Ugly Hedgehog® - Photography Forum
How do I get "clear" images when printing a 16x24 of a hi-res .jpg
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Page: <<prev 1 2 3 4 5 next>>
Dec 22, 2013 09:24:10   #
wowbmw
 
mikegreenwald wrote:
Most of the factors mentioned above are correct. RAW is always the way to go when manipulation is planned; it tolerates PP better than any other format.
The was a plug-in called Genuine Fractals, later Adobe called it Perfect Resize, that did an exceptional job of upsizing photos with excellent retention of detail etc.. It seems to be missing from Photoshop CC, though I believe it was still present on Photoshop CS5. It is probably still available somewhere.


The application is not part of Adobe PS but rather OnOne software now in version 8.0. Works well.

| Reply
Dec 22, 2013 09:40:12   #
jr168
 
To get sharp prints you must first have a sharp image. Since you are using continuous lighting which will not freeze action, you have to have a higher shutter speed to reduce movement. Using the one to one ratio of focal length to shutter speed (50mm to 1/50 sec)or using image stabilization will help reduce camera shake. To stop movement of the pets, 1/250 or higher would be required.

| Reply
Dec 22, 2013 09:40:29   #
Morning Star
 
Dave_TX wrote:
If the printer operates at 300dpi and you supplied the JPEG image at 288dpi it will have to be converted yet again. JPEG compression is "lossy". . Whatever you do, avoid running an image through the JPEG conversion process multiple times.


dpi has nothing to do with the image resolution and everything with the number of drops of ink your printer lays on the paper.
The resolution of images is expressed in ppi - pixels per inch.
300 ppi is generally accepted as a good resolution for most pictures.
If in fact your images of 16 x 24 inches were printed at 288 ppi, you wouldn't be able to tell the difference with the 300 ppi resolution, especially when for that size of image, you wouldn't be holding it close to your face like you might do with a 3 x 5 or 4 x 6 picture.

No converting is necessary. Even a photo at 72 pixels per inch will print well, as long as the overall dimensions are sufficient.
Lets go back to the 16 x 24 image for a moment. At 288 ppi, the overal dimensions are then 4608 x 6912 pixels.
Keep those last two numbers and make the resolution 72 ppi,
the computer will tell you the actual measurement is now 64 x 96 inches. Take both those files to the printer and tell them you want them printed at 8 x 12 inches (1/4 of the 16 x 24). They won't blink an eye and just print them.
Now ask them to print at 64 x 96 inches - still keeping the same dimension of 4608 x 6912 pixels - they will tell you the resolution isn't high enough for that.

When it comes to dpi - when I was using an inkjet, I noticed the default was 600 dpi: 600 drops of ink per inch. However, I never worried about that, as I had my photos printed at a photo lab anyway.

beauxPatrick, what was the resolution of your images straight out of the camera? The best thing is to always shoot in the highest resolution you camera offers.

| Reply
Dec 22, 2013 09:43:16   #
f8lee
 
Wow, a lot of misinformation here…

First, beaux - you do not mention what lens you used (or camera, for that matter) - this is an important factor, as is the aperture used. The lack of sharpness you perceive could be due to camera shake (at 1/60th second quite possible) and/or the depth of field at whatever f-stop you used not being enough to cover the zone you wanted in focus. For the former issue, a tripod could help. For the latter, s smaller aperture (which in turn might require more lighting).

Now the JPEG lossy compression thing has little to do with actual sharpness - the "loss" involved has to do with bit depth which relates to color at a given pixel. "Lossy" does not refer to the actual number of pixels, unless you compress further to save yet more space on the drive.

And know that the manner in which JPEG works, in an effort to save disk space, is such that if you were to open a JPEG image with ANY image editing program and do absolutely nothing to it but save it again (not "save as" but just re-save) the the compression algorithm will actually change the image file a little bit and a little more data would be lost in the process. And this is regardless of how high you set that slider. Compressing an image further would just make for additional data loss. In the analog world, this would be somewhat analogous (get it?) to making a copy of a copy.

