 |
 | 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 / July 2004can one print at actual pixels size?
Can one print at actual pixels size (100 %)? How is this done?  Signaturenobody >Can one print at actual pixels size (100 %)? Of course, as long as you don't mind cropping your your image, or you have access to a big enough printer. You can go bigger too. I've printed a number of my 10D (6.3MP) photos at 10"x15" & 400DPI, which is roughly 400%. > How is this done? Depends on the image, the software you have, & the printer. If you explain what you're trying to achieve, we can probably give you response that are much more useful than this one. :) SignatureW . | ,. w , "Some people are alive only because \|/ \|/ it is illegal to kill them." Perna condita delenda est ---^----^--------------------------------------------------------------- > Can one print at actual pixels size (100 %)? How is this done? Can you rephrase that question? A pixel will have whatever size you print it, or do you mean printing at the same size as the sensor? Bart >From: nobody nowhere nobody@jwhite.demon.co.uk > >Can one print at actual pixels size (100 %)? How is this done? If you mean "can you print so the print is the same size as what's on the screen at 100%" (which is what I *think* you mean), then yes, you can do this IF you can figure out the rez of your screen. Many older monitors have a dot pitch of 72 dpi (which is why that's a common default number in many graphics programs) and many new ones are around 96 dpi. So I'd try setting the image size rez at one of these numbers and printing at that size, but the print will be really big. Note that most printers like 200-300 ppi input file rez so the print will be pretty thin and low rez for, say, the Epson desktops. To put numbers on this, a typical 6 Mpixel dSLR file might be 3,072 x 2,048 pixels (for the Canon 10D, say). Set the resolution to 72 ppi and it will print 28.444 x 42.667 inches, set the rez to 96 ppi and it will print 21.333 x 32 inches, set the rez to 240 ppi (enough to get a nice print on most printers) and it's 8.5 x 12.8 inches. Bill Thank you Bill, that is exactly what I meant. Probably my reputation in this NG will be damaged for the next 3 days because of my mental laziness to work out the answer myself. When I scan a 6 x 6 at 14 bit, I get huge files (450MB or so), which at 100% magnification still look good. The further question would be if I treat this huge image at 72 ppi with Neatimage and perhaps sharpen it with unsharp mask, would it look on paper as good as it does on the monitor? Would the printer (who likes 266 ppi and higher ) make it look worse than on the monitor? Has anybody tried this? (I mean somebody who owns an Epson 7600 or even 9000 or "bigger"). n article <20040629133141.23881.00000902@mb-m28.aol.com>, Bill Hilton <bhilton665@aol.comedy> writes >>From: nobody nowhere nobody@jwhite.demon.co.uk >> [quoted text clipped - 18 lines] > >Bill Signaturenobody > Thank you Bill, that is exactly what I meant. Probably my reputation in > this NG will be damaged for the next 3 days because of my mental [quoted text clipped - 3 lines] > with Neatimage and perhaps sharpen it with unsharp mask, would it look > on paper as good as it does on the monitor? No, of course not unless it's a rather large print (20x30 maybe?) viewed from an extreme distance. I like my printed images to be at least 200ppi; I always shoot for 300. > Would the printer (who > likes 266 ppi and higher ) make it look worse than on the monitor? Yes. > Has > anybody tried this? (I mean somebody who owns an Epson 7600 or even 9000 > or "bigger"). I only have a 870, but I know it would look like crap w/o trying it. Mark hmm this is or appears to be an extension of what is a pixel? A pixel is a picture element A printer pixel is not the same as a monitor pixel and both are not the same as a sensor pixel (sometimes called sensels) A monitor relies on physical attribute of light - usually called additive which combines 3 primary colors to obtain "white" A printer (usually) relies on subtractive properties of inks so comining a combination of all 3 inks produces something called "black" See the difficulties? What is intended is bto have 1 to 1 correspondence across the physical properties of each device in order to render an accurate image. The image will always be subject to the constraints of the physical limitations on the hardware, software & drivers in place. Plus - there is the aspect of human perception. While eyes allow light to excite receptors the brain interprets these signals into an image. Now the above is a gross over simplification. What was the question? Oh - yeh. It depends upon having the correctly configured hardware, software firmware. After that it's easy within the constraints listed above das B > Can one print at actual pixels size (100 %)? How is this done? ps - seel Gisle's link below for even deeper reading on the effect called interpolation it's sorta important and for any budding mathematical genius out there - yeh, it really is all math das B > Can one print at actual pixels size (100 %)? How is this done? nobody nowhere <nobody@jwhite.demon.co.uk> wrote in news:ytcGk $DC7Z4AFwOk@jwhite.demon.co.uk: > Can one print at actual pixels size (100 %)? How is this done? A pixel has no size. So - no you cannot print at pixel size. > nobody Hmmm .... methinks you are trolling? /Roland [] > A pixel has no size. So - no you cannot print at pixel size. [] > Hmmm .... methinks you are trolling? > > /Roland Most pixels have a finite and well-defined size - can you quote examples where the pixel has "no size"? Of course, printing at the original pixel size may not be quite what the OP wants! Cheers, David > Most pixels have a finite and well-defined size - can you quote examples > where the pixel has "no size"? The pixels stored in a file has no size. A pixel as a concept has no size. The pixels in my graphic card has no size. A pixel can be matched to something that has a size, e.g. a sensor, a monitor or a print. You often also call this physical realisation a pixel, but this is misleading. > Of course, printing at the original pixel > size may not be quite what the OP wants! Yepp - and this (probable) mistake shows that you shall think of a pixel as having no size. /Roland > > Most pixels have a finite and well-defined size - can you quote examples > > where the pixel has "no size"? > > The pixels stored in a file has no size. It is sometimes not stated what the size is, but and real-world image stored in a file does have a size. > A pixel as a concept has no size. .. although a zero-sized pixel could capture no information... > The pixels in my graphic card has no size. .. an interesting idea. Can the pixel exist without being displayed? <G> > A pixel can be matched to something that has > a size, e.g. a sensor, a monitor or a print. [quoted text clipped - 8 lines] > > /Roland Nearly back to confusion caused by the 72ppi or the 300ppi image! Thanks for the diversion. David > It is sometimes not stated what the size is, but and real-world image > stored in a file does have a size. A real world image stored in a file has a size? OK then - how big is a real world 5 Mpixel image? In inches or meters, whatever you prefer. >> A pixel as a concept has no size. > > .. although a zero-sized pixel could capture no information... Uh? Have I said zero size? No size does not mean zero size. No size means that it is not measured in meters. Just like a bit or a digit has no size, a pixel has none either. The number 45 has no size, it can be displayed at any size and still be 45. Just as a pixel can be displayed at any size and still be a pixel. And both contain information, without having any size. >> The pixels in my graphic card has no size. > > .. an interesting idea. Can the pixel exist without being displayed? Yes - and I think it is here you go wrong. There are lots of pixels in graphics files, in RAM and on graphics cards. Neither needs to be displayed to exist. > Nearly back to confusion caused by the 72ppi or the 300ppi image! > > Thanks for the diversion. You are welcome. But the diversion and confusion is totally on your side :) /Roland > A real world image stored in a file has a size? OK then - how > big is a real world 5 Mpixel image? In inches or meters, > whatever you prefer. You need to know the taking conditions, so that the angular resolution can be computed. > >> A pixel as a concept has no size. > > [quoted text clipped - 7 lines] > and still be a pixel. And both contain information, without > having any size. My point is that a pixel which is of zero physical size (angular or linear) could capture no information in any real-world sensor. > >> The pixels in my graphic card has no size. > > [quoted text clipped - 3 lines] > in graphics files, in RAM and on graphics cards. Neither needs to > be displayed to exist. But in any real world image the pixel has a size. You take my "displayed" argument too seriously - perhaps I missed the <G> or perhaps it is a philospohical discussion! > > Nearly back to confusion caused by the 72ppi or the 300ppi image! > > [quoted text clipped - 4 lines] > > /Roland No, I am never confused, I think! Cheers, David >My point is that a pixel which is of zero physical size (angular or >linear) could capture no information in any real-world sensor. You're talking about different pixels. A pixel in a physical image sensor has a size. (Two sizes, actually - the sensing area, and the spacing between adjacent pixels). A pixel in a physical print has a size too. But a pixel in an image data file, with no physical instantiation, does not have any particular size. That's what the other writer is saying. Dave > >My point is that a pixel which is of zero physical size (angular or > >linear) could capture no information in any real-world sensor. [quoted text clipped - 8 lines] > > Dave Thanks, Dave. I am aware of that, and simply trying to clarify my statement about zero physical size. A pixel in a data file requires meta-data to specify its size, though, as is often required in (for example) GIS applications. Cheers, David > > "David J Taylor" > > <david-taylor@blueyonder.co.not-this-bit.nor-this.uk> writes: [quoted text clipped - 17 lines] > meta-data to specify its size, though, as is often required in (for > example) GIS applications. Perhaps zero physical size is misleading, undefined would probably be better. For instance I can create a 3D model with no units and render it to a 800x600 image. The pixels then have no defined size at all. > My point is that a pixel which is of zero physical size (angular or > linear) could capture no information in any real-world sensor. A sensor has a size; a print has a size; a monitor has a size. The grid of pixels can be mapped on any of those and then the pixel (or rather the pixel spacing) is associated with a size, the pixel pitch. But this size has nothing to do with the pixel; the pixel is sizeless and can be mapped to any size. To give the size of a pixel is the same as give the length of a string. OK - a string might be 10 inches, and then _this_ string is 10 inches. But, in general a string has no length; it can be any length. The same goes for pixels. This sensor might have a pixel spacing of 10 um, but the pixel in the photo taken by this sensor has no size. If it had, then all prints made from this photo would be very small indeed :) > But in any real world image the pixel has a size. You take my > "displayed" argument too seriously - perhaps I missed the <G> or > perhaps it is a philospohical discussion! It is both philosophical and practical :) I think it is a good habit not to use sloppy language. I try my best (even thogh I am not English speaking) to use tha language correct. And - when you talk about a pixel you mean the abstraction pixel. And the abstraction pixel has no size. When you talk about size and pixels, you talk about the actual pixel spacing (or pitch) on a sensor, a monitor or a print. So - even when you talk about a sensor you don't talk about the size, you talk about the spacing (or pitch). The term "pixel size" has no meaning. Pixels in a file has absolutely no size and pixels on a detector has a pitch. /Roland groovio - way to go Perhaps another way to look at it is like this <o> <o> unless it is a really good picture element then look at it like this: <0> <0> or <-> <-> Anyway - a pixel is seemingly assumed context sensitive Like Roland's length of string but the string can be a cable, a strand, a thread, or ... depending upon Hey, thing of it as a strand! A strand of what? Exactly! das B > > My point is that a pixel which is of zero physical size (angular or > > linear) could capture no information in any real-world sensor. [quoted text clipped - 33 lines] > > /Roland [] > The term "pixel size" has no meaning. Pixels in a file has absolutely > no size and pixels on a detector has a pitch. > > /Roland I am afraid we will have to disagree there. Any real-world pixel has some sort of associated dimension, and it may be important to retain that dimension when it is in a file. Pixels in a detector will have finite size as well as pitch. Cheers, David >[] >> The term "pixel size" has no meaning. Pixels in a file has absolutely [quoted text clipped - 6 lines] >dimension when it is in a file. Pixels in a detector will have finite >size as well as pitch. no, I'm afraid you are wrong .. just like the definition of a mathematical "point" which has no size as well although it has an unambiguous spatial definition .. the "size" of a pixel (picture element) would only start to make sense if the size of the picture was defined if one would like to stretch the "definition" ... any "ccordinate" on a map is of the same nature ie it may have several defining parameters but it has no size .. I think people confuse the "3D" physical presence of "light sensors" and the mathematical result of what they produce and even the way this info is represented in realtime (phosphorus dots on a screen) FWIW [] > >I am afraid we will have to disagree there. Any real-world pixel has some > >sort of associated dimension, and it may be important to retain that [quoted text clipped - 12 lines] > info is represented in realtime (phosphorus dots on a screen) > FWIW To prove my statements incorrect: - please quote an example of a real-world pixel (i.e. one taken by a digital camera or similar) that does not have an implied physical size. and - please quote an example of a detector in which the pixels do not have a finite size. Cheers, David >[] >> >I am afraid we will have to disagree there. Any real-world pixel has [quoted text clipped - 19 lines] >- please quote an example of a real-world pixel (i.e. one taken by a >digital camera or similar) that does not have an implied physical size. I've been browsing the internet and the definition has become ambiguous to say the least to me the implied physical size you mention is only relevant when the pixel is represented on a medium .. in return I could ask you what the size of a pixel is in a file ? .. :-) which is real-world as well .. to me "pixel" means a picture element that has all the parameters to represent the whole gamut .. and this is independent of the media it is represented in. If you think pixels have a size then plse compute me the size of it in a CCD sensor, a screen and a printer ? hopefully find the same size in each... see http://www.micron.com/products/imaging/technology/pixel.html for "something" (aka solid state sensor) that to me is not a pixel as such but which seems to be the present days perception of pixel in technology .. notice the statement :"When the image sensor is read out, line by line, the pixel sequence comes out GRGRGR, etc., and then the alternate line sequence is BGBGBG, etc. This output is called sequential RGB (or sRGB)" so where is the "pixel" you represent on your screen ? :-) ..and the end statement : "...Since each pixel has been made sensitive only to one color (one spectral band), ...." given the nature of bayer sensors, extrapolating "pixels_with_a_size" to get "pixels_with_probably_some_other_size" is more than a funny concept to say pixels have a size .. so given the bayer sensor architecture, as provided in the above website, kindly provide me with the calculated size of the resulting extrapolated "pixel" ? :-) >and > >- please quote an example of a detector in which the pixels do not have a >finite size. exactly as I said .. a detector and its signal .. NOT a pixel .. FWIW [] > I've been browsing the internet and the definition has become ambiguous to say the least [] So, all I am saying is that real-world pixels do have an associated size, either from the camera sensor, from any subsequent interpolation (when the luminance and chromanance resolutions may not be the same as you point out), or when displayed or printed. Therefore such a real-world pixel can only be completely described when meta-data about its dimernsions is