RESOLVE FX · DIFFUSION FILTER · TONE LIGHTS
ColorTone Diffuser.
Build an optical look in layers.
Know what a LUT can and cannot carry.
ColorTone Diffuser models a virtual diffusion filter surrounded by user-controlled colored lights. Its useful mental model is a chain: signal description → resolution/diffusion → tonal eligibility → Tone Lights → center-weighted falloff → internal compensation → optional post-adjustment clipping → final Blend. 3D LUT Compatible is a separate representability boundary, not a quality switch.
Resolution · Diffusion
Bias RGB · Falloff
Full FX · 3D LUT subset
Use the lab to switch between the full spatial Resolve look and a LUT-representable subset. Deliberately break the input contract, raise Radius/Amount, push Tone Light bias, and watch which kind of problem appears instead of treating every difference as “strength.”
Teaching model only. It visualizes the separation between signal contract, spatial diffusion, Tone Light bias, falloff, image compensation, and LUT representability. It does not reproduce Resolve’s proprietary ColorTone Diffuser OFX filtering, LUT export, color science, or parameter ranges.
5 actual Resolve captures + independent optics/LUT teaching model
29 PRIMARY CONTROLS · CONDITIONAL ADVANCED CLIPPING01SIGNAL → OPTICS → LIGHT → COMPENSATION
The effect is easier to control when each layer has one job.
The physical metaphor matters because spatial diffusion, content-dependent Tone Lights, tonal protection, and LUT compatibility are not interchangeable controls.
It models a diffusion filter lit by colored perimeter lights, not merely a soft blur.
Blackmagic describes a virtual diffusion filter surrounded by a rectangle of user-controlled colored lights. Those Tone Lights scatter off diffusion elements in the glass. Resolution/Diffusion therefore control spatial softening/scatter, Tone Lights add a content-dependent color contribution, and Falloff models a stronger center with weaker edges. Treating the whole plug-in as one “glow amount” hides the purpose of half the panel.
3D LUT Compatible is a representability boundary, not a quality mode.
A 3D LUT stores a pointwise RGB→RGB mapping and cannot encode spatial blur, diffusion radius, or position-dependent falloff. Resolve therefore disables controls that cannot be represented by a LUT. If the LUT is the deliverable, enable compatible mode before designing the look; if the Resolve render is the deliverable, the full spatial look and LUT-compatible subset are not expected to match.
Input Color Space / Gamma describe the signal at the node — they are not creative controls.
Use timeline is correct only while the timeline color-management state truly matches the OFX input. An upstream CST, DCTL, or other transform can change gamut/gamma before this node. The overrides must then describe the actual input. A wrong contract changes how the effect interprets luminance and color, so compensating with Bias RGB or Contrast merely hides an architectural error.
Pre-effect Soft Clip and Advanced Clipping are two different stages in the chain.
The upper Shadow/Highlight Soft Clip blocks protect tonal extremes from ColorTone Diffuser itself: values beyond the threshold bypass the filter. Show Advanced Clipping Options adds a second equivalent soft-clip stage after Image Adjustments. That matters when in-plug-in Exposure/Contrast lifts highlights or alters shadows again. The stages are not interchangeable: the first limits entry into the look, the second protects the compensated output.
Resolve 21.1 manual boundary: ColorTone Diffuser is documented in Film Emulation without a Studio-only label. Blackmagic describes the virtual diffusion filter, perimeter Tone Lights, LUT compatibility, color-space overrides, pre-effect soft clips, falloff mask, and post-adjustment clipping explicitly.
02REAL RESOLVE 21.1 CONTROL MAP
Twenty-nine primary controls across four responsibilities, plus conditional Advanced Clipping.
The screenshot viewer uses only actual Resolve captures from the archive. Conditional Advanced Clipping is documented from the manual because the expanded state was not captured; no synthetic Inspector is presented as real UI.
AA · Entry / signal contract
Define the starting look and actual input state first.
- Presets
Ready-made starting configurations. The manual says Clean Slate makes no changes and is intended as the base for a Custom look; the other presets simply set the controls to different combinations.
VERIFYA preset is a direction, not a “film stock” or separate engine. After choosing one, inspect which actual controls changed.
- 3D LUT Compatible
Disables all parts of ColorTone Diffuser that cannot be passed through a 3D LUT, leaving only the LUT-representable subset.
VERIFYEnable it before developing a LUT. Do not expect the exported LUT to reproduce spatial diffusion/falloff from the full FX.
