which can make the end result look ugly. >I've had digital zoom fucntion turned off on every camera I've owned. Whether or not I'd leave DZ off would depend on how it is implemented. On my Fuji P&S camera, if I jump from optical, and then go all the way into the digital, what it gives is a crop, so if most of the frame without the DZ would just be empty, worthless space, then the cropping is useful, inasmuch as it saves space, lets you see what you have better, in review, and may reduce the time required before you take the next shot (and you may see the subject better as well, depending on the camera). Another thing is that if digital zoom is exactly 2x, then the lower quality of JPEG is slightly abated by the oversampling. Of course, values like 1.3x and 1.6x are detrimental to image quality, if it is done by interpolation rather than cropping. This would probably be best implemented, however, by having a separate "2x" control that just turns on and off. Yet another way to look at it is that if you are shooting the camera set to small-sized output images, digital zoom is actually necessary to get most of the detail the sensor is capturing, as opposed to an after-the fact crop. If, for whatever reason, you have a camera set to output 640*480 JPEGs, you can use a good deal of digital zoom to your advantage in that context. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< I don't use the highest pixel setting on my Dcam, but it is a 3 MP. My pics are 1024 x 768. I burned some pics to CD and had them printed up at Wal-Mart as 4 x 6's. Since that particular kiosk supposedly prints water-resistant prints, I favored it based on what others who have used it said about it. My mom took some of those Wal-Mart prints and had them blown up for a calendar but the resolution looked awful to me. So I printed those same prints on my computer (7.5" x 10" size) and told her to try these prints. Since the paper I used was glossy, the scanner tinted the pictures badly and the prints looked rather "flat" (whatever that is); but the resolution looked much improved by using my large prints instead of Wal-Marts' small prints on the scanner. We kept with the poor-resolution prints from Wal-Mart for the calendar, unfortunately. My point is that small prints from Wal-Mart will have very poor resolution when blown up. Using only 786,432 pixels (1024 x 768) on my camera and printing up big from my computer ain't all that bad in comparison. I rarely print anyway, and I take several thousand pictures a year now (which I could never do before I got this camera...heck I couldn't even do 5 rolls a year for many years!). I prefer to view 95%+ of my pictures in a computer moitor. If I have a special project, I might up my resolution on the camera. I just took 69 pictures this morning! It sounds like you should go from camera to PC, and then to whatever you want to print with. Scanning a gloss finish picture and then trying to upsize is a very lossy process. It also sounds like your scanner settings need a bit of "tweaking". Default scanner resolution is usually 300 or less. Scanning a moderately sized picture should use a resolution no less than 300, and likely 600 for better results. smaller pictures generally require higher settings. For instance, I use 1200 for scanning slides. Resolution and color accuracy are often compromised by retail photo processors such as the one you mentioned. It seems that everyone prefers contrasty pictures with more saturation. The photo processing equipment is usually setup with this in mind. Had you used the 3mp setting, and printed the picture with a decent printer on 8 1/2-11 paper,you would normally get a very acceptable picture. A great picture at 8 1/2-11 might require about a 5-6MP camera. Many forget to find out what resolution the printer is using when they print pictures. Usually "standard" resolution is around 320, "Hi res" around 720, and maximum 720 or more. The resolutions higher than 720 tend to involve using "tricks" with dot size, positioning, and multidot printing. Many of the printers state in the details that printing above 720 or so can only be accomplished when using the OEM printer driver. (Epson and Canon, for example) > I don't use the highest pixel setting on my Dcam, but it is a 3 MP. My pics are > 1024 x 768. I burned some pics to CD and had them printed up at Wal-Mart as 4 x [quoted text clipped - 18 lines] > I have a special project, I might up my resolution on the camera. I just took 69 > pictures this morning! From what I've read, a 3 MP camera provides good resolution up to about a 4 inch by 6 inch print. If you enlarge it to 8 by 10 or there about, degradation will occur. You need a 5 MP camera for that. > From what I've read, a 3 MP camera provides good resolution up to about a 4 > inch by 6 inch print. If you enlarge it to 8 by 10 or there about, > degradation will occur. You need a 5 MP camera for that. Not so, in my experience. I've made quite nice 8X10 prints from my 3.1 Mp camera. It depends in part, of course, on the amount of fine detail that you need to show. >> From what I've read, a 3 MP camera provides good resolution up to about a >> 4 inch by 6 inch print. If you enlarge it to 8 by 10 or there about, >> degradation will occur. You need a 5 MP camera for that. > Not so, in my experience. I've made quite nice 8X10 prints from my 3.1 Mp > camera. It depends in part, of course, on the amount of fine detail that > you need to show. Hi Marvin. Which camera do you own? >>>From what I've read, a 3 MP camera provides good resolution up to about a >>>4 inch by 6 inch print. If you enlarge it to 8 by 10 or there about, [quoted text clipped - 5 lines] > > Hi Marvin. Which camera do you own? I have an Oly C-3020Z, which I bought in early 2002, to replace a stolen C-3030. Whenever i've thought of getting something newer, I've had better uses for the money. >>>>From what I've read, a 3 MP camera provides good resolution up to about >>>>a 4 inch by 6 inch print. If you enlarge it to 8 by 10 or there about, [quoted text clipped - 8 lines] > C-3030. Whenever i've thought of getting something newer, I've had better > uses for the money. I know exactly what you mean. My camera is a Concord Eye-Q 3103 (3.1 MP, no optical zoom, and no macro). It just died, and is not worth it to repair, so I'm looking around for a replacement. My budget is pretty low, so I'm hoping I can find something on Ebay used. I have taken, printed and had developed excellent 8X10 pics using 3 megapixels. The most importand feature on a camera is the lense. Many 3 meg cameras with good lenses can get much higher quality shots than a 5 meg camera with a cheap/insufficient lense. > From what I've read, a 3 MP camera provides good resolution up to about a > 4 inch by 6 inch print. If you enlarge it to 8 by 10 or there about, > degradation will occur. You need a 5 MP camera for that. The higher pixel cout cameras (assuming a decent lens) give you the option to do serious cropping, and still maintain the detail when enlarging to fit larger prints. Also remember that many lenses are less than perfect at the corners and edges of a picture. There are other issues, perhaps not as obvious. More pixels and a larger CCD device mean less light is available for each pixel. In addition, more pixels can mean more in camera processing time. > I have taken, printed and had developed excellent 8X10 pics using 3 > megapixels. The most importand feature on a camera is the lense. Many 3 meg [quoted text clipped - 3 lines] > > 4 inch by 6 inch print. If you enlarge it to 8 by 10 or there about, > > degradation will occur. You need a 5 MP camera for that. Ben, One other thing to consider, even in your theorecital of just the CCDs being different....the CCDs can be different SIZES too. That is, one could be 4 MPx and be .5" in size and the other could be 16 MP and .8" (or whatever) in size. And the size of the sensor related to the resolution of the lens plays a part. A high quality lens can function with sharp focus on a smaller sensor than a lower quality lens. Remember that the common sensors size is much less than the 35mm film camera frame. > Ben, > > One other thing to consider, even in your theorecital of just the CCDs > being different....the CCDs can be different SIZES too. That is, one > could be 4 MPx and be .5" in size and the other could be 16 MP and > .8" (or whatever) in size. Today Chuck commented courteously on the subject at hand > The difference between a lower and higher pixel count > camera becomes more obvious when the picture is cropped, > and then expanded to fit a standard picture size. That's true, but the OP has to be doing an apples-to-apples comparion for their test to have any meaning at all. No fair cropping the two images differently and/or resizing one of them - up or down. The OP also said nothing about aspect ratio, which is far different for 8 x 10 than 4 x 6, and no camera will be exact for both without cropping in one or the other direction. >> >The 10" x 8" print from camera No.2 will be better >> >quality than the 10" x 8" print from camera No.1. Agreed? [quoted text clipped - 14 lines] >> Austin, TX Spam block in place, no emil reply is expected >> at all. SignatureATM, aka Jerry Today Impmon commented courteously on the subject at hand >>The 10" x 8" print from camera No.2 will be better quality >>than the 10" x 8" print from camera No.1. Agreed? > > Correct. 