available, although such meta-data is not essential to our everyday use of pixels. Cheers, David <psst> I am surprised nobody pulled me up about a product being prime <tee hee> das B <0> <0> "imbsysop" <imbsysop@yahoo.com> wrote in message > - please quote an example of a real-world pixel (i.e. one taken by a > digital camera or similar) that does not have an implied physical size. Now - you say real world pixels again. What does it mean? My disk is full of images, stored in image files. Those images consists of pixels. The image files are real world things I assure you. I don't think that Seagate has started making disks of imaginary material yet. Those pictures I can move between several real world medias. The size the pixels take on those medias has no meaning for the image. Now - as I said elsewhere to you - a sensor is NOT an image. Nor is a monitor an image. A digital projector is also not an image. None of those divices has pixels - picture elements, because none of the divices has pictures/images. Those devices can detect or display images though. The projector is a nice example. It has probably a small LCD screen that is projected on a wall or something. Or you can project the image on dust steam or smoke. So - whats the size of a pixel of the image a projector displays? It depends on the distance from the wall to the projector, does it not? Now, let us forget that it is really the pixel pitch we are talking about. The actual size of the displayed pixel depends on the fill factor, which we almost always ignore. But let us call the pitch a size, just to simplify. So, whats the pixel size of a camera? The actual size of the sensor is rather uninterestng for the picture. It is the reality outside the camera that can be said containing the "real world pictures" - just as for the projector. So - when taking a picture, the pixels of the coming picture are all projected upon the real world - you have thousands of pixels in your face whne the picture is taken. So - how big are those pixels? It all depends on the distance from the camera - the further away from the camera, the bigger the pixels. Furthermore, for parts of the pucture that is out of focus - the puxels are not sharp. So - what do I want to tell you with all this rambling? Yepp - the sensels in your sensor are not pixels, because your sensor is not a picture. The picture is out there in the real world, either as a projection upon your wall, a printing on paper or ultimately as the real world picture you are taking. The actual size of the intermediate device and its sensing elements (sensels) has nothing at all to do with a size of a pixel. /Roland [] > Now - you say real world pixels again. What does it mean? It includes, but is not limited to, pictures of real-world images. As opposed to purely computer generated images - an artist using a graphics program, for example. [] > Now - as I said elsewhere to you - a sensor is NOT an image. > Nor is a monitor an image. A digital projector is also not > an image. None of those divices has pixels - picture elements, > because none of the divices has pictures/images. Thre is an image on the surface of the sensor, display, screen. [] > So, whats the pixel size of a camera? The actual size of the > sensor is rather uninterestng for the picture. It is the reality [quoted text clipped - 6 lines] > the bigger the pixels. Furthermore, for parts of the pucture that > is out of focus - the puxels are not sharp. There is a fixed well-defined angular size for each pixel taken a digital camera. It is given by the pixel pitch divided by the focal length. Some cameras (e.g. Fuji?) post process the sensor data to interpolate, and that will alter the angular size of the pixel. > So - what do I want to tell you with all this rambling? Yepp - > the sensels in your sensor are not pixels, because your sensor [quoted text clipped - 7 lines] > > /Roland The linear size will vary with the display or sensing device and taking or viewing distance, yes, but the angular size at taking time is defined. Cheers, David >[] >> Now - you say real world pixels again. What does it mean? [quoted text clipped - 10 lines] > >Thre is an image on the surface of the sensor, display, screen. as said, you fail to see the difference between the data that "describe/define" a picture and the representation thereof on any medium .. :-) [] > as said, you fail to see the difference between the data that > "describe/define" a picture and the representation thereof on any > medium .. :-) No, I am quite happpy to differentiate, but I argue that any real-world derived pixel requires some dimensional specification to be a complete representation, i.e. there needs to be more than just the pixels. Cheers, David > No, I am quite happpy to differentiate, but I argue that any real-world > derived pixel requires some dimensional specification to be a complete > representation, i.e. there needs to be more than just the pixels. Many cameras add 72 dpi in the file, some add 225 or 300 or something. Thats the only information in the image file that gives any kind of size information. Now --- the actual sensor pitch is much smaller than that; and that number is not stored in the file. Those numbers that are stored are supposed to help you in some way or another. I know that some cameras store a dpi value that corresponds to printing the picture with their own standalone printers. Some store 72 although such a coarse screen resolution is now obsolete and you almost never print at 72 dpi. Now David - let us be reasonable. Currently we don't agree. There are more people in this discussion that don't agree with your view. But frankly - it does not matter all that much. I understand where you are coming from and can appreciate your urge to convince us, just as I hope that you understand my urge to do the same. Discussing language semantics is fun :) Understanding the language makes us grow and learn. I recommend you to think about it once more. My (and others) point is that a pixel is an abstraction. The pixels you are talking about are just instances of that abstraction. Those instances have size, or rather, is organized in an 2D array with a pitch. It is perfectly OK to call the elements in such a physical array for pixels. The problem arises when you ask the question - "How big is a pixel?". That question has no answer. I hope you can see that? You must ask a question of the type - "How big are the pixels in my camera?". And even that is not totally unproblematic - because what you probably mean with that question is "What is the pixel pitch in my camera?". But maybe not - maybe you mean the actual size - maybe you wonder how many photons the sensor element can collect - then the size is much better to know than the pitch. /Roland [] > I recommend you to think about it once more. My (and others) point is > that a pixel is an abstraction. The pixels you are talking about are [quoted text clipped - 10 lines] > > /Roland Yes, a real world pixel will have a size (something like an equivalent capture area) and a grid spacing. I had omitted that distinction originally because, in a well-designed system, the sampling interval should be sufficiently small to avoid aliasing, and hence the pixel area will only affect the light capture efficiency. Of course, well all know that there are systems out there in the real world which do not avoid aliasing! I guess I am talking about pixels within the context of a digital photography newsgroup! Cheers, David > Yes, a real world pixel will have a size (something like an equivalent > capture area) and a grid spacing. I had omitted that distinction [quoted text clipped - 6 lines] > I guess I am talking about pixels within the context of a digital > photography newsgroup! We still don't agree :) Real world pixel is a very starnge notion. I personally think that all pixels are real world, except the very abstraction pixel. I have lots of real world pixels on my hard disc. I assure you that they are not from some other reality than ours :) And digital photographers does not only handle cameras. They also handle images stored on discs. But - as I said earlier. It is not all that important. What is important is that we understand and respect each other. Although we in the heat of the discussion (pretends?) not to understand a bit of the "nonsens" the other part is writing, we actually do. Or at least most of us do. /Roland [] > But - as I said earlier. It is not all that important. What is important > is that