- Input Color Space
Describes the input gamut/primaries actually reaching the OFX. The default Use timeline reads the project state.
VERIFYRe-check after an upstream transform: timeline state and node input may no longer match.
- Input Gamma
Describes the input transfer/encoding. It is semantic context for the effect, not a “gamma grading” control.
VERIFYDo not choose gamma by eye. Describe the real encoding state at this point in the chain.
BB · Diffusion / tonal protection
Separate softening from light scatter and protect tonal extremes before the look.
- Resolution
Amount of blur/softening used to mimic the slight resolution loss of light passing through a physical diffusion filter.
VERIFYJudge at 100% Viewer. Excessive Resolution quickly removes fine texture even with moderate Diffusion.
- Diffusion
Enables spatial spreading of bright areas into neighboring parts of the frame.
VERIFYThis is a separate spatial operation. In a LUT-compatible workflow it belongs to the kind of processing a 3D LUT cannot store as spatial behavior.
- Amount
Strength of spatial diffusion: how strongly bright information spreads around the source structure.
VERIFYDo not compensate an oversized Radius by blindly changing Amount; the controls own different properties.
- Radius
Size of the scattering region. A larger Radius spreads the effect wider but is not equivalent to greater strength.
VERIFYFind the correct scatter scale first, then set Amount.
- Shadow Clip · Threshold · Softness
Shadow Clip enables shadow protection; everything below Threshold bypasses ColorTone Diffuser. Softness controls how gradually that bypass begins.
VERIFYThis is not Lift. Use it as an eligibility gate for the effect, not as ordinary black-level grading.
- Highlight Clip · Threshold · Softness
Highlight Clip protects the bright end; values above Threshold bypass ColorTone Diffuser, and Softness feathers the boundary.
VERIFYUseful when speculars should remain cleaner than the look. Check the transition on waveform and in motion, not only on one frame.
CC · Tone Lights / Falloff
Colored lights and spatial weighting are separate axes of the look.
- Intensity
Overall level of the virtual Tone Lights scattered by the diffusion elements.
VERIFYToo much Intensity turns the optical bias into a flat color wash. Start lower than the urge to “see the effect” suggests.
- Bias Red · Bias Green · Bias Blue
Relative RGB composition of the Tone Lights. The manual explicitly notes content-dependent behavior and stronger visibility of the bias in shadows.
VERIFYIdentical numbers do not have to look identical across shots. Match the image/scopes, not copied values.
- Apply Falloff Mask
Enables the center-weighted spatial mask: more effect in the center and less toward the edges, mimicking the light distribution around the filter.
VERIFYDisable it for an even full-frame look; for an optical footprint, do not leave its shape accidental.
- Preview Falloff Mask
Displays the mask in the Viewer: darker means less effect, white means more.
VERIFYThis is the primary diagnostic mode for Size/Softness/Amount. Turn preview off before final evaluation and render.
- Size · Softness · Amount
Size defines the center region, Softness shapes the falloff, and Amount controls how strongly the mask modulates the effect.
VERIFYDo not confuse mask Amount with Diffusion Amount: one controls spatial weighting of the overall look, the other controls the diffusion branch strength.
DD · Compensation / output trim
Compensate the low-contrast look inside the FX deliberately.
- Enable Adjustments
Enables the built-in Exposure/Contrast/Saturation block so compensation can live in the same FX node and participate in the LUT-compatible portion where representable.
VERIFYIf neighboring grade nodes already own these jobs, do not duplicate them automatically inside the FX.
- Show Advanced Clipping Options
Reveals a second Shadow/Highlight soft-clip stage that the manual places at the end of the plug-in, after Image Adjustments.
VERIFYUse it to protect the compensated result. It is not a replacement for the upper pre-effect soft clips; it is a different stage.
- Exposure
Built-in brightness/gain compensation after creative processing.
VERIFYMatch perceived brightness when comparing bypass; otherwise the brighter version is likely to feel “better” by bias alone.
- Contrast
Built-in contrast compensation for the low-contrast result created by diffusion/Tone Lights.
VERIFYDo not automatically restore the original contrast if the softer contrast is part of the creative intent.
- Saturation
Built-in saturation compensation after ColorTone processing.
VERIFYCheck skin/neutrals and the Vectorscope: saturation compensation plus strong RGB Bias can make the look chromatically fragile very quickly.
- Global Blend → Blend
Final wet/dry for the complete ColorTone Diffuser result, mixing the assembled effect with the original image.
