The image should be between 100 and 125 dpi at full size.

RGB (not CMYK)

BMP or TIF (JPG loses color info.

I know it is pretty confusing. Below are some links I'd like to suggest reading. They do a much better job of explaining it than me.

I hope you find this useful.

http://www.scantips.com/basics1e.html
http://www.wolfenet.com/~workshop/scan-print.html
http://www.nilbs.com/techbabl/dpi_ultim.htm
http://nilbs.com/techbabl/dpi_woes.htm
http://www.dtss.com.au/White%20Papers/General/Resolution.html
http://www.ulead.com/new/scanner.htm
http://lcweb.loc.gov/preserv/guide/guid_exp.html
http://www.academic.uidaho.edu/cti/scanning/

[ November 24, 2001: Message edited by: Glenn Taylor ]

PPI has to do with resolution and nothing to do with DPI.......

http://www.nilbs.com/techbabl/dpi_ultim.htm

As for JPG to TIF, once an image is in JPG, you've pretty much got all of the color information you're gonna get. TIF and BMP contain the most color info of any of the formats. You can step down from TIF/BMP to JPG lose color information, but you can't do the same in reverse and gain color information. The image must start with TIF/BMP in order to have the most color information available. That is why most of your high-end printing starts with scanning color slides.

Digital cameras are good for everyday printing, but you will never capture some of the vibrant colors and details that you see other professional digital print services produce.

[ November 24, 2001: Message edited by: Glenn Taylor ]

Thats because inkjet printers use a higher DPI printhead. The higher the DPI printhead count, the higher your DPI count should be for the file. For example, if the printhead is rated 600dpi, then it is reasonable to have the image at 175 to 200 DPI. However, if your printhead DPI is only 300, then any image DPI higher than 108 is just a waste of HD space.

The trick is to have the image DPI set according to how it is going to be printed. I just think its a waste of time (time = money) to have the file size larger than is necessary.

Just a hint, if any of you have seen some of the samples created for the Edge by Mike Jackson (which were available on Gerber's GA 6.x software CD's), guess what DPI he used.

[ November 24, 2001: Message edited by: Glenn Taylor ]

0: TIFF 2048x1536 uncompressed
1: TIFF 1600x1200 uncompressed
2: TIFF 1280x960 uncompressed
3: TIFF 1024x768 uncompressed
8: TIFF 640x480 uncompressed
3 or more: SHQ 2048x1536; 2.7:1 JPEG
10 or more: HQ 2048X1536; 8:1 JPEG
5 or more: SQ1 high 1600x1200 low compression
15 or more: SQ1 normal 1600x1200 std. compression
8 or more: SQ1 high 1280x960 low compression
24 or more: SQ1 normal 1280x960 std. compression
13 or more: SQ2 high 1024x768 low compression
38 or more: SQ2 normal 1024x768 std. compression
33 or more: SQ2 high 640x480 low compression
82 or more: SQ2 normal 640x480 std. compression

As in all things, the larger PPI number the better. A 2.5 megapixel is sufficent for an 8x10 photo.

I would start with ....

0: TIFF 2048x1536 uncompressed

Then bring the image into your image editor (Corel, Photoshop, etc....) enlarge the image to full size (the size you intend to print) and export it as a 110 DPI BMP image. Print a sample of it. Then save the same image in your editor as a 300 DPI BMP image and do a test print of it.

Note how much longer it takes to "rip" the 300dpi image than it does the 110dpi image and that there is little if any difference in the final printed image.

Try it. You'll see what I mean.

[ November 24, 2001: Message edited by: Glenn Taylor ]

Trust me.....you are over-analyzing it.

1) Just take a picture at as large a PPI as you can. This is to get as much "color information" as you can - nothing more.

2) Bring the image into Corel Draw. Scale it to the size you want to print.

3) Export the image as a BMP file at 110 DPI.

4) Bring the BMP image into Omega/GA Composer.

5) Prep the file for printing as you normally would and then print it.

That's all there is to it.

As for the difference between the print quality between your friend's inkjet printer and the Edge, it has to do with the difference in printing technologies.

Inkjet printers spray the ink kinda like miniture airbrushes. This ink is absorbed into the substrate and spreads out a little bit. This causes each dot to visually blend the colors better for the human eye.