we understand and respect each other. Although we in the heat > of the discussion (pretends?) not to understand a bit of the "nonsens" > the other part is writing, we actually do. Or at least most of us do. > > /Roland Agreed. And our discussion has reminded me that both the active area and the inter-pixel spacing are important. Thanks, David >> Now - you say real world pixels again. What does it mean? > > It includes, but is not limited to, pictures of real-world images. As > opposed to purely computer generated images - an artist using a graphics > program, for example. The definition of a pixel has nothing to do how the picture is generated; if it is computer rendered, taken with a camera or a manipulated image taken with a camera, or a mix of all those things. You are making life complicated by introducing notions that is not neccessary. You are talking about "real world pixels". I have tried to tell you before that this has no useful meaning. The only meaning I can think of (and it is not really useful) is that when taking a picture of you, the pixels are in your face. >> Now - as I said elsewhere to you - a sensor is NOT an image. >> Nor is a monitor an image. A digital projector is also not >> an image. None of those divices has pixels - picture elements, >> because none of the divices has pictures/images. > > Thre is an image on the surface of the sensor, display, screen. If there is an image on the surface of your screen, I recommend you to take it back to the shop where you bought it. My screen is totally blank until I use it to display any image I want in any size I want. Your monitor is only a display device, there should be no images on the screen. I do understand what you are trying to say, but you are not using the words correctly. Misuse of words makes it easier to misinterprete what you are trying to say. /Roland >> - please quote an example of a real-world pixel (i.e. one taken by a >> digital camera or similar) that does not have an implied physical size. [quoted text clipped - 3 lines] >My disk is full of images, stored in image files. Those >images consists of pixels. .. (snipped for bandwidth) perfect explanatory concept ! :-) if they don't get it now it is hopeless .. :-) Nice website BTW pixel image sensor makes more sense as picture element image sensor n'est pas? we sorta assume that a 4MB sensor has 1-1 correspondence with output mage -wrong assumption! a sensor provides data that is used to compose pixels so 1 way to go is RAW to camera lcd RAW to monitor RAW to printer (or maybe RAW2 to RAW1 ?) but jpeg sorted this out some years ago perhaps the future holds TIFF for comsumers and taking it from the high end digital image processing suites? das B ps - it's good to see some high level intellect around here - I was getting bored I am the OP. I am sorry that I started a fight at the OK Coral! "Actual pixels" is a Photoshop term, as everybody knows, and means 100% magnification of the image on the monitor (I hope). Therefore, by printing at "actual pixels size", what I was really asking was whether a printer could print at that size, implying resolutions of perhaps 72 ppi or 96 ppi, and still get acceptable results. But I agree that my question was not clear enough. >> The term "pixel size" has no meaning. Pixels in a file has absolutely >> no size and pixels on a detector has a pitch. [quoted text clipped - 8 lines] >Cheers, >David Signaturenobody > I am the OP. I am sorry that I started a fight at the OK Coral! > "Actual pixels" is a Photoshop term, as everybody knows, and means > 100% magnification of the image on the monitor (I hope). Correct. > Therefore, by printing at "actual pixels size", what I was really > asking was whether a printer could print at that size, implying > resolutions of perhaps 72 ppi or 96 ppi, Yes, it would print. Let's say your monitor i 72 ppi, and when you look at the image in at 100% (i.e. Actual Pixels) it fills a space on the screen that's 10x15 in - and you want to print it as 10x15 in. Let's also assume that your printer is much higer resolution than your screen - for example 240 ppi. If you sent at 72 ppi image to a 240 ppi photoprinter, the printer will silently upsample he image to 240 ppi and print it. It will probably use some clever built-in upsampling algorithm, so the results will probably be better than if you printed in on a printer that actually had a resolution of 72 ppi (the latter would look really bad). > and still get acceptable results. That depends on a number of factors, including what you would consider acceptable and the quality of the built-in upsampling algorithm. I would not be happy with the quality, but YMMV. Signature- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ] ======================================================================== «To live outside the law, you must be honest.» (Bob Dylan) >From: Gisle Hannemyr gisle+njus@ifi.uio.no >Let's also assume that your printer is much higer resolution than >your screen - for example 240 ppi. [quoted text clipped - 3 lines] >probably use some clever built-in upsampling algorithm, so the >results will probably be better ... Actually they use a not-so-clever upsampling algorithm, 'nearest neighbor' instead of something better like bicubic :) And they don't do any sharpening ... that's why it's usually better to upsample and sharpen yourself rather than let the printer software do it for you. You can prove this easily enough with an Epson that allows you to write out the print file and save it ... no matter what size input file you provide the printer output file will be the same size ... you can duplicate this size *exactly* only if you resample using 'nearest neighbor'. Bill >>From: Gisle Hannemyr gisle+njus@ifi.uio.no > [quoted text clipped - 17 lines] > > Bill It might not be wise to conclude that all printers use the same method, even Epson printers. Unless, of course, you have something to document that they DO use the same method. I should think there would be as many methods as there are printers, and printer drivers. > I am the OP. I am sorry that I started a fight at the OK Coral! > "Actual pixels" is a Photoshop term, as everybody knows, and means 100% > magnification of the image on the monitor (I hope). Not being a Photoshop user, but with some experience in the digital imaging field, I have never heard of the term "actual pixels" before. No, not everybody knows some manufacturers unusual and proprietary definition of such terms. Cheers, David > I am the OP. I am sorry that I started a fight at the OK Coral! "Actual > pixels" is a Photoshop term, as everybody knows, and means 100% [quoted text clipped - 3 lines] > or 96 ppi, and still get acceptable results. But I agree that my > question was not clear enough. Simple answer to your revised question is NO. I am pretty sure you wouldn't be satisfied with a print with only 96 dpi resolution, at least not unless you intend to stand across a room to view it. >> The term "pixel size" has no meaning. > I am afraid we will have to disagree there. Any real-world pixel > has some sort of associated dimension, and it may be important to > retain that dimension when it is in a file. Pixels in a detector > will have finite size as well as pitch. Sure they have, but this is not a very interesting metric, at least not in the context of the question from the OP (see subject). For instance: Nobody in their right mind is going to print with a pixel pitch of 4 micrometers just because that happen to be the "size" of the distance between the pixels in the sensor array in their P&S. And what about pixels that doesn't originate from a sensor? I have a friend who uses a computer to make very nice geological maps based on seismic data files. Are the pixels in his image files different than pixels in image files that has been recorded by a CCD? In what way? I go with Roland Karlsson. The term "pixel size" does not make sense. Signature- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ] ======================================================================== «To live outside the law, you must be honest.» (Bob Dylan) [] > > I am afraid we will have to disagree there. Any real-world pixel > > has some sort of associated dimension, and it may be important to [quoted text clipped - 3 lines] > Sure they have, but this is not a very interesting metric, at least > not in the context of the question from the OP (see subject). Well, it now turns out that the OP was using some unusual