The reason RAW files allow for more image manipulation is due to that bit depth. JPEG is an 8 bit format, while your camera may record 10, 12 or 14 bits worth of light at each photo site (remember that each one of your 11megapixels the CCD is merely recording the intensity of light as filtered through the Bayer filter at that site).

So what does THAT mean? Well, bit depth simply means how many variants can be recorded. A bit depth of 2 (on or off, black or white) would yield a purely black and white image with zero tonality because each pixel could only be black or white. JPEGs with 8 bits of depth for each color, allow for 2^8, or 256, possible shades for each of R, G and B. 12 bits of depth (it's all exponential) yields 4096 potential shades at a given pixel. So doing things like bringing up shadow detail or possible pulling back portions that were overexposed ("blown out" ) is more possible with the RAW file, since it still has the fully detailed set of image data as recorded by the imaging chip.

Now, on a separate note, printer drivers (the invisible program that actually commands the printer) will rejigger resolution as required by the printer itself (which as programmed by the manufacturer). That is, there is a native resolution for digital printers, often around 300DPI. However, that's not a god-given standard - Kodak dye-sub printers are natively 301DPI, and Epson ink jets are natively 720DPI (for the small models) and 360DPI (for the large ones) - at least they were the last time I looked. But the entire idea of the driver software is to take care of resolution mismatches, same as screen driver software does for the image you see on screen (screens can be 72PPI up to 200-300+PPI on a Retina display, depending on if it's a Macbook or iPad).

And RAW files cannot be viewed or printed directly (in this they are not quite like negatives) - they MUST be interpreted by some software to even display as an image on screen or on paper. So printing a "RAW" file is actually nonsense. When you shoot RAW and 'chimp' the shots on the LCD on the back of your camera, you are in fact looking at the JPEG image embedded in that file as interpreted by the camera's software. Think of a RAW file as the binary code that makes up an e-book - a gaggle of zeros and ones but not readable by a human as words until it is converted by other software.

Mr Greenwald is mistaken about the fate of Genuine Fractals - it was not in purchased by Adobe but rather by On One software, who now offers it (under the name Perfect Resize) as part of their suite. And it is quite good at what it does - I've used it to make poster sized prints from a D200 (10MP) which I've sold. But it's again dependent on your own camera's MP count to see if it's needed to start with.

Hope that makes sense.

| Reply
Dec 22, 2013 09:52:52   #
WNC Ralf
 
You were in av, for moving subjects you need to open up, or increase your ISO to get shutter speeds up.

| Reply
Dec 22, 2013 10:34:19   #
BobHartung (a regular here)
 
beauxPatrick wrote:
I just completed a shoot and they were Pet Portraits... shot under "cool" studio lighting (5000K) with out strobe. Av at 16 with shutter speed as low as 1/60 when zoomed... the slightest movement and I got a blur...

There is a reason for the "cool" lighting and no strobes... I am a Pet Friendly photographer and don't do anything to startle or upset the animals... 99% of my shots are good... a few did not turn out as well as I would have liked...

I had to have them printed at 16x24(20) and I saved each image at 16x24 at 288 dpi... when printed, I am disappointed in the quality of the enlargement... close up, they seem too soft... at a distance they look great... did I screw up by changing the dpi... even though I used a multiple of 72?

What should I have done...

beaux
I just completed a shoot and they were Pet Portrai... (show quote)


Did you perform any edge or "smart" sharpening after all other corrections were performed? Digital cameras impart a certain unsharpness by the Bayer algorithm used to distribute the R,G, & B receptors on the sensor. I have been taught to add a little (3-4% in Photoshop) sharpening after all other corrections have been performed, before printing. Others may differ.