3 mpix theorically looks good at 4x6" and smaller > but will show some artifact (barely noticeable) at 10x8" JPEG artifacts and posterization/pixelation are two entirely different things. The former is caused almost exclusively by over compression, while the latter is most often caused by too little PPI for the desired print size. >>My question is this: >>Will the 4" x 6" prints be the same quality or will the 4" [quoted text clipped - 3 lines] > won't see the difference between 3 mpix or 6 mpix on a 4x6" > prints. SignatureATM, aka Jerry >Suppose I have two cameras. >No.1 has 3 megapixels [quoted text clipped - 11 lines] >Will the 4" x 6" prints be the same quality or will the 4" x 6" print >from camera No.2 be better? That depends on the exposure levels. All other things being equal, the 6 MP is going to have more noise, which is more noticeable with weaker exposures. If the noise is bad enough, it will show even on the 4*6. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< Today commented courteously on the subject at hand > In message > <43ed1ad4$0$6978$ed2619ec@ptn-nntp-reader02.plus.net>, Ben [quoted text clipped - 20 lines] > more noticeable with weaker exposures. If the noise is bad > enough, it will show even on the 4*6. All other factors being equal, but they seldom are, shooting at the "natural" resolution of the sensors will yield the least noise. That /may/ be the highest mega pixels or maybe one down from the max. And, the OP not only said no-thing about the camera(s), they also said no-thing about ISO, the #1 cause of noise if improperly used, particularly when the image is underexposed. I'm making a ton of assumptions in my replies, but a 6 MP camera from just a couple of years ago will be less sharp and much noisier than a 6 MP from the same manufacturer's newer model today. If you don't want a lot of noise, /never/ buy a bleeding edge camera for max mega pixels in your budget range. Camera maker's first attempt at attaining the next higher marketing bullshit number will almost always be noisier than the "improved" model they quietly release in 6-12 months. In the final analysis, when all the "experts" have had their say, their isn't anything that'll match a test drive. My last 2 digitals were all bought from a store that would allow a return if I didn't like them for any reason. If your store won't do that, find another one! SignatureATM, aka Jerry Today Ben Hardy commented courteously on the subject at hand > Suppose I have two cameras. > No.1 has 3 megapixels [quoted text clipped - 7 lines] > The 10" x 8" print from camera No.2 will be better quality > than the 10" x 8" print from camera No.1. Agreed? Not necessarily. What exactly are the 2cameras, how do you have them set to save the images, and what different did you do, if anything, when you post-processed them? It seems to me there is a world of difference between a $100 P & S and a Canon 20D in your example. Or, even a decent EVF vs. a Nikon D70S or Canon Rebel XT. > My question is this: > Will the 4" x 6" prints be the same quality or will the 4" > x 6" print from camera No.2 be better? Prevailing wisdom says you need a minimum of 200 PPI to get a decent print, 300 if you can. But, viewing distance has a lot to do with it, as does exposure, contrast/brightness, detail level of the subject and background, etc. So, for your example you could get by with an image as small as 1200 x 800 = 0.9 mega pixels for 4 x 6. And, for the 8 x 10 example, you'd likely be OK with just 2000 x 1600 = 3.05 MP. That does /not/ mean that more mega pixels won't be better, I just don't have enough information to even hazard a guess. If you're saving to JPEG, make damn well sure you're using the absolute /least/ compression. Cameras usually call this "fine", as opposed to "normal", or something like that. And, without knowing anything about your camera, I'll just say that the lens quality and sensor size/quality are really big factors, as is the "quality" of the camera's image processing algorithms. SignatureATM, aka Jerry |
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.