Thermal resin printers (Edge, ColorCamm, DuraChrome, Fargo) use heat to fuse precise dots of pigment onto the substrate. There is no bleeding of one dot into another.

Thermal resin printers will never provide the same image quality as an inkjet printer. Inkjet printers have a higher DPI printhead count than thermal printers. Inkjet printers will produce an image that will hold up to close-up scrutiny more so than thermal printers. This is where you decide on purchasing equipment according to the need and expected usage of the printed product.

[ November 24, 2001: Message edited by: Glenn Taylor ]

Banding is partially due to too high an LPI for too low a DPI printhead. The trick is to find the right combination.

Would you like to go over that as well? I'll be more than happy to go over it with everyone if you'd like.

[ November 24, 2001: Message edited by: Glenn Taylor ]

An inkjet print may look great at .5 megapixel, but on outdoor signage it may look the same from the same intended viewing distance, until 6-10 months later when most of them would have faded beyond an acceptable degree.

BTW Glenn, I would be interested in the LPI to DPI balancing topic.

From re-reading your posts, I get the feeling that you've not been getting the results that you expect.

Are there any specific problems or observations about Edge prints you've received from your current provider? The reason why I ask this is because we are really about to get into the deep stuff about color reproduction.

Lets take a look at the "banding" that you mentioned. I assume that you are refering to the banding that occurs with a gradient image, not the mechanical banding that would occur with a ColorCamm.

First, the Edge does not have PL2 (postscript level 2). Because many of your better laser printers come with PL2 or higher, there is virtually no banding with gradients. On a basic level, all printers print gradients by printing bands of dots, stairstepping through 256 (this number varies) levels of concentration of dots.

The graphic below represents 11 shades of grey.

Inkjet printers use 256 shades of grey in the same amount of space. Combine this with the fact that the ink from inkjet printers bleed or fuzz-out and you will get a gradient with a very smooth transition from 100% to 0%. Its this same effect that allows low resolution images to print very nicely on paper.

Now you come to a piece of equipment that does not have the benefit of "fuzz-out" or as high a DPI printhead as an inkjet printer. This would include laser printers as well as thermal printers such as the Edge, DuraChrome, and ColorCamm (you could even include screen printing). These devices lay down very precise, very clean dots and do not bleed or "fuzz-out". Because of this, the operator has to pay a great deal more attention to detail as to how a file/image is prepared.

"Banded" gradients are the result of trying to print too high an LPI (lines per inch) with too low a DPI. The end result looks like this....

One way to compensate is to use a device that has PL2. What PL2, Postscript Level 2, does is add noise (dots) between each step to create the illusion of a smooth gradient.

However, it is not full-proof. I use a 24" Aspect 400 imagesetter to generate film positives for screen printing. It comes with its own RIP software and PL2. If I try to print a gradient at 45LPI and 400DPI, even with PL2, I will get a stairstep gradient. However, if I print a gradient at 45LPI and 800DPI, I get a perfectly smooth gradient. But, If I print at 55LPI and 800DPI, I get the stairstep effect again.

So, what has this got to do with the Edge? Plenty.

The Edge does not have PL2. And, at 300DPI, printing a smooth gradient can be a problem.

So, what is the solution? There are several.

One is to print at a lower LPI. However, this creates larger dots. Another is to change the pattern of the dots. However, when you change dot size and/or dot patterns, you also affect color saturation. This is why one combination of LPI, DPI, and dot pattern may work well for one image, but not necessarily for another image.

In the end, it is up to the operator to look at an image, and through experience, determine what combination will work best for your particular image.

If you are not getting satisfactory results from your current provider, then you may need to find another provider.

But, you have to remember, there are distinct pro's and con's to each method of printing. Each comes with its own difficulties and requirements. Just because one piece of equipment cost $300 and the other cost $20,000, there is no guarantee nor expect a better, clearer, smoother image from the more expensive device than from the cheaper. (this does not contradict my assertions about the ColorCamm vs. Edge - a totally different matter).

Anyhow, I hope some of this helps. Like I've said, its confusing. But, if you'll go back and read the articles I listed earlier, they do a much better job of explaining things than my meager attempt.

[ November 25, 2001: Message edited by: Glenn Taylor ]

Yes, but that was in the context of creating graphics in preparation for printing. In Jack's case, if I'm reading him correctly, he has a disatisfaction with the difference between inkjet printers and thermal printers.