HTH

| Reply
Dec 22, 2013 11:26:33   #
bunuweld
 
beauxPatrick wrote:
I just completed a shoot and they were Pet Portraits... shot under "cool" studio lighting (5000K) with out strobe. Av at 16 with shutter speed as low as 1/60 when zoomed... the slightest movement and I got a blur...

There is a reason for the "cool" lighting and no strobes... I am a Pet Friendly photographer and don't do anything to startle or upset the animals... 99% of my shots are good... a few did not turn out as well as I would have liked...

I had to have them printed at 16x24(20) and I saved each image at 16x24 at 288 dpi... when printed, I am disappointed in the quality of the enlargement... close up, they seem too soft... at a distance they look great... did I screw up by changing the dpi... even though I used a multiple of 72?

What should I have done...

beaux
I just completed a shoot and they were Pet Portrai... (show quote)


The pictures are adorable as they are, but if downloaded your original, they certainly are at a resolution borderline to poor for the enlargement you want. Using programs such as onOne software may help you to display them in a larger size. I am puzzled why you chose such a harsh overhead light. I hope you don't mind if I play with the dodge tool in Adobe to brighten the very dark areas. Obviously that's not as good an approach as setting up the lights in the original scene, but if you intended the harsher contrast your original is very attractive



| Reply
Dec 22, 2013 12:15:58   #
flyguy
 
Dave_TX wrote:
If the printer operates at 300dpi and you supplied the JPEG image at 288dpi it will have to be converted yet again. JPEG compression is "lossy". That means that every time you open a JPEG image, do something with it, and then reconvert to JPEG you degrade the quality of the image. When you PP JPEG images you should always PP a copy of the original. I always lock my JPEG originals so that I won't inadvertently write over my original with a processed version. Merely rotating the JPEG image so that it is easier to view in the folder degrades the image. You can convert your original JPEG to uncompressed TIFF, do all of your processing, and then when you are ready to
send the image to the printer convert it back to 300dpi JPEG.

Whatever you do, avoid running an image through the JPEG conversion process multiple times.
If the printer operates at 300dpi and you supplied... (show quote)


Excellent advice.

| Reply
Dec 22, 2013 12:43:03   #
mikegreenwald
 
f8lee wrote:
Wow, a lot of misinformation here…

First, beaux - you do not mention what lens you used (or camera, for that matter) - this is an important factor, as is the aperture used. The lack of sharpness you perceive could be due to camera shake (at 1/60th second quite possible) and/or the depth of field at whatever f-stop you used not being enough to cover the zone you wanted in focus. For the former issue, a tripod could help. For the latter, s smaller aperture (which in turn might require more lighting).

Now the JPEG lossy compression thing has little to do with actual sharpness - the "loss" involved has to do with bit depth which relates to color at a given pixel. "Lossy" does not refer to the actual number of pixels, unless you compress further to save yet more space on the drive.

And know that the manner in which JPEG works, in an effort to save disk space, is such that if you were to open a JPEG image with ANY image editing program and do absolutely nothing to it but save it again (not "save as" but just re-save) the the compression algorithm will actually change the image file a little bit and a little more data would be lost in the process. And this is regardless of how high you set that slider. Compressing an image further would just make for additional data loss. In the analog world, this would be somewhat analogous (get it?) to making a copy of a copy.

The reason RAW files allow for more image manipulation is due to that bit depth. JPEG is an 8 bit format, while your camera may record 10, 12 or 14 bits worth of light at each photo site (remember that each one of your 11megapixels the CCD is merely recording the intensity of light as filtered through the Bayer filter at that site).

So what does THAT mean? Well, bit depth simply means how many variants can be recorded. A bit depth of 2 (on or off, black or white) would yield a purely black and white image with zero tonality because each pixel could only be black or white. JPEGs with 8 bits of depth for each color, allow for 2^8, or 256, possible shades for each of R, G and B. 12 bits of depth (it's all exponential) yields 4096 potential shades at a given pixel. So doing things like bringing up shadow detail or possible pulling back portions that were overexposed ("blown out" ) is more possible with the RAW file, since it still has the fully detailed set of image data as recorded by the imaging chip.