VERIFYUse it as a final trim after Radius, Tone Lights, Falloff, and compensation are correct. Blend does not repair a wrong input contract or poor diffusion geometry.
03FULL RESOLVE LOOK vs LUT DELIVERABLE
Treat the full spatial look and the LUT-compatible subset as two different deliverables.
Switch the graph between the complete Resolve pipeline and the 3D LUT workflow. A LUT cannot encode spatial radius or a position-dependent mask, so parity must be defined before export rather than argued about afterward.
Build spatial softening first
Tune Resolution, Diffusion Radius, and Amount at 100% before adding strong colored Tone Lights.
The graph shows ownership and decision order rather than a mandatory node count. The full Resolve look may contain spatial operations that cannot exist inside a 3D LUT; treat those as separate deliverables, not as an export bug.
04BUILD IT IN RESOLVE
Four exercises cover subtle diffusion, center-weighted Tone Lights, LUT-compatible delivery, and post-adjustment clipping.
These are method exercises, not numeric presets. Material, working space, exposure, subject contrast, and final display all change the useful amount of diffusion and color bias.
Subtle optical diffusion without a milky image
Take the digital edge off while adding a slight colored scatter without destroying texture or specular structure.
PrepareKnown working-space input, Clean Slate, dedicated look node, Viewer at 100%.
- 01
Keep Use timeline only if timeline state truly equals node input; otherwise set the overrides.
- 02
Raise Resolution minimally; then enable Diffusion and find Radius before Amount.
- 03
Add low Tone Light Intensity and a very restrained RGB Bias.
- 04
Compare bypass at matched brightness; use Blend last.
The optical softness reads in highlights/texture without looking like a generic blur filter.
Center-weighted Tone Lights with controlled falloff
Create stronger colored scatter in the center with cleaner edges without turning the look into a normal vignette.
PrepareTone Lights active, Apply Falloff Mask enabled, Preview available.
- 01
Enable Preview Falloff Mask and find the center Size.
- 02
Use enough Softness that the footprint does not read as an obvious mask.
- 03
Raise Amount until the required spatial difference is achieved, then turn Preview off.
- 04
Then tune Tone Light Intensity/Bias separately rather than compensating with the mask.
The center feels lit through a filter without an obvious vignette edge.
LUT-compatible look without false spatial parity expectations
Build the portion of the look that can actually be exported as a 3D LUT.
Prepare3D LUT Compatible enabled before look design; a full Resolve version is kept separately.
- 01
Enable 3D LUT Compatible and note which controls Resolve leaves available.
- 02
Build the color/tonal portion only from representable controls.
- 03
Export the LUT and compare it in the target device/app under the same input signal contract.
- 04
Keep spatial diffusion/falloff as a separate Resolve-only stage if required.
The LUT matches the compatible subset, and the difference from the full Resolve look is documented rather than mistaken for an export failure.
Advanced Clipping after internal compensation
Raise Exposure/Contrast inside the FX without reintroducing highlight blow-outs.
PrepareImage Adjustments enabled; upper soft clips already set for creative eligibility.
- 01
Set the upper Shadow/Highlight Soft Clip first as a pre-effect gate.
- 02
Apply the required Exposure/Contrast/Saturation compensation.
- 03
Enable Show Advanced Clipping Options and tune the post-adjustment soft clip separately.
- 04
Check waveform and specular transitions in playback.
The creative effect keeps its tonal intent while post-adjustment clipping prevents unwanted peaks from returning.
05DO NOT CONFUSE THE JOBS
ColorTone Diffuser is neither Glow, Halation, nor a technical color-space transform.
| Tool | What it owns | Use it when |
|---|---|---|
| ColorTone Diffuser | Spatial diffusion + colored Tone Lights + tonal gates + falloff + compensation. | When you need one creative optical look with colored scatter and spatial weighting. |
| Glow | Spreads glow around bright regions with a different physical/visual intent. | When the job is emissive bloom/glow rather than a diffusion-filter + Tone Light model. |
| Halation | Models dye-layer reflections and edge halos associated with film response. | When you specifically need film-halation character around bright edges. |
| Color Space Transform | Technical gamut/gamma transform with tone/gamut mapping. | When the task is a correct technical transform rather than a creative diffusion look. |
06SYMPTOM → CAUSE → FIX
Most failures come from a wrong signal contract, excessive spatial processing, or confusing pre- and post-effect protection.