Something every Edge owner needs to do is to scan, or in this case take a picture with a digital camera, of an object. Then, do a series of prints, each with a different dithering setting - Gerber Photo, Gerber Artwork, Classic Dot (do several different size dots) and Stochastic. You will be amazed at the difference in color saturation, gradation quality, and the change in the overall image.

****

Jack,

The articles I posted, while written for scanning, are applicable to your particular situation. After all, what is a digital camera but a different type of "scanner." It collects information in nearly the same identical manner. And, the collected information is handled in the same identical manner.

As for the "resolution overkill", as I stated before it is simply to collect as much "color information" as possible; and where I would suggest you "start". You may find later that going to such a high PPI may not be necessary in some applications and absolutely necessary in another application. Only you, based on your experience, can make that determination at that time. My point is that it is better to start with "overkill" and have too much color information and work your way down, rather than to have to "make do" with too little color information and struggle to work your way up. That was the point I was trying to make.

Back to your original question - "I need to know the resolution needed for them, format to e-mail, and price for these items."

My answer are as follows......
*Resolution - as high as you can provide.
*Format - whatever the format the camera uses.
*Prices - I'll e-mail them to you this morning.

[ November 25, 2001: Message edited by: Glenn Taylor ]

Here's the camera setting I would start off with.

quote:

---

0: TIFF 2048x1536 uncompressed

---

Reasons for this choice are as follows:

1. Uncompressed images contain as much color info as possible, as Glenn has already stated. This will provide you with the maximum amount of image information to work with. If after some experience you decide you do not need all this image info, by all means, use a lower setting.

2. As stated in those articles Glenn posted, the DPI is an irrelevant term when talking about the image file. DPI refers to the number of dots the printer is going to lay down. The print head is going to lay down 300 dots per inch whether the image is "set" for 600 DPI or 25 DPI, but the driver has to invent those missing dots and it does so by averaging out the existing "dots" around it.

If your image has a pixel size of 2048x1536, you can print an image that's almost 19"x14" in size using a 108 DPI setting.

This 108 DPI setting in no way relates to the printer at all, it's an imaginary number (actually it's the number that should be called PPI) that the software uses to determine the actual printed size of the image.

So, yes, you will be starting off with a 10MB uncompressed TIF file. That's 2048 x 1536 pixels, multiply that by 24 (24 bits per pixel in an RGB image).. then divide that by 8 (8 bits per byte, computer lingo) and you end up with a 9.437 MB file.

That is the best possible source image you can get from that camera.

If you dont want to deal with 10MB files you dont have to, but *you* need to experiment with the settings to find one that is acceptible for the work you do and your expectations. Using the compressed file formats from the camera limits the quality you can get from the camera so you may as well start off with the best you can.

Part of Glenn's point is you wont tell much of a difference between a 300 DPI print and a 108 DPI print when printed on a 300 DPI print head, and he's right. Your eye won't see much of a difference at all, so why use a 300 PPI file and burn up drive space and processing time when there's not going to be any discernable difference between the two? Don't take my word or Glenn's word for it, try it for yourself.

300 PPI is NINE TIMES more image info than 100 PPI which translates into an 85MB file based on the pixel dimensions provided by your camera.
If you want to print an image at 300 PPI on a 300 DPI print head at 11"x17", you need a 5100x3300 pixel size image, which is 50MB uncompressed. Since there wont be much difference between the 300 PPI print and 108 PPI print, you may as well go with a 108 PPI file which is only 6.5 MB in size.

In short, yes your new camera will provide what you need, in fact it will provide more than you need.
There is no relation between the optimum PPI file setting and the DPI of the printer, it just takes experimentation to find the right PPI settings for the software and hardware being used by whoever is producing the output.
The 108 PPI figure is probably the number Glenn has come up with (ie: lots of experimentation) which produces the optimum image output from his Edge while using the smallest image pixel dimensions possible, thus saving processing time and harddrive space.

i have had that camera for about a month and love it! it paid for itself the first week i got it. the quality is excellent. enjoy...

You've cleared up information for me in 30 minutes that has befudled me for 3 years.

You've cleared up information for me in 30 minutes that has befudled me for 3 years.