Now, on a separate note, printer drivers (the invisible program that actually commands the printer) will rejigger resolution as required by the printer itself (which as programmed by the manufacturer). That is, there is a native resolution for digital printers, often around 300DPI. However, that's not a god-given standard - Kodak dye-sub printers are natively 301DPI, and Epson ink jets are natively 720DPI (for the small models) and 360DPI (for the large ones) - at least they were the last time I looked. But the entire idea of the driver software is to take care of resolution mismatches, same as screen driver software does for the image you see on screen (screens can be 72PPI up to 200-300+PPI on a Retina display, depending on if it's a Macbook or iPad).

And RAW files cannot be viewed or printed directly (in this they are not quite like negatives) - they MUST be interpreted by some software to even display as an image on screen or on paper. So printing a "RAW" file is actually nonsense. When you shoot RAW and 'chimp' the shots on the LCD on the back of your camera, you are in fact looking at the JPEG image embedded in that file as interpreted by the camera's software. Think of a RAW file as the binary code that makes up an e-book - a gaggle of zeros and ones but not readable by a human as words until it is converted by other software.

Mr Greenwald is mistaken about the fate of Genuine Fractals - it was not in purchased by Adobe but rather by On One software, who now offers it (under the name Perfect Resize) as part of their suite. And it is quite good at what it does - I've used it to make poster sized prints from a D200 (10MP) which I've sold. But it's again dependent on your own camera's MP count to see if it's needed to start with.

Hope that makes sense.
Wow, a lot of misinformation here… br br First, b... (show quote)


Thank you for a very informative response!

| Reply
Dec 22, 2013 13:05:55   #
Kuzano
 
mikegreenwald wrote:
Thank you for a very informative response!


I agree with the comment about "misinformation" in this post, but not surprised.

First, the comment that since you fell short of 300DPI (capability of the printer????), is incorrect. Depending on the capability of the printer, I get excellent prints using as low as 225DPI, and do often print at 250 to 275 DPI. Beyond a certain point, extra DPI is simply wasted ink that takes longer to dry or absorb into the surface of the paper. In fact, most printers in home situations will print to 360DPI, but ink is surely wasted.

Test your printer on various print settings of DPI from 225 to 300. Optimally 300 is usually the best, but may be wasting ink, where 275 DPI may be excellent as well, with a savings on ink. You did NOT mess up your print by falling 12 DPI short of someone's optimal.

Secondly, I checked the post rather thoroughly, and see NO mention of the amount of pixels in the image you printed from. 72DPI will print like crap normally, if you throw it out to the printer at that res. That is for monitor viewing, and web site use only.

HOW many megapixels were in your image, and remember that if you crop you lose megapixels.

The math for printing is complex. If there are not enough pixels to really create a 16X20 print (you will need 28.8 megapixels in the image to print best sharpness at 300DPI), you compromise the print of the image. This is why I often print at 250DPI (at which point you would need only 20 megapixels to reach the dimension at 250 DPI)

Since I don't know what size pixelwise your original image was after any cropping, I will surmise you did not have 20 to 28 megapixels.

In that case, your print function of your editing program if you force the issue by changing the print dimensions x DPI, your editing program will be forced to interpolate additional pixels to reach those dimensions and DPI.

Now regular editing software is just not up to this task in a way that interpolates truly usable pixels by the printer.

It is best to use software that is designed to do this "upping of resolution" with a final result of good print quality. One such was mentioned in a prior post as a plug in to the edit program you may be using. There are also free standing programs that will provide quality UPRESSING, as it is called.

Without understanding the process of increasing the resolution of files to print quality prints, starting with too few pixels in the first place, it will be difficult to get a handle on why the quality of a printed image is so variable.