terminology, something not made clear at the time. But the finte size of a pixel does matter, because it means that you need to have a rising spatial frequency response if a flat MTF is to be maintained. > And what about pixels that doesn't originate from a sensor? > I have a friend who uses a computer to make very nice geological [quoted text clipped - 4 lines] > I go with Roland Karlsson. The term "pixel size" does not make > sense. These seismic data files will surely have some associated dimensions, otherwise the feature scale would be unknown? Do his maps not have scales? Cheers, David >From: "David J Taylor" david-taylor@blueyonder.co.not-this-bit.nor-this.uk >Well, it now turns out that the OP was using some unusual terminology, >something not made clear at the time. It was clear to anyone who knows Photoshop :) > >From: "David J Taylor" david-taylor@blueyonder.co.not-this-bit.nor-this.uk > > >Well, it now turns out that the OP was using some unusual terminology, > >something not made clear at the time. > > It was clear to anyone who knows Photoshop :) Then the post might have been better entitled: Can one print at "actual pixels" size? Cheers, David >>From: "David J Taylor" david-taylor@blueyonder.co.not-this-bit.nor-this.uk >>Well, it now turns out that the OP was using some unusual terminology, >>something not made clear at the time. >It was clear to anyone who knows Photoshop :) But "actual pixels" in Photoshop is *not* a physical size specification for the image, since it depends on monitor pixel pitch. All it means is that each pixel in the image is mapped to 1, and only 1, pixel in the frame buffer. Nothing more. Dave >>It was clear to anyone who knows Photoshop :) >From: davem@cs.ubc.ca (Dave Martindale) > >But "actual pixels" in Photoshop is *not* a physical size specification >for the image It just means the file is viewed at 100% ... click on the Hand tool and you'll even see an option for it in the menu bar, where it says "Actual Pixels" and holding the cursor over it says "Zoom current window to 1:1". That's where the OP got the phrase, but he should have mentioned that when he first posted. I can see how people who don't know Photoshop terminology didn't know what he meant though ... > These seismic data files will surely have some associated dimensions, > otherwise the feature scale would be unknown? Do his maps not have > scales? Ehem? Are you now claiming that the pixel size must be related to the real world size of the things you are photographing? /Roland > > These seismic data files will surely have some associated dimensions, > > otherwise the feature scale would be unknown? Do his maps not have [quoted text clipped - 4 lines] > > /Roland No, simply questioning your statement that pixels comprising map or siesmic data contained no size information. Cheers, David > No, simply questioning your statement that pixels comprising map or > siesmic data contained no size information. I don't think the poster said that the map of seismic date was the pixels. I think he said that he made pictures (consisting of pixels) from seismic data. Then the pixels has no size information. For the pixels to contain any size information, you must also add distance information for all pixels. If the constructed picture has a well defined geometry, like a rectilinear lens, a fish eye lens or a panorama lens, then the pixels contain angular information. But no size information whatsoever. /Roland [] > If the constructed picture has a well defined geometry, like > a rectilinear lens, a fish eye lens or a panorama lens, then > the pixels contain angular information. But no size information > whatsoever. > > /Roland Which is just what I said several posts ago! Typical digital camera pixels are of a defined angular size, depending on the pixel spacing and lens focal length. Cheers, David > Which is just what I said several posts ago! Typical digital camera > pixels are of a defined angular size, depending on the pixel spacing and > lens focal length. Yes, I know. But it is not "angular size", it is just "angle". And - if we shall take this with angles too seriously, then the original poster will get a very strange answer indeed :) /Roland [] > Yes, I know. But it is not "angular size", it is just "angle". I think in English one might say: "a pixel subtends an angle of 1 milliradian", or "has an angular subtense of 1 milliradian". You wouldn't say it has "an angle of 1 milliradian" but you might say "it has an angular size of 1 miiliradian". Yes, you should be giving two dimensions, but you know what I mean! Angular subtense is the term I would use if I was trying to be "correct" in my English. > And - if we shall take this with angles too seriously, then the > original poster will get a very strange answer indeed :) > > /Roland Actually he gets just the same answer: the lower resolution pixels subtend an angle that is 2048/640 times greater than the higher resolution pixels. (or (2048/640)^2 [squared] times the solid angle. Cheers, David > [] > [quoted text clipped - 10 lines] > Cheers, > David David, The pixels have no 'size'. Or rather their apparent size varies depending on the display medium. erm how can a pixel have any real meaning independent of the picture it is part of? A picture composed of picture elements has pixels These pixels are in an ordered array - end of story :-) what may be confusing people is the way data are used to determine the ordering of picture elements to form -erm- a picture dB > > [] > > [quoted text clipped - 14 lines] > The pixels have no 'size'. Or rather their apparent size varies > depending on the display medium. [] > David, > The pixels have no 'size'. Or rather their apparent size varies > depending on the display medium. That sounds like you are trying to have it both ways! At the same time, pixels do, and do not, have size! <G> David > [] > > The term "pixel size" has no meaning. Pixels in a file has absolutely [quoted text clipped - 9 lines] > Cheers, > David "Real-world pixel" seems to me to be something of an oxymoron and much of this discussion is has turned upon mere semantics, perhaps because there's no precise, generally-accepted terminology for describing individual photosites in a CCD or CMOS array. Let's call the object associated with each independent photo site on the imager a "photel", for "PHOTo ELement." (And, yes, I just made the term up, so feel free to change it if you think it's as ugly as I do.) A pixel, then, would be an abstract mathematical entity of zero spatial extent comprising an RGB triplet instantaneously measured at a point in space. A photel, on the other and, would have the geometrical dimensions of the individual photosite as its spatial extent and would also have a datum associated with it, the latter consisting of an average/integration of pixel values taken across the photel's spatial extent over the exposure time. A photel would then be described by five elements: 1)-3) its coordinates in space, 4) a mathematical function or mapping describing its spatial extent and 5) a datum indicating its time- and space-averaged color value. I think this seems a more reasonable starting point for discussions of "pixel dimensions" Well, it all sounds pretty good, anyway. :-) [] > "Real-world pixel" seems to me to be something of an oxymoron and much of > this discussion is has turned upon mere semantics, perhaps because there's [quoted text clipped - 19 lines] > > Well, it all sounds pretty good, anyway. :-) Maybe!! Who said that a pixel/photel had to have a colour value? Unfortunately, as Hoover =<> vacuum cleaner, photel <=> pixel Good marks for trying, though!! Cheers, David > "Paul H." <xxpaulhtck@zzcomcast.yycom> wrote in message <snip> > > Well, it all sounds pretty good, anyway. :-) > > Maybe!! > > Who said that a pixel/photel had to have a colour value? Who said a pixel had to have a luminance value? Or a position? Or even have physical reality? Let's make the defintion of pixel so general that anything from elephants to rings over manifolds can be called pixels! Then we can really have one of those legendaryheat-without-light usenet discussions where everybody is right and everybody is wrong. You know you've been on usenet too long when "I Am the Walrus" starts making sense. I am the Eggman. We are the Eggmen. I am the Pixel. Goo-goo-ka-joob. > Unfortunately, as Hoover =<> vacuum cleaner, photel <=> pixel There's a Freudian slip: Just like your Hoover, this thread has indeed started to suck. :-) > You