Fading to nothing AKA halftone trickery

[Gilligan]

This probably isn't exactly the right place and feel free to move it if it isn't.

We got a bid for this job and the art got me wondering how this could be handled... even if we don't get the job I'm curious.

As you can see it's got some "outer glow" going on fading to a black shirt.

My thoughts were that you could fade out the underbase and a quicker rate than the red so that the red would then start to fall off onto the shirt at the tips... maybe even just change the curve so there are a mix of underbased and non-underbased red dots and THEN just non-underbased dots at the very end.

I'm thinking this would give the best "fade to nothing" look on a design like this.

My concerns are, will accurip lay down the dots in the right place or would they be white dots falling outside of the red dots?  I really have ZERO understanding of how this works.  We send out all separations that get intricate like this so we end up just blindly printing what we are given and it all works out.  But, I'd like to get a better understanding of how the software works in this regard.  Plus this certainly isn't something we should HAVE to send out.  We would likely recreate the artwork in Illustrator and just do the gradients as spot color gradients... will this help or hurt the concept that I'm putting forward?

Again, I think it really comes down to not understanding, AT ALL, the math behind RIPS.

[Colin]

Yes, you basically will have a smaller white glow under your red.  The white dots will stop somewhere around your 25% point on the red (This is an estimate - it is all artwork dependent).

You will get a better visual fade to nothing if you can then print your red and highlite white wet on wet.

[Screened Gear]

Jobs like this you have to do a high LPI. If you use too low of a LPI your fade will only be a few dots and look like a rastered edge. Stuff like this I do 55 LPI.

[jvanick]

The doing the smaller white glow now makes sense of why the job I did looked so good with the seps from Scott from myseps...

Really makes a difference vs what separation studio outputs.

[Gilligan]

Scott is really good... I've been very impressed with the stuff we've gotten.

I tried to copy a 3 color grey scale that he did for us... LOL, what a joke... some serious skills go on in doing those seps!

[jvanick]

I'm looking forward to the next job that he makes us look good on.

It REALLY proved for me how much the art makes so much of a difference in how good the print looks.

[Sbrem]

I'm looking forward to the next job that he makes us look good on.

It REALLY proved for me how much the art makes so much of a difference in how good the print looks.

No 2 ways about it, if the art is bad, it doesn't matter how well you print it. (But, some people are pickier than others.) A skilled separator knows what to do with each channel or layer so that they print at their best; testing and experience will get you there. Before digital, when we shot film in a process camera, we would make halftones by laying a halftone screen over the film during exposure. The results were less than desired; a little digging and we learned that a good halftone made that way required 3 separate exposures, one each for highlight, midtones, and shadow. Today, that is done in Photoshop or whatever your favorite raster editing software is with Curves or Levels. The more you dig in, the more you'll understand and the better control you'll have.

Steve

[mk162]

you can nail dot on dot, it's tough and I don't think I would try it manually.

or...discharge....or DTG

[Gilligan]

you can nail dot on dot, it's tough and I don't think I would try it manually.

or...discharge....or DTG

That's a good point... I'll have to look and see what we've done that might be comparable.  We have some sim process stuff but that is much more forgiving if it doesn't line up perfect.

[Croft]

I'm looking forward to the next job that he makes us look good on.

It REALLY proved for me how much the art makes so much of a difference in how good the print looks.

X3  After a lot of years in the business , doing flat colour stuff Scott has made me look at art again in different ways , the way he uses white backers with top colours really works, I did some zombie shirts with red on black and it really looked like blood some red was just on the black and some on a white under base , that glow I assume could be done in the same way.

Every job he has sent me has worked with excellent results

[myseps]

Thanks for the positive feedback guys.   

The RIP won't create a base like you are describing.  You will need to make a base that is "Pulled back" in the glow area so that you will have a smaller "glow" of base as suggested above.  The red glow will extend further into the shirt.   I find that its easier to do this in Photoshop, where you can actually see how it will look by making your red channel 20% opacity.

[Gilligan]

Thanks for the positive feedback guys.   

The RIP won't create a base like you are describing.  You will need to make a base that is "Pulled back" in the glow area so that you will have a smaller "glow" of base as suggested above.  The red glow will extend further into the shirt.   I find that its easier to do this in Photoshop, where you can actually see how it will look by making your red channel 20% opacity.

Yeah, I knew the rip wouldn't do it automagically... I was mostly wondering if I did it manually would the rip put half of the 40% dots on top of the 20% base dots or would they arbitrarily fall where ever.

I think I know the answer but that's only based on gut feeling on how it MUST work vs experience or knowlegde.

Thanks!

[Dottonedan]

This is a job you can do yourself. If I were doing it, a simple job like that would cost $30.00 but it's about one of the easiest jobs out there to separate in raster.

1 Base white
2 Red
3 Top white.

In layers, (if it's not already flat), make a new channel, move to the top, MERGE visible,  (Select all) in that new layer, COPY, PASTE into new channel. Invert (command or control I) Keep this as your master.

Copy that master to create a template for knocking out the grayscale parts to eventually get your red.On the copy, adjust curves till you burn out the grayscale halftones, (and use that) as your selection to knock out of yet another copy of the master. This will be your red screen, (the grayscale halftone art).

The other, (solid white type) is the one being used to knock out what is essentially the white in the image on your red screen. This hard white (template) will eventually be your TOP WHITE. Create another copy of the master. This will be your base.

On the base, adjust curves. You want to beef up the deepest shadow areas. Stuff that is 85%, make them 100.  Stuff that is in the 1-5% range, should be burned out.  This will allow the top red color to overflow or overlap the underbase.On that red screen,

I would look at beefing up the red in the mid tone range out to the qtr tone range and fill that in a little more (as this is the area) that will begin to fade to the shirt more and you want more ink down. The areas less than the Qtr tone will fall straight to the shirt. All of this will overlap your underbase dots.

Thats pretty much it in a nut shell. There are many more different ways to do it and the % adjustments are all arbitrary.I will send you a version I did of this. Took less than 30 min.Yes, your dots will (or should) line up on top of each other if using the same line screen and angle on each sep as long as the registration is great. They will be dot on dots. Just some (like the underbase dots) will and should be smaller (as the %'s are lighter) like stacking a quarter on top of a dime.

I have tho, seen some RIPs that can do chokes on an underbase and the underbase was automatically made by the RIP.  Ours does that on the STE. It gives you a halftone underbase with options as to how much you want to choke the underbase by # of pixels (e.g.) 2-3 or even 5 pixels etc. on a 300ppi file or 1 pixel on a 150ppi file. It chokes dot for dot.  Everyone does their own underbases typically, as people like more control over the underbase in specific areas.  I was impressed to see that it actually does choke (all parts) or every dot and not just the outside. Still tho, this process is not for every sim process job. It would work great for any solid art.

[Gilligan]

Thanks Dan... I'll give it a whirl tomorrow.

[Sbrem]

Nice share, Dan, thanks.

Steve