01The look changes dramatically after moving the node.+
Check Input Color Space / Input Gamma. The new placement may receive a different signal state after CST/DCTL/LUT.
02The image became too milky and lost texture.+
Reduce Radius and Resolution separately, then Amount. Judge at 100% Viewer; do not fix excessive scatter with Blend alone.
03Tone Lights look like a flat color wash.+
Reduce Intensity/Bias, then enable Preview Falloff Mask and inspect spatial distribution. Bias is especially visible in shadows, so different shots require visual matching.
04A visible boundary appears at Shadow/Highlight protection.+
Increase Softness for the relevant pre-effect clip and check the transition in playback. Threshold sets eligibility; Softness sets transition quality.
05Highlights blow out again after Image Adjustments.+
Use Show Advanced Clipping Options: it is the post-adjustment soft-clip stage. Do not retune the upper pre-effect clip for a problem created later in the chain.
06The exported LUT does not match the full Resolve node.+
This is expected when the full look uses spatial diffusion/falloff. Develop the LUT only in 3D LUT Compatible mode and document Resolve-only stages separately.
07Copied Tone Light values do not match across shots.+
RGB Bias is content-dependent and reads more strongly in shadows. Use a shared baseline, but match with Vectorscope/RGB Parade and the image rather than numeric values.
100% Viewer for spatial decisions
Resolution, Radius, and fine texture cannot be judged reliably in a small Viewer. Inspect hair, pores, fine lines, specular edges, and texture at 100%.
Scopes for compensation and bias
Waveform is useful for pre/post soft clip and Exposure/Contrast; RGB Parade/Vectorscope help with Tone Light bias and neutrals. Scopes do not replace spatial-diffusion judgment, but they catch tonal/color drift.
Bypass at matched perceived brightness
A diffusion look often lowers contrast, while Image Adjustments can restore brightness. Compare on/off at similar perceived brightness so loudness bias does not make the test misleading.
07BEFORE YOU DELIVER
Approve the signal contract, texture, Tone Light color, falloff, and the actual target pipeline.
The Resolve render and an exported LUT are not automatically the same product. Test each one in the path where it will actually be used.
08FAQ
The distinctions that keep ColorTone Diffuser predictable instead of “mysteriously cinematic.”
Is ColorTone Diffuser Studio-only?
In the current Resolve 21.1 manual, ColorTone Diffuser is not marked “Studio Version Only,” unlike Camera Shake, Film Grain, and several neighboring FX. This guide therefore does not label it Studio-only.
How is Resolution different from Diffusion Amount/Radius?
Resolution mimics overall resolution loss/softening through the filter. Diffusion enables a separate spatial scatter: Amount controls strength, Radius controls area size.
What does 3D LUT Compatible do?
It disables controls that cannot be represented in a 3D LUT. It is a compatibility boundary, not a quality setting. Spatial diffusion/falloff should not be expected in the exported LUT.
Why can the same RGB Bias look different on different shots?
The manual explicitly calls the Bias controls content-dependent and more visible in shadows. Different tonal/color structure changes the perceived result.
Why use Input Color Space/Gamma if the project is already color managed?
Use timeline works when project state matches the actual input of this node. If an upstream CST/DCTL changed the signal earlier, the overrides let you describe the real input.
How is the upper Soft Clip different from Advanced Clipping?
The upper Shadow/Highlight Soft Clip operates before the main effect and excludes tonal extremes. Advanced Clipping appears after Image Adjustments and protects the compensated output.
Does ColorTone Diffuser replace Glow or Halation?
No. It models a diffusion filter with colored Tone Lights and falloff. Glow builds emissive bloom around bright regions; Halation models film-like dye-layer reflections/edge halos.
Where should ColorTone Diffuser sit in the node tree?
A dedicated creative-look node after the signal state is known is usually easiest to manage. This is not a hard rule: the overrides allow different states. The important requirement is explicit input contract and consistent placement across shots.
Sources and verification boundaries
Visible names and dropdown states were checked against the five actual Resolve captures in the project archive. Control semantics, Advanced Clipping behavior, and the physical filter metaphor were cross-checked against the Resolve 21.1 manual and Blackmagic’s official Resolve 20.1 New Features Guide. The interactive lab is deliberately an independent teaching model; it does not reproduce proprietary OFX filtering, LUT generation, color science, or pixel output.
This web pass does not claim live Resolve-render certification. Final acceptance belongs to the installed Resolve build, the real shot, the exported LUT where applicable, and the rendered delivery file.