know you've been on usenet too long when "I Am the Walrus" starts making > sense. I am the Eggman. We are the Eggmen. I am the Pixel. > Goo-goo-ka-joob. That song has always made sense if you dropped enough acid. it holds provided the picture has no size too of course, as soon as the picture has size then a pixel will have size too a pixel can have size even if a picture is not displayed another question seems to be: does the pixel know its place in the big picture? the 'el' part suggets not, convention or common understanding suggests yes interesting ... > > [] > > > The term "pixel size" has no meaning. Pixels in a file has absolutely [quoted text clipped - 33 lines] > > Well, it all sounds pretty good, anyway. :-) [] > another question seems to be: does the pixel know its place in the big > picture? > > the 'el' part suggets not, convention or common understanding suggests yes > > interesting ... It suggests that "ordered array" should be in there somewhere. "A picture comprises an ordered array of pixels" perhaps? Cheers, David >> [] >> > The term "pixel size" has no meaning. Pixels in a file has absolutely [quoted text clipped - 33 lines] > >Well, it all sounds pretty good, anyway. :-) thought this was pretty obvious for anyone working in this field :-) > > "David J Taylor" > > <david-taylor@blueyonder.co.not-this-bit.nor-this.uk> [quoted text clipped - 7 lines] > It is sometimes not stated what the size is, but and real-world image > stored in a file does have a size. True, many image formats have a dpi field or whatever which tells you the size. But mostly this is meaningless, I mean (as Roland says) how big is an image form a X-megapixel camera? It really does depend entirely on how you display it. Also don't forget that although a monitor is 96dpi and a printer is 300dpi, a monitor can display any one of 16 million colours for each dot, whereas (I believe) a normal "household" printer can only display 8 different colours for each dot. The printer needs to use clusters of dots to give the impression of many more colours. Hence if you have a 300dpi printer you can print an image at 150dpi (or even 75dpi) and it not look any worse quality to printing it at 300dpi. > > A pixel as a concept has no size. > > .. although a zero-sized pixel could capture no information... Why not? A pixel doesn't have physical size. If I had my JPEG image that was 800x600 pixels but I set the dpi fields to zero, then the pixels would have zero size but there would still be an image. > > The pixels in my graphic card has no size. > > .. an interesting idea. Can the pixel exist without being displayed? > <G> Of course it can. At the moment I am working on some image processing software that captures images from a webcam in real time and outputs just a few numbers. At no point are the pixels displayed (apart from for testing purposes) but the pixels certainly exist and are fundamental to the operation of the software. [] > True, many image formats have a dpi field or whatever which tells you the > size. But mostly this is meaningless, I mean (as Roland says) how big is an > image form a X-megapixel camera? It really does depend entirely on how you > display it. The pixel from a camera will have a precise dimension, depending on the taking lens and the sensor. It will have a precise angular size. > Also don't forget that although a monitor is 96dpi and a printer is 300dpi, > a monitor can display any one of 16 million colours for each dot, whereas (I [quoted text clipped - 3 lines] > print an image at 150dpi (or even 75dpi) and it not look any worse quality > to printing it at 300dpi. This is one of the major confusion points for beginners. Those image processing programs which display image size in inches or millimetres ..... don't get me started on that one! > > > A pixel as a concept has no size. > > [quoted text clipped - 3 lines] > was 800x600 pixels but I set the dpi fields to zero, then the pixels would > have zero size but there would still be an image. If a pixel is of zero physical size, it could not capture any photons. Your JPEG file would be lying, which many do, of course. > > > The pixels in my graphic card has no size. > > [quoted text clipped - 6 lines] > purposes) but the pixels certainly exist and are fundamental to the > operation of the software. There was a <G> there! Of course, your pixel is not actually a pixel when it is stored, it is just a few numbers......a mathematical representation of a pixel. For example, is a real pixel quantised in levels, as the output from your WebCam is? Cheers, David > The pixel from a camera will have a precise dimension, depending on > the taking lens and the sensor. It will have a precise angular size. Hmmm ... I can agree that a pixel corresponds to certain "angular size" - yes. But "angular size" is commonly not called a size, it is called just angle. A pixel in a picture corresponds to a certain angle. > If I had my JPEG image > that [quoted text clipped - 4 lines] > If a pixel is of zero physical size, it could not capture any photons. > Your JPEG file would be lying, which many do, of course. The JPEG file is not about photons, pictures might e.g. be rendered in software. It is perfectly valid to set the dpi size to 0. Then the entire picture should (by a faithful printing software) be printed to one dot. A good software would of course warn you before doing it. > There was a <G> there! Of course, your pixel is not actually a pixel > when it is stored, it is just a few numbers......a mathematical > representation of a pixel. For example, is a real pixel quantised in > levels, as the output from your WebCam is? The pixels stored as numbers are also pixels (i.e. picture elements). Graphic file formats (such as TIF and JPEG) store pictures and pictures consists of pixels. You can store other things in your computer, e.g. text. This text is still text and consists of words and letters even though it is binary coded. The binary representation is not less text than ink on a paper. /Roland > The JPEG file is not about photons, pictures might e.g. be rendered > in software. It is perfectly valid to set the dpi size to 0. Then > the entire picture should (by a faithful printing software) be printed > to one dot. A good software would of course warn you before doing it. I have to pick you up on that one :-) If the resolution was 0 dpi, the image would not be one dot - it would be infinitely large. Image size = pixels / dpi I do, however, agree with you that talking about the size of a pixel in a file is meaningless. Pixels are used, in conjunction with a display medium of a certain resolution, to give a size. They have no size until a display medium is introduced. Of course, this is because I (and it seems many others) take the word 'pixel' to mean an abstract piece of information. David uses the word to describe an element of a display. Different definitions of the same word, as is the cause of so many disagreements... [] > Of course, this is because I (and it seems many others) take > the word 'pixel' to mean an abstract piece of information. > David uses the word to describe an element of a display. > Different definitions of the same word, as is the cause of so > many disagreements... I tried to avoid the ambiguity by referring to "real-world pixel", i.e. an abstract pixel but one related to a real-world image. The pixel might have been taken by a camera, existed on the surface of a CCD or display as a light pattern, or on a printer as an ink pattern. Cheers, David SNIP > I tried to avoid the ambiguity by referring to "real-world pixel", i.e. an > abstract pixel but one related to a real-world image. The pixel might > have been taken by a camera, existed on the surface of a CCD or display as > a light pattern, or on a printer as an ink pattern. I've followed the discussion (with some amusement), but it all is rather simple if you stick to practical descriptions and discriminate between: 1. a photosensitive sensor element (sensel) 2. pixel or sample (dimensionless data, e.g. representing luminance or color) 3. physical or output pixel (data output to a physical medium) 4. printer dots (used to dither intermediate ink colors for each output pixel) The ISO hardly ever talks about pixels when describing camera sensor arrays, they do refer to photosensitive sensor elements or samples (with a sample pitch and area which determine its quality). They consider the physical characteristics of the sensel (e.g. spectral sensitivity), but