Creating halftones / detail for small prints

[kidink]

My client is printing a 5 color sim process job, the halftones at 65lpi for a full size chest print at 15" wide look great, loads of detail.

However when I resize the chest print down to say 4" wide and generate the halftones via photoshop at 65lpi because the art is so small the halftones look massive and there's barely enough information to replicate the details.

How do you tackle this? Should i keep the artwork at 15" wide, halftone and then resize?

[cbjamel]

What percentage at 65lpi, thats the question. are you creating the halftones or is the seps/rip creating them. should let the rip create them.
Shane

[Sbrem]

If the original art is continuous tone, then no matter what size image you output, if it's at 65 line, all dots are the same size. However, a smaller image like a left front has a lot
less dots representing the area than at full size...

Steve

[MNTS]

Can you share the image? If I can see it, I may be able to give better pointers as to what you should do to keep it looking nice for a left chest print. 15" to 4" is a steep change. When you do the resize are you doing that before processing the separations and tones or after?

[kidink]

See attached. If the client had asked for this at 15" wide i'd feel confident that when I'm creating the halftones in photoshop there will be plenty of information there.

However, if i've separated the art into channels and then resize it down to 4" when i go to image-mode-bitmap and create the halftones at 65lpi the dots look massive (i know they're not) and there's barely any information there as the print is so small.

What's the best way to tackle this?

[kidink]

Did anyone get a chance to look at this? Still wondering what's the best way when wanting to halftone something thats only 4" wide but contains a lot of detail.

[inkman996]

Cannot open your linked image.

[Frog]

Did anyone get a chance to look at this? Still wondering what's the best way when wanting to halftone something thats only 4" wide but contains a lot of detail.

Link downloads a file with 0 bytes. I see it says that it's a png file. If so, and no more than 20000KB, it should attach just fine and not even need a link.

[kidink]

Try: https://imgur.com/kH2Y3et

[kidink]

What percentage at 65lpi, thats the question. are you creating the halftones or is the seps/rip creating them. should let the rip create them.
Shane

I don't currently have a RIP so am relying on halftoning through photoshop.

I can't really see why this makes any difference however as before sending to the rip i'd still have to resize the art to 4" wide. There's 2 ways i could resize though;
a) I could resample the image to 10cm and keep the resolution at 300ppi
b) I could resize the image to 10cm which would bump the resolution up to something like 1900ppi (as it's much larger than 4" wide at 300ppi)

[kidink]

.

[Croft]

are there a lot of shirts ? , that looks like a good candidate for a digital print transfer.

[kidink]

Way too many and the separations have already been created, it's just making the halftones when it's going to be such a small chest print has me stumped.

[zanegun08]

Way too many and the separations have already been created, it's just making the halftones when it's going to be such a small chest print has me stumped.

You do the same process for both the 15" wide print as the 4" wide print.

A 65 LPI dot is the same size wether printing a 4" image or printing a Boeing 747, it is 65 Lines Per Inch.

You won't be able to get the same details on the left chest as the back print, for the left chest I would lighten the darker ranges and darken the lighter and try to simplify the image a bit to make the transitions bigger.

To answer your question, you can scale down the image to 4" whatever DPI you want doesn't make a difference (higher is better but 300DPI is fine), then you'll create your bitmaps at 300 DPI / 65 LPI / Elliptical Dots / Whatever Angle you commonly use.  The benefit to photoshop is that you can see that output instantly and adjust from there on what you know will happen on press.

If I were tasked with printing the left chest, I'd try to make it look more vector looking, and then have the back be higher detailed, at such a small scale packing in to much detail will make it look fine at close viewing distances, but as you move away from it (6ft social distanced viewing) it will turn into a blob in my opinion.

Good luck!  Post your results.

[Dottonedan]

Way too many and the separations have already been created, it's just making the halftones when it's going to be such a small chest print has me stumped.

You do the same process for both the 15" wide print as the 4" wide print.

A 65 LPI dot is the same size wether printing a 4" image or printing a Boeing 747, it is 65 Lines Per Inch.