It's based on first, mathematically understanding when you go beyond the optimal print size due to the megapixels you have to work with. Beyond that, you increase the dimension of the print with an upward gain in resolution (more pixels) and a grasp of what DPI is best for your printer. That varies with the subject content. So proofing is one way of doing that.

NOTE: Pardon me, but I used 16x20 on my math of printing examples, instead of the OP 16X24. my mistake, but idea the same.

The best book I have run across on the math of printing is a book called "The Magic of Digital Printing" by Derek Doeffinger. The ISBN number is ISBN - 10: 1-57990-689-3, or ISBN - 13: 978 - 1 - 57990 - 689 - 4

Very good informative soft cover 8X11 w 160 pages
The Magic Of Digital Printing
The Magic Of Digital Printing...

| Reply
Dec 22, 2013 13:07:16   #
dpullum (a regular here)
 
f8lee wrote:Wow, a lot of misinformation here… Indeed f8lee you are, thank you, very informative .... and of course we must all learn to bracket. Quite possible in the case of these dogs, since lights are used, not flash. also the light should have been much lower as mentioned by others.

Unquestionably, as it has always been, even in wet days, photography is very technical. However we digi world people are very lucky few are allergic to 0s or 1s, but many are allergic to how to manipulate them. I printed the 17 pages I suggested and will digest them. (and the f8lee info)

In the mean time, I copied the thumbnail of the single dog and played with Topaz Adjust and Topaz Detail. I was able to bring forth definition of the burned head top without over sharpening or darkening the rest of the photo. Surprising how much resurrection that Topaz or On-one can do.

With regard to changing the "size" of the photo (upsize), Faststone.org has the free program "resize"

| Reply
Dec 22, 2013 13:25:19   #
pauleveritt
 
To ME, it seems like you have a depth of field issue. If of all you need to crank up your ISO as far as it will go "Native". My Nikon D90 goes up to 3200 native. Beyond that you risk getting digital noise. You then need to set your aperture to f8 or higher so that you get a greater depth of field. FINALLY, with all of this set, set your shutter speed to say 1/200 to avoid any motion blur. Depending on the lighting, you might be able to come down to 1/150. FINALLY, I would get a stuff animal with some long hair, a like a lion with a mane that was fairly LONG and shoot the stuffed animal straight on with the body set at an angle to give you a "deep" object to photograph. Once you get the hair in the lion's mane in sharp focus AND the hair in the lion's tail in sharp focus you should have your sweet spot for small dogs. Adjust accordingly for larger animals.

For the print, crank the resolution up as far as the PRINTER will go. If you have a 4800 DPI printer, why leave any of the resolution on the table?

| Reply
Dec 22, 2013 14:06:51   #
lbrandt79
 
Nonsense. I admit there are things you can do with Raw images to help salvage/improve some things, but if the shutter speed, aperture, focus is out of wack, it is out of wack in Raw too. If it ain't sharp and in focus, Raw is not going to help. If you get the exposure right in jpeg, it will be as good as Raw. We do not have enough info on his shots, would like to see his exif data on each.
Sorry, this was meant as a reply to Alfresco.

| Reply
Dec 22, 2013 14:12:29   #
dpullum (a regular here)
 
Kuzano suggested: The best book I have run across on the math of printing is a book called "The Magic of Digital Printing" by Derek Doeffinger.

Thanks, just bought it $4.29 used from
www.betterworldbooks.com
New is about $8
BWBooks pay the shipping.... great site.

| Reply
Dec 22, 2013 14:28:47   #
mikegreenwald
 
wowbmw wrote:
The application is not part of Adobe PS but rather OnOne software now in version 8.0. Works well.


Thank you. You're right - I must have added the plug-in to Photoshop myself.

| Reply
Page: <<prev 1 2 3 4 5 next>>
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
UglyHedgehog.com - Forum
Copyright 2011-2019 Ugly Hedgehog, Inc.