the pixel/sample is just integrated (over time and in area) data, until output to a physical medium. Bart [] > I've followed the discussion (with some amusement), but it all is rather > simple if you stick to practical descriptions and discriminate between: [quoted text clipped - 4 lines] > 4. printer dots (used to dither intermediate ink colors for each output > pixel) [] > Bart Actually, for my work, that isn't quite enough, because some of my pixels do actually represent true ground measurements - in the sense that interpixel spacing is well defined by the inter-line spacing of a satellite scan. The vertical size (approximately north-south) of the pixel is constant, but the horizontal size is not because the satellite is much close to the earth than the size of the earth. In angular terms, the pixel is well described and a constant size. The term sensel is not used at all. Yes, clarity in what you are describing is essential. Cheers, David SNIP > Actually, for my work, that isn't quite enough, because some of my pixels > do actually represent true ground measurements - in the sense that > interpixel spacing is well defined by the inter-line spacing of a > satellite scan. a.k.a. sample pitch? Bart > SNIP > > Actually, for my work, that isn't quite enough, because some of my pixels [quoted text clipped - 5 lines] > > Bart You could say that - north-south it's 1.09km, east-west a step of 0.95 milliradians. Cheers, David > You could say that - north-south it's 1.09km, east-west a step of 0.95 > milliradians. So - David - you are meassuring the pixel size just as I described earlier. I gave you two options - either degrees or size mapped to the object you are photographing - and you use both! /Roland > > You could say that - north-south it's 1.09km, east-west a step of 0.95 > > milliradians. [quoted text clipped - 4 lines] > > /Roland Well, here I'm actually talking about the sampling interval, which in the X direction is determined by a uniform sampling of a rotating mirror, and in the Y direction by the motion of a spacecraft. Hence it is a linear angle measurement in X and a linear distance measurement in Y. You can convert knowing the object distance, but I prefer to quote the fundamental sampling interval relating it to the physical cause of the sampling. Each pixel will also have an angular subtense determined by the dimensions of the detector and the focal length of the optics. Knowing the spacecraft to ground distance (approximately 833km) you can also compute the size of the pixel on the ground. It is about 1.1km square, but varies in width and height across the scan. Cheers, David > Well, here I'm actually talking about the sampling interval, which in > the X direction is determined by a uniform sampling of a rotating [quoted text clipped - 9 lines] > can also compute the size of the pixel on the ground. It is about > 1.1km square, but varies in width and height across the scan. Nice! Just to get back to our old discussion. In this case the "pixels" could be said to be approx 1.1 km square. But, this is only a mapping. If you look close at the ground, you find no 1.1 km large pixels. So - there are no pixels on the ground - they only exist in your image file. In this setup you have (with a scanning device), there are no pixel anywhere, except in the file. /Roland [] > Nice! > [quoted text clipped - 6 lines] > > /Roland The video stream exists as an analogue voltage from the detector and amplifiers. At the point where it is sampled, the pixel is defined as an analogue voltage which is then digitised. You could see that pixel with an oscilloscope. The angular size of the pixel is defined by then, and is required as meta-data. Further meta-data is required to define the pixel spacing. The pixels exist in many places, one of which is a data file. By the way, that same file (i.e. binary data downloaded from the spacecraft) also contains a whole lot of calibration information about the pixels it contains. Cheers, David > [] > > Nice! [quoted text clipped - 12 lines] > analogue voltage which is then digitised. You could see that pixel with > an oscilloscope. ROTFLMAO mate .. when I will tell this in the Ultrastructure dpt here they're probably going to faint .. laughing that is .. [] > ROTFLMAO mate .. when I will tell this in the Ultrastructure dpt here > they're probably going to faint .. laughing that is .. Laugh they might well do, but that pixel in your file, whether a greyscale value or an RGB triplet, originated from somewhere in the analog world (we are talking real-world pixels) and at the point you can see the value with an oscilloscope, meter etc. Indeed, if you look at the specification sheets for digital camera sensors the outputs are specified in voltage or current terms..... Cheers, David and, of course, there are sensels which work with wavelengths we can see and thos that work with wavelengths we cannot see <i wish I hadn't typed that> > SNIP > > I tried to avoid the ambiguity by referring to "real-world pixel", i.e. an [quoted text clipped - 19 lines] > > Bart > [] > > True, many image formats have a dpi field or whatever which tells [quoted text clipped - 4 lines] > The pixel from a camera will have a precise dimension, depending on > the taking lens and the sensor. It will have a precise angular size. Yes you are correct, but how that maps to a linear dimension (eg on a monitor or printer) is completely different. There is no linear information stored with pixels from a camera. > > Also don't forget that although a monitor is 96dpi and a printer is > > 300dpi, a monitor can display any one of 16 million colours for [quoted text clipped - 20 lines] > If a pixel is of zero physical size, it could not capture any photons. > Your JPEG file would be lying, which many do, of course. But who says my pixel needs to capture photons? I can make a JPEG (or bitmap) file from any source and set the pixels to any colour value and then set the size to zero. In fact many image formats don't have any information about the physical size of the pixels. > > > > The pixels in my graphic card has no size. > > > [quoted text clipped - 11 lines] > when it is stored, it is just a few numbers......a mathematical > representation of a pixel. Yes you're right, my pixel is actually three transistors, three capacitors and a bit of liquid crystal wedged between electrodes, but in software terms a pixel refers to an element of a picture, the smallest representable element of a picture to be precise. They don't have to be displayed, I can have 640x480 pixels stored in an array, but I don't have to show them for them to be pixels. > For example, is a real pixel quantised in > levels, as the output from your WebCam is? Yes, a real physical pixel is always quantised. Your printer or monitor can only display a set different number of "levels". Exactly the same way a pixel from your camera can only be a set number of levels. good point Scott We don't need to spread the pixels in XY space We can stack them and shuffle them just like cards <o> <o> dB > > [] > > > True, many image formats have a dpi field or whatever which tells [quoted text clipped - 68 lines] > only display a set different number of "levels". Exactly the same way a > pixel from your camera can only be a set number of levels. [] > > For example, is a real pixel quantised in > > levels, as the output from your WebCam is? > > Yes, a real physical pixel is always quantised. Your printer or monitor can > only display a set different number of "levels". Exactly the same way a > pixel from your camera can only be a set number of levels. Well, only in the sense that a physical pixel will be quantised because of quantum-mechanical effects. The quantisation because of conversion from the analog to the digital domain is something else.... David It IMHO is really about mindsets eg: take sensel data and process it so every separate picture element has 3 references the first is x-y position data the third is3-color representation (hmm a recursive triplet!) anyway, take a big bag of colored jelly beans if the product of the x-y is prime place a red jelly been at the x-y position if the x-y product is divisible by 2 place a purple jellybean at the x-y position make up lots more rules like this unti the X-Y space is filled (note that the 'white' jelly bean is really a light murky grey) There, you have an image based on sensor data - it may not look much like the original image but it may taste better. That, in a nutshell, is an