How to do post-render editing on image - resolution too low for digital - should I print/scan?
heisdave
Posts: 130
Hi I am rendering some characters and scenes and I want to add some post-render linework on them. I have tried the shaders that are supposed to add outlines auomatically and I couldnt get them working properly.
So I have two choices..
choice 1 -save the rendered image files and work on them digitally in photoshop. When I open the file in PS and try to draw a thin line it looks clunky and pixelated. I think the problem is that the rendered image needs to be much higher resolution. So I try to adjust the render parameters but i just end up expanding the SIZE of the render which includes more of the viewport, but does not increase the resolution of the image. Anyone know what parameters to adjust for this?
choice 2- Go old school, print out my images, do my line work with pens, and scan them back in. But i dont want to lose quality when transferring. If i print to printer paper, can I do line work on printer paper or will that look like crap ? Plus I have an older printer and scanner, what resolution (DPI?) is adequate for the printer and the scanner to faithfull reproduce the art?
Anyway, Im sure someone has worked through this before. Any ideas? Thanks
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
you might try the shadow box trick. I am not sure if you know what a shadow box is But
Render your image you want first.
then use your rendered image as a back ground on a pane. and go back and reload your scene leaving your render in the pane as the background.
Then place your characters over the first rendered characters in the background pane. adjust distance from background pane as needed for shadow line.. this trick leaves a distinct shadow line around the characters making them pop out like using a drop shadow in Photoshop layer. only your doing it inside daz studio overlaying your characters over your first render. and adjust your distances that is same as a shadow box
First question would be : what shader did you use to do outline and in which renderer? Last time I did one, 3delight outline worked in DS 4.10 but not DS 4.11 beta
Second point : in render settings->general you can specify the size of the render in pixel. There is no dpi setting for what I know but you can adjust the render size to conform to the resolution you want
Ivy, thanks Ill try that. How do you insert an image into the viewport to use as a background?
Double size the image, apply the line art on a separate layer, save the line layer separately, shrink it down to the original size, apply it to the art, make sure you use anti aliasing, and save. Should solve the jagged edge issues.
You might try one of my free panes, That way It won't cost you anything if the shadow box trick does not work for your needs.. ( http://www.sharecg.com/v/92902/gallery/21/DAZ-Studio/Ivys-Magic-Pane-prop) You can add your pre render image to the pane under the surface tab and make your adjustments according to how you want your shadow to outline your characters., you may need to add a distant or spot light source in front to make your shadow outline more pronouced. But the effect is really cool :)
If you do decide to go down the "add lines by hand" route, then one option would be to print the image in draft (quality is not important), lay some tracing paper over the print and draw on that. Then scan in the tracing paper, and use an image editor to remove the tracing paper in the scan (fuzzy select should do this), and then layer that on top of the original render. That way original render quality is not comprimised in any way via the print/scan process.
that is what I do and never had any issues doing it that way
The rendered art is not your DPI source. You could or should* (to each their own) create a (new) document that matches your size requirements (in Photoshop) - first.
Say 3,000 wide X 2400 tall (arbitray choice or pick size by ratio, HD sizes, etc...) and create a NEW document in photoshop with that size in pixels and 300 DPI (anything past 300 should do fine/450 is probably the max that's needed) and then DRAG your Daz render into that NEW DOCUMENT. Photoshop works at the resolution of the master document settings and therefore, filters and lines and text will be scaled (visual quality) based on the ORIGINAL DOCUMENT SETTINGS. Daz, if I recall renders at 150 DPI, too low for most photo-editing standards.
And you have 3 choices. You could always ask for help with that shader that gave you problems. Sort out why you didn't get the results you wanted or expected, use a different (simpler?) shader and still go forward with drawing on top of that too.....
And maybe choice 4 is/are all the filters and the Photoshop actions that give the traced line-artwork look.
That is not correct. PPI is just a number at the front of the file, saying how big a pixel should be in print (by specifying how many there are to the inch). The pixels are not affected by the PPI setting in any way, each pixel is just a square with one of 16 million colours. DS does not assign a PPI value to the renders. If you drag an image into another Photoshop document it will be placed at 1 pixel/pixel - PPI has no bearing. There's no set PPI needed for editing, you need enough actual pixels to work with.
There's no set PPI needed for editing, you need enough actual pixels to work with.
I should have said "printing" and not editing. And it's Photoshop that does/looks/works better at higher resolutions/DPI, not the Daz Render.
I need to get in front of my machine. There's a 150 DPI number somehwere in this mix.
DS does not assign a PPI value to the renders.
Do you mean DPI, instead? Because Daz is consistently landing at this 96.012 PPI setting.