You won't be able to get the same details on the left chest as the back print, for the left chest I would lighten the darker ranges and darken the lighter and try to simplify the image a bit to make the transitions bigger.

To answer your question, you can scale down the image to 4" whatever DPI you want doesn't make a difference (higher is better but 300DPI is fine), then you'll create your bitmaps at 300 DPI / 65 LPI / Elliptical Dots / Whatever Angle you commonly use.  The benefit to photoshop is that you can see that output instantly and adjust from there on what you know will happen on press.

If I were tasked with printing the left chest, I'd try to make it look more vector looking, and then have the back be higher detailed, at such a small scale packing in to much detail will make it look fine at close viewing distances, but as you move away from it (6ft social distanced viewing) it will turn into a blob in my opinion.

Good luck!  Post your results.

I would ditto all of this with one additional comment.


When doing my bitmaps, I like to do so at the same maximum resolution as my output device. If my film printer is a 600dpi printer, (Epson 1430) then I would use 600dpi (for the sole purpose of producing better or more formed dots). This part has nothing to do with the LPI being used. 300 works, and 600 works.


Some people get that idea confused with the old method for determining your files resolution you need. Typically 2 or 2.5 x your lpi (at a minimum) will be your resolution you would need. That may be accurate for printing a photo with halftone from photoshop, but that assumes your are printing to a printing device with a good RIP and not building your own halftones (using Photoshop).


What we are doing here with pre-halftoning and using Photoshop to do our bitmap halftones requires a higher resolution to form the best dots possible when sending to your device. Like they say, “crap in, crap out”. The 600ppi will be output to your 600dpi printer more minutely than 300ppi. If you need proof, just take a PSD 5% fill of black, at 1”x 1”. Bitmap one copy at 300 and bitmap another at 600 and add that to any film you print out next and you will see the difference. Actually, you can see the difference in the dots when you blow up in photoshop.  If my film device was a 1200dpi printer, I would bitmap my files at 1200ppi but I don’t have one of those.




For others coming along to this later, who might be unfamiliar with pre halftoned files (that might get the idea of reducing after the fact), to left chest or any size smaller, don’t. To reduce a pre halftoned 13”wide image to a left chest would be like 211 line screen. A 65lpi at 50% reduction becomes 130lpi. (in case anyone was thinking it would be the same when reduced). Side note. That’s how I used to do my halftones before digital printers and shooting film using a processing stat camera. I would print the image at say 35lpi but at 2 times my final print size on standard copy paper using the old laserwrighters. We then would shoot that to film but with a 50% reduction and they would come out much cleaner, better quality. Back then those were 300dpi printers, so my end result was a 600dpi 70lpi. Workarounds were a thing.

Pertaining to your file,
To create your left chest, I would start with your original larger greyscale image seps. I would reduce those to the size needed, but FIRST, I uncheck RESAMPLING. This will maintain the tonal integrity of the image (keeps all of the data)...but makes a very high resolution in proportion to the reduction in size.
Once, I’ve done that, I will either KEEP the resolution at that same REALLY HIGH Rez, and then do my bitmap at 600PPI or, I will lower that resolution to a safe Rez such as 600ppi. This does noting negative to your file. If you were to ever use that with say 1300ppi rez, all that would happen is that any additional resolution would be kicked out of the file at the ripping stage (if you used a rip later)...and you can re size it later larger (if ever needed).  I’ve had that happen once or twice where I needed to take my left chest back up. Doing it this way, maintains the integrity of the file.


Like everyone was saying, reducing that image to 4”wide (in 65lpi) still does not give you much room to showcase your tonal transitions. Like, how do you show a transition from a 10% to a 5% in a .0625” area? Basically, it would be putting two dots next to each other that are roughly the same size.


This is why when we artist are designing, we won’t use the exact same image from the back and put it at a left chest size. We would build another file for the left chest using only some elements of the back (and enabling us to make those elements much larger). Maybe just some type...and the car emblem or something. Can’t pack 10lb of crap into a qtr lb bag.