example of what happens rebdering data to a visible picture :-) das B > > "David J Taylor" <david-taylor@blueyonder.co.not-this-bit.nor-this.uk> > wrote in news:OciEc.6076$zZ5.66173165@news-text.cableinet.net: [quoted text clipped - 34 lines] > > David > Can one print at actual pixels size (100 %)? How is this done? The way I do it is: Adjust the image so it looks the way I want it (Photo Shop or Elements) Save it in TIFF on the Desktop (on the machine connected to the printer) (Epson 750) Open it in PSP on the printer machine Tell PSP to print it Use PSP's screens to Choose paper and output mode Fit the image to the paper Push Print PSP chews on it for a while, sends it to the printer (Quite a bit later) It comes out looking just like it did on the monitor Lucky (simple) me. Frank ess "Frank ess" <frank@fshe2fs.com> wrote in news:YgiEc.19884$Fy.4790 @twister.socal.rr.com: > The way I do it is: > Adjust the image so it looks the way I want it (Photo Shop or Elements) [quoted text clipped - 9 lines] > (Quite a bit later) > It comes out looking just like it did on the monitor No it don't. Why should it? > Lucky (simple) me. Lucky at least - if you got it right :) /Roland > "Frank ess" <frank@fshe2fs.com> wrote in news:YgiEc.19884$Fy.4790 > @twister.socal.rr.com: [quoted text clipped - 18 lines] > > Lucky at least - if you got it right :) (Funny, you don't *look* trollish) Sure, it *do*. Same colors, contrast, etc., same shapes, same size within 10% (8x10); in *the common picture-viewing experience*, the only considerable difference between looking at one and the other is... one is more difficult to hold at arm's length. (Lucky) Frank ess "Frank ess" <frank@fshe2fs.com> wrote in news:oNiEc.20161$Fy.15909 @twister.socal.rr.com: > Sure, it *do*. > [quoted text clipped - 3 lines] > > one is more difficult to hold at arm's length. You left out that your monitor is the same size as your paper. But the one that is harder to hold at arms lenght is much easier to have standing at arms length. /Roland > It comes out looking just like it did on the monitor Well, unless you have a /really/ good monitor, or a /really/ useless printer - the print should look much better than the screen (higher res., larger gamut). (And I think the OP was trolling.) Signature- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ] ======================================================================== «To live outside the law, you must be honest.» (Bob Dylan) > Can one print at actual pixels size (100 %)? How is this done? Set up the print so it's sent to the printer at 72dpi, that's what most monitors are. It'll look pretty lousy, though. Mark For every pixel output to a monitor assign it a particular colored jelly bean I thinh the comment "a pixel has no size" can be rephrased as "a pixel has no standard universal size" onviously pixels do have size, shape and form otherwise there would be no pix Remember the el bit is important too Equally, a pixel is not a centrimetre but is does have size :-) they usually have color lol :-) dB > > Can one print at actual pixels size (100 %)? How is this done? > > Set up the print so it's sent to the printer at 72dpi, that's what most > monitors are. It'll look pretty lousy, though. > > Mark > Can one print at actual pixels size (100 %)? How is this done? Frankly, I prefer printing only one blown-up pixel per page, then arranging the pages in order somewhere in the Nevada desert. Plenty of room and no one's the wiser, except for a few scorpions. Seriously, if you've got a row of X pixels printed at Y dots-per-inch, then you'd need a print that's X/Y inches wide. The rest is left as an exercise for the student. > Frankly, I prefer printing only one blown-up pixel per page, then > arranging the pages in order somewhere in the Nevada desert. Plenty > of room and no one's the wiser, except for a few scorpions. Nice project! Let us assume that each paper is 10 by 10 cm. Then a 6 Mpixel picture would be 300 x 200 meters and consist of 6 million papers. Now - that would be something to see! You need an helicopter or high building or something though. /Roland > > Frankly, I prefer printing only one blown-up pixel per page, then > > arranging the pages in order somewhere in the Nevada desert. Plenty [quoted text clipped - 7 lines] > > You need an helicopter or high building or something though. Or just wait for some satellite to pass over... > > > Frankly, I prefer printing only one blown-up pixel per page, then > > > arranging the pages in order somewhere in the Nevada desert. Plenty [quoted text clipped - 9 lines] > > Or just wait for some satellite to pass over... Ah, but if you took a picture of the area you were going to use to display the photo and had really good color matching, the guys in space wouldn't even know they were looking at a digital picture after you laid it out! Well, the lighting might look funny during certain parts of the day. :-) Find out what the format (chip) size of your camera is, resize photos to that size, and set number of pixels to equal number of pixels in chip. Most printer drivers will attempt to print as best they can. Image will not be that great unless you have a REALLY high res printer. > Can one print at actual pixels size (100 %)? How is this done? > -- > > nobody SignatureDon Stauffer in Minnesota stauffer@usfamily.net webpage- http://www.usfamily.net/web/stauffer > Find out what the format (chip) size of your camera is, resize photos to > that size, and set number of pixels to equal number of pixels in chip. > Most printer drivers will attempt to print as best they can. Image will > not be that great unless you have a REALLY high res printer. Like the ones used to "print" the ICs in the first place! Cheers, David > > Find out what the format (chip) size of your camera is, resize > > photos to that size, and set number of pixels to equal number of [quoted text clipped - 3 lines] > > Like the ones used to "print" the ICs in the first place! Hehe, don't they "print" them at (say) 100x and then use optics to create the masks at the correct size? > Can one print at actual pixels size (100 %)? How is this done? Simple - use the printer software. Which pixel do you want to print? Can one print at actual pixels size (100 %)? How is this done? Signaturenobody ------------------------------------------------------------ ------------------------------------------------------------ Sure. As long as the printer machine has resolution capacity equal to or finer than the image resolution I don't see why not. mark_ > Can one print at actual pixels size (100 %)? How is this done? I've read your later post where you clearified that this was about PS and that you had an image on the screen displayed at 100% (i.e. at screen resolution) and that you wanted to print this image at screen resolution - so I'll leave the ongoing phenonomelogical debate about the true nature of the pixel, and move on to how to print at Actual Pixel size in PS. First, find out the resolution of your monitor: Then in PS 7.0, load the image and: Image->Image Size Untick: Resample Image (Important!) Set: Resolution = the resolution of your monitor. Click: OK File->Print with Preview Untick "Scale to Fit Media" Make sure scale is 100% Click: Print I am not sure why you want to do this tho'. By printing at monitor resolution (typical 72 ppi), you'll lose a lot of quality compared to printing at a resolution closer to the resolution of your printer (a whole lot more than 72 ppi if it is a photoprinter). Your printer will upsample the image - so the final size will match what you see at the monitor - but the quality will suffer. Signature- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ] ======================================================================== «To live outside the law, you must be honest.» (Bob Dylan) Thank you, that's all I wanted to know. In other words, that the image on the monitor, acceptable as it is, will not be matched in quality by the print at that huge size, but reduced ppi below the acceptable level of 266 or so. n article <q5smcbhmts.fsf@nelja.ifi.uio.no>, Gisle Hannemyr <gisle+njus@ifi.uio.no> writes >> Can one print at actual pixels size (100 %)? How is this done? > [quoted text clipped - 24 lines] >will upsample the image - so the final size will match what you see >at the monitor - but the quality will suffer. Signaturenobody |
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
|
©2010 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.