I still don't know why that 150 resolution is sticking in my mind.
If you set the size in inches (or cm, or points - any physical unit) then of course PPI matters - without it you are not telling the application how many pixels to use. But if you set the size in pixels then PPI has no bearing on image quality per se, just on how big the pixels (and so the image) are.
I meant PPI - pixels per inch. DPI - dots per inch - is a measure of printer capability, and will be higher than the optimal PPI for the device (since printer dots have fewer possible colour values than pixels it takes multiple dots to match a pixel).
When you load an untagged image Photoshop (or whatever you use to edit) will assign a value - but that isn't in the file, and in some cases you can choose what value is assigned (72 and 96 PPI are the common choices).
That's good to remember. It would explain why the only choices seem to be longer render times from the render-quality-settings or rendering at a larger size and letting it be (automatically) scaled down by placing the image in a (I won't say higher resolution- I'll say a) denser-pixel-environment.
denser pixels is a good concept.
Simply placing an image in another won't change its pixels in any way (well, other than pre-multiplying any alpha belnding)
See, I knew we talking about two different things. Forget DPI and substitute every PPI for DPI.
The idea is that Photoshop works at the base resolution. So an image that was created at 72 PPI has its fonts working at 72 PPI and the lines and other graphics work at that (72 pixel) resolution.
That's why the OP was having jagged lines and fonts. It's because of the PPI setting "of the render".
If you just bring the render into Photoshop -- that's the resolution....the original document's setting.
That's fine for LOOKING at a Daz Render and that obviously has enough pixels to edit pixels, but for pixel-intensive-operations, you (may) get issues.
That's why you don't (really) do text heavy stuff in Photoshop. That's what Ilustrator is for. If you set the resolution (PPI) high enough you can get over.
And that's where that 150 comes from. When I converted my art to poster-printable-formats, they demanded 150 DPI.
I actually called the company and asked why not the industry-standard (low-end) of 300 PPI and they said they had a process.
And I admit whatever they did worked because I have a massive showroom banner that has no jaggies or artifacts even though the original wasn't as crystal as I thought it should have been.
STAPLES does this too...They say they take any format, but I have seen large files stall their set up so I think they do some resizing/scaling too.
If you add text or lines in Photoshop they have to be rendered into pixels. Since text size, and line width, are specified in real-world units those values will depend on the PPI setting - but if you want text that is a certain visual size relative to the image size then it's the number of pixels that matters, not the PPI.
The resolution in digital images is mesured i pixels not ppi. There is no ppi in digital images.
However digital images can have ppi (often called dpi when you look at the file) in the metadata. This is so software like photoshop can make a size in inches when you open them, also when you open the image size calculator it gives a starting number. This can of course be changed at any time and does not affect the actual image.
Monitors dont care about the ppi tag and printers dont care about the ppi tag.
Printers need pixels, nothing else. If you want to print 10x10inch print at 300ppi you need to send 3000x3000 pixel image to the printer. You can export the image in pixels or ppi x inches, 300ppi x 10 inches or 600ppi x 5 inches for example (and create ether 300ppi or 600ppi tag in the image metadata at export). The result will be the same, 3000x3000 pixels that will be 300ppi at 10 inches in the print.
As for how many pixels you need for good print? 300ppi will be great for up close viewing. 240ppi should be fine. At 180ppi I can see a bit of a difference and consider this as lower limit. At 150ppi things start to break down for up close viewing, big prints can go lower because of viewing distance.
The fact that you can assaign ppi value to image (in the metadata) seems to confuse alot of people.
Hi, OP checking in.
So it seems that there is no way to achieve "denser pixels" when rednering with 3ddelight? or did I miss something? Adjustting the shading rate slider or pixel sampling slider would not increase the resolution?
The best solution seems to be to double size the image in PS, do my linework on a layer, then shrink the layer when done.
Thanks
Yes that is correct and no you ddin't miss anything.
Adjusting shading rate is only a quality setting, finer image so to speak.
Yes
Rendering larger is the way to achieve "denser pixels", in as far as the term makes any sense in relation to a bare image (rather than an image placed at 100% in an application that measures its pages in physical units, or an image sent to a printer at 100% scaling). Resizing the image is not going to give you more real information in the image, it's going to try to guess what should go in the additional resolution (menaing pixel dimensions, not PPI).
Correct; rendering the image with a higher resolution is the way to go. Same as in photography; picking the right tool/process for the job (e.g. a super-hi-res single capture, or compositing several captures of lower res shots).
It's best to start with the desired output. Many clients that need print materials should specify one of two things: