Colored Flood Lights: RGBW Control, Beam Mixing, Facade Use, and Specification Risks

Specification guide for lighting designers, controls engineers, procurement teams, and project owners · Updated August 2026

Colored flood lights are a system, not a color swatch. The approved result depends on the exact luminaire, color channels, driver or decoder, controller, control personality, optics, power conditions, mounting geometry, facade surface, viewing position, and scene file. Renderings and the label “RGBW” can’t prove that those parts will produce an even, repeatable scene on site.

This guide turns that uncertainty into two usable procurement tools: the Color-Flood Evidence Spine for testing what each document actually proves, and the Scene Commissioning Dossier for binding an accepted scene to its configuration and owners. It doesn’t claim an unpublished product option, universal mixing distance, blanket code compliance, or guaranteed project result.

Colored Flood Lights Are a System, Not a Color Swatch

Colored flood light system boundary map from luminaire configuration through acceptance.
Colored flood light system boundary map from luminaire configuration through acceptance.

A color-changing flood light uses multiple controllable light channels to produce selected colors, but the fixture is only one part of the result. RGB commonly refers to red, green, and blue channels; RGBW adds a separate white channel. Whether a specific product changes color, how it’s addressed, and what its mixed output looks like must be confirmed for the exact configuration.

Write one system scope line before comparing samples: exact fixture configuration + driver or decoder + controller and interface + control personality + optics + supply and protection boundary + mounting and surface geometry + acceptance owner and scenes.

CIE S 025 sets requirements for reproducible photometric and colorimetric measurement and advises how to report the data. That makes an unspecified setup inadequate as exact-model evidence.

“Photometric and colorimetric performances are considered individually for each configuration.” CIE S 025 public description

Eight boundaries to freeze before a color review
Boundary Minimum record If omitted
Luminaire Model, revision, emitter/channel configuration, quantity A sample or report may describe a different build.
Drive chain Driver or decoder model, channel capacity, supported modes “Compatible” parts may still behave differently together.
Control Protocol, controller/gateway, personality, addresses, loss state Without those fields, a scene cannot be reproduced or assigned to an owner.
Optics Lens or distribution, emitter layout, shielding Beam angle becomes a poor proxy for color uniformity.
Electrical Supply, protection design, dimming and operating state Measurements may not represent the installed condition.
Geometry Mounting, aiming, throw, setback, zone, surface A tabletop test is mistaken for a facade result.
Observation Viewpoints, ambient condition, spill boundary, critical receptors Approval can miss separation, glare, nuisance, or habitat effects.
Acceptance Scene IDs, criteria, evidence, sign-off, retest trigger A subjective approval cannot be audited later.

What kind of flood lights can change color?

Flood lights can change color when the complete system provides independently controllable color channels, such as RGB or RGBW, through a compatible driver or decoder and controller. The product label alone is not enough. Confirm the exact channel map, control personality, supply and wiring topology, available scenes, dimming behavior, and optical configuration for the named model. Then test intended colors and white scenes at representative project geometry instead of approving a catalog image.

RGB or RGBW? Specify Channels Before You Specify “White”

RGB and RGBW channel comparison for project specification.
RGB and RGBW channel comparison for project specification.

RGBW is not automatically “better RGB.” A dedicated white channel adds another input that may serve functional white, pastel colors, efficiency, or a particular tuning strategy. It doesn’t by itself prove white quality, color consistency, brightness, dimming behavior, channel balance, optical mixing, or controller compatibility. The procurement risk is assuming the extra channel resolves those questions.

A 2025 peer-reviewed RGBW study is useful precisely because it’s configuration-specific: the researchers evaluated color coordinates, color difference, correlated color temperature, thermal conditions, and selected operating modes in an automotive interior-lighting system. Its results don’t transfer to a facade flood light. The author affiliations include Infineon Technologies, so this guide treats the paper as configuration evidence, not as independent product endorsement. For this guide, the transferable lesson is to test the actual channel architecture and operating condition rather than infer performance from the four-letter label.

Conditional RGB versus RGBW comparison
Question RGB submission should show RGBW submission should show Neither label proves
Channel map Order, range, personality, shared functions The same, plus the exact white-channel position and function Correct addressing in the integrated system
White scene How the three color channels create the specified scene Whether white is used alone, blended, or constrained by mode Target color quality or consistency
Dimming Critical levels and transitions in the intended configuration Critical levels for color and white channels together Smooth fades, camera behavior, or absence of visible artifacts
Optical result Distribution and color-uniformity evidence Evidence across both saturated and white-influenced scenes A universal mixing distance or clean beam edge
Approval Representative mock-up with named scenes The same, including why the white channel exists Suitability for the final surface and viewpoints

Ask the supplier for the exact manual and channel table, not a generic family brochure. Whether the operator uses a show controller, building interface, or remote control, define scene names as numbered, version-controlled states. If the white channel has no written purpose, the schedule hasn’t yet specified RGBW, it has only repeated a product category.

Beam Mixing Is a Mock-Up Question, Not a Beam-Angle Promise

Beam mixing risks under short throw and oblique aiming conditions.
Beam mixing risks under short throw and oblique aiming conditions.

The channel choice now needs an optical check. A beam-angle value describes an intensity-distribution convention; it does not establish multi-channel color uniformity. Visible red, green, blue, or white separation can change with emitter spacing, lens design, focal relationship, throw, aiming, surface texture, incidence angle, dimming level, and viewer position. Hot spots and colored edges may be inconspicuous on one surface and prominent on another.

Request luminous-intensity distribution for the exact optical option and angular color data if the manufacturer has it. Record the emitter and lens configuration to which the file belongs. Then test the worst representative combinations: short throw, oblique aiming, grazing texture, beam overlaps, low dimming levels, pastel scenes, saturated scenes, and a viewing position that can see the beam edge.

Turn an optical claim into a hold point
Claim Evidence before mock-up Representative check Hold point
“Uniform color” Named configuration, measurement method, sample images or angular data Observe critical scenes across the illuminated zone and approved viewpoints Do not approve if the acceptable variation is undefined.
“Wide beam” Exact distribution file and aiming basis Check overlap, edges, spill, and surface luminance pattern Beam angle alone cannot close the color-mixing review.
“Smooth dimming” Driver/controller configuration and measured behavior at specified states Run fades, static low levels, and transitions used by the project Test the delivered scene file, not a separate demo controller.
“Facade ready” Environmental, optical, control, and installation evidence Reproduce mounting, surface, viewpoints, ambient light, and receptors A showroom wall is not the final geometry.

There is no defensible universal “mixing distance” in this article because the available evidence does not support one across products and geometries. The supplier can resolve the uncertainty with configuration-linked data; the project team can resolve application risk with a representative mock-up and explicit acceptance criterion.

Map the Control Chain Before the Fixture Schedule Is Frozen

Colored flood light control chain from controller to scene file.
Colored flood light control chain from controller to scene file.

A protocol label moves data; it doesn’t assign integration responsibility. Map the chain from supervisory control or show controller through gateways, distribution, decoder or driver, fixture personality, addresses, scene file, power state, and loss-of-data behavior. Integration risk appears when responsibility between those interfaces is left undefined. For deeper functional background, use the LED controller compatibility guide and DMX controller selection and address planning; this section stays focused on the project interface.

ESTA’s Spring 2024 revision article prioritizes backward compatibility and describes a standardized loss-of-data classification and declaration method. It supports an interoperability framework, but the public update doesn’t prove that two unnamed products implement the same personality, features, loading assumptions, or fallback state.

Control interface and ownership schedule
Interface Freeze in the submittal Accountable party Acceptance evidence
Controller to gateway Protocol/edition, universe or network map, configuration backup Controls integrator Approved topology and restored-backup test
Gateway to field distribution Physical interface, segmentation, termination, isolation basis Electrical/controls design As-built schedule and witnessed communication test
Field device to fixture Personality, channel order, address, supported features Supplier plus integrator Exact manual and point-to-point checkout
Scene data Scene ID, file revision, channel states, fade behavior Designer and owner Approved scene file and ledger record
Abnormal state Power recovery, signal loss, default output, restart ownership Project team by interface Witnessed loss-and-recovery test under an approved procedure

Don’t overpopulate this schedule with generic cable lengths, device counts, or loading limits. Those values depend on the chosen interface, equipment manuals, topology, and installation. Compatibility is a testable integrated behavior, not a noun copied from multiple data sheets.

Facade Geometry Changes How the Same Scene Looks

Facade geometry, viewpoints, and receptor conditions for a lighting mock-up.
Facade geometry, viewpoints, and receptor conditions for a lighting mock-up.

Once the control chain is mapped, the same scene file can still look different across facade zones because throw, setback, aiming, surface material, texture, reflectance, curvature, recesses, glazing, ambient light, and overlapping beams alter the observed result. Freeze those conditions before approving a sample. The separate facade and architectural lighting design process covers broader concept development; here the concern is whether a colored scene survives the actual geometry.

A 2024 peer-reviewed complex-facade case used geometry, simulation, luminance-distribution analysis, and iteration rather than a product-only selection. Its project result is not a forecast for another building, but its method supports zone-specific review where a flat sample cannot represent the final elevation.

Geometry-to-observation plan
Variable Record Observe in critical scenes
Mounting and aiming Coordinates, height, tilt, rotation, setback, throw Beam overlap, scallops, hot spots, cut-off, access conflicts
Surface Representative material, texture, finish, color, condition Perceived hue, saturation, brightness, shadowing, edge separation
Viewpoints Primary design views plus close, oblique, road, and neighboring views Uniformity, direct view, glare, spill, visible sources
Ambient context Nearby lighting, signs, glazing, operating hours, weather condition Scene legibility and conflicts at the intended time
Receptors Residents, drivers, pedestrians, wildlife, observatories, adjacent property Intrusion, distraction, skyward light, habitat exposure, complaints

DarkSky and the Illuminating Engineering Society frame responsible outdoor lighting around useful, targeted, low-level, controlled light and warmer color where possible. Dynamic facade colors need a project-specific justification, schedule, aiming and spill review. Also check local planning, transport, environmental, heritage, curfew, and signage requirements; a website article cannot establish the rule for an unnamed jurisdiction.

Wall-washing form factors may also change the optical approach. Review LED wall washer light options as a form-factor contrast, not as proof that a particular distribution, color engine, or control mode suits the project.

Treat IP, Power, Dimming, and Safety as Separate Evidence Lines

Separate evidence lines for enclosure, electrical, temporal light, and photobiological safety.
Separate evidence lines for enclosure, electrical, temporal light, and photobiological safety.

That geometry review still leaves separate product and installation risks. “Outdoor ready” compresses them into one phrase. Split the decision into enclosure evidence, installed water and material management, electrical coordination, temporal-light behavior, jurisdictional requirements, and photobiological-safety assessment. Each line needs its own exact-model evidence and responsible reviewer.

What an IP code proves: IEC 60529 classifies the protection provided by an enclosure against access, solid foreign objects, and water under defined tests. What it does not prove: the integrity of the completed cable system, field connectors, orientation, drainage, condensation management, corrosion resistance, cleaning method, workmanship, or long-term maintenance.

Temporal-light behavior also needs configuration-aware evidence. United States Department of Energy guidance describes multiple temporal-light metrics and measurement conditions rather than one universal “flicker-free” threshold. Test the exact luminaire, driver, controller, firmware or personality, scene, channel balance, and critical dimming levels relevant to people, cameras, or other project equipment.

Do not transplant a regulatory number without its scope. European Union Regulation 2019/2020 includes temporal-light requirements for covered products, but the modulation metric provision has stated exceptions that can include light sources intended for outdoor or industrial applications. Determine the market, product scope, amendments, exceptions, and contract requirement before using a value as an acceptance limit.

IEC 62471-7:2023 provides a current scope for photobiological-safety assessment of luminaires that predominantly emit visible radiation. Its public standard page does not establish the risk group of an unnamed fixture or the safety of a specific mounting distance, exposure duration, maintenance task, or viewing condition. Ask for exact-model assessment and have the responsible project professional evaluate intended use.

Independent evidence lines for an outdoor colored system
Line Request Project check Do not infer
Enclosure Exact-model report/certificate, orientation and instructions Entries, connectors, drainage, condensation, cleaning, corrosion “Waterproof system” from one enclosure code
Electrical Supply, input, driver, protection and wiring data Site supply, coordination, fault protection, loading, qualified installation A universal surge device or conductor size
Temporal light Measurement method, waveform/metrics, exact configuration and levels Critical scenes, dimming, cameras, exposed users, local requirements Zero flicker from a driver label
Photobiological safety Exact-model assessment, test setup, classification and limitations Distance, exposure, viewing, aiming, maintenance and access Safety from a generic family statement

What are the downsides of using flood lights?

Flood lights can produce glare, spill, skyward light, intrusive brightness, uneven facade patterns, visible color separation, control complexity, maintenance-access risk, and unwanted exposure for nearby receptors when geometry and operation are poorly controlled. Colored systems add channel, scene-file, dimming, optical-mixing, and integration variables. These are not reasons to reject every flood light; they are reasons to aim, shield, schedule, dim, mock up, and document the exact system. Use the lowest output that achieves the defined task and review every receptor.

Use the Color-Flood Proof Chain: 6-Part Evidence Spine

Color-Flood Evidence Spine from claim through acceptance record.
Color-Flood Evidence Spine from claim through acceptance record.

Even a thick submittal can be weak if no reviewer can trace a claim to the exact model, configuration, method, issuer, date, result, limitation, and project condition. Build the Color-Flood Evidence Spine from the six evidence layers below. Those glare, spill, and exposure risks need the same common review order, and each higher layer answers a different question without erasing the limits of the earlier one.

  1. Manufacturer declaration: identifies the claimant and what the supplier says. Record the legal entity, model and revision. It is not independent verification.
  2. An exact-model report or certificate links a stated method, issuer or laboratory, date, configuration, result, and scope. A logo or family certificate without linkage is insufficient.
  3. Photometric and color data: describes measurable output under stated electrical, thermal, optical, and operating conditions. It does not reproduce the building.
  4. A physical sample checks build, connectors, markings, basic operation, and requested documentation. It does not validate full mounting geometry.
  5. Representative mock-up: tests selected scenes, surface, mounting, aiming, viewpoints, control chain, spill, and critical receptors under agreed conditions.
  6. Finally, a commissioned acceptance record binds the delivered configuration and scene revision to evidence, exceptions, owners, and retest requirements.
Filled example: “Uniform pale-blue scene on Zone A” is the claim. First, the declaration names fixture model/revision and optical option. Next, the report identifies the tested configuration and color/photometric method. A sample confirms markings and channel map. During the mock-up, the project team uses the surface, mounting, aiming, viewpoints, and scene file. Finally, the acceptance record names the scene revision, evidence files, criterion, verdict, deviations, and signatories. No layer promises performance outside its recorded conditions.

A protocol label moves data, a beam angle describes distribution, and an IP code grades an enclosure; none of the three approves a commissioned colored facade. That approval emerges only when evidence and project conditions converge at one traceable hold point.

For transparency, the public GQLAMP company background contains manufacturer-reported company and capability information. Treat it as first-party context. Don’t infer facility ownership, certificate coverage, laboratory independence, or applicability to a specific product unless the underlying exact-scope evidence is supplied and reviewed.

Run the Scene-to-Evidence Ledger as a Commissioning Dossier Before Acceptance

Scene Commissioning Dossier fields for repeatable acceptance.
Scene Commissioning Dossier fields for repeatable acceptance.

Once the proof layers are separated, the project still needs one shared acceptance record. A controller scene name is not an acceptance record. The Scene Commissioning Dossier lets the designer, supplier, installer, controls integrator, quality lead, and owner refer to the same physical and digital state. Use one row per critical scene and keep file names and revisions immutable after sign-off.

Scene Commissioning Dossier, eight reusable fields with a filled example
Point Filled example Your project field
1. Scene ID SC-03 / Pale Blue / revision 2 Unique name, number, purpose, revision
2. Exact configuration Fixture/optic/driver/controller identified in approved submittal Models, revisions, firmware/personality, addresses
3. Channel and dimming state Recorded channel values and transition file; no generic percentage claim Values, curve/mode, fade, steady-state duration
4. Zone and geometry Zone A, mounting/aiming coordinates, surface sample reference Zone, throw, setback, angle, surface, overlap
5. Viewing and ambient conditions Primary plaza view plus neighboring-road view, date/time and context Viewpoints, weather, ambient light, receptor conditions
6. Evidence file Photographs, measurement file and observer record under one revision File names, method, instrument, calibration/status
7. Criterion and verdict Project-defined optical, control and spill checks: pass with one noted exception Tolerance, pass/fail, exception, corrective action
8. Owner and retest Designer/owner sign-off; retest after scene, driver or fixture replacement Names/roles, date, revision, trigger, status

The cited Pacific Northwest National Laboratory outdoor-lighting report emphasizes clear project goals, qualified input to requests for proposals, mock-ups or pilots, and evaluation of affected surroundings. This ledger turns that general method into a colored-facade record; it isn’t a standard-mandated form.

Acceptance is not permanent proof. A 2019 United States Department of Energy update on accelerated stress-testing reports that aging effects varied by multi-source product architecture and affected luminous flux, chromaticity, and tuning range in the tested products. That doesn’t predict a specific fixture’s drift. It does justify project-defined revalidation after a fixture, driver, optical, firmware, controller, or scene-file change; after visible color mismatch; and at an owner-selected maintenance interval.

Send an RFQ That Exposes Missing Evidence

Colored flood light RFQ deviation categories and evidence checks.
Colored flood light RFQ deviation categories and evidence checks.

Comparable quotations require the same application boundary, evidence deliverables, and hold points. Copy this table into the request for quotation, replace “project-defined” with your adopted values, and require suppliers to list every deviation. Any missing value should remain visibly open; don’t fill it with a universal web recommendation.

RFQ checklist — copy these into your quote request:

Parameter Recommended range Why it matters How to verify
Application and scene set Project-defined zones, purposes, critical scenes and operating schedule Prevents a demo effect from replacing the actual lighting task Scene schedule, receptor map and owner-approved design brief
Fixture/channel configuration Exact model/revision; declared RGB or RGBW channel map and white-channel purpose Stops family labels from hiding a different delivered build Exact manual, submittal, markings and physical sample
Optics and geometry Project-defined optic, mounting, aiming, throw, setback, surface and viewpoints Controls mixing, overlap, spill and observed facade pattern Configuration-linked distribution data and representative mock-up
Control interfaces Named protocol/edition, personality, addresses, topology, loss and recovery states Assigns integrated behavior instead of assuming compatibility Interface schedule, approved scene file and witnessed tests
Electrical/environmental scope Site supply and protection design; exact enclosure evidence plus installation conditions Separates product evidence from installed-system water and electrical risk Reports/instructions, electrical review, delivered-state inspection
Temporal light and safety Project-defined scenes, levels, users/cameras, market scope and exposure conditions Prevents generic “flicker-free” or safety claims Configuration-linked measurements and exact-model safety assessment
Acceptance and lifecycle Project-defined mock-up criteria, ledger fields, sign-off, deviations and retest triggers Makes the scene reproducible after handover and maintenance Completed ledger, punch list, commissioned record and revalidation plan
Translate catalog and search wording before comparing quotes

Fixture and family labels: bulb, light bulbs, floodlight, led flood light outdoor, outdoor led flood, rgb flood light, rgb led flood, rgb floodlight, smart rgb floodlight, light fixture, spotlight, security light, adjustable, built-in, waterproof flood light, and ip66 waterproof.

Control labels: timer, bluetooth, bluetooth smart, wi-fi, wifi, music sync, strobe, landscape lighting with app control, color changing light, rgb color changing, color changing uplight, and rgb lights.

Use-context labels: landscape lighting, landscape, rgb landscape, patio, outdoor garden, large areas, stage lighting, uplighting, indoor, indoor outdoor, indoor ambient lighting, atmosphere, decoration, decorate, holiday lighting, halloween, and diy.

Color and output labels: multi-color, multicolor, blue led, orange light, daylight, 5000k, lumen, 1200lm, 30w, 300w, 120v, low voltage, 180°, 16 million color, 16 million, million colors, and million colors 23 modes.

Pack and comparison labels: 2 pack, 4 pack, pack 50w, 800w equivalent, 300w equivalent, energy efficient, and Govee. Treat every phrase in this list as search or catalog wording to normalize—not as an accepted specification, equivalence, brand recommendation, or proof of availability.

GQLAMP’s current flood-light page presents commercial white-light flood specifications in its public structured content; the reviewed page didn’t independently establish an RGB or RGBW option. Use RGB and colored LED flood lights as the commercial handoff, then ask GQLAMP to confirm current exact-model availability, channel configuration, optics, controls, documentation, and project fit in writing.

How do I choose a flood light?

Choose a flood light by starting with the task and installed geometry, not wattage or a color rendering. Define the zones, surfaces, mounting and aiming, viewpoints, spill boundaries, color scenes, operating schedule, control interfaces, supply and environment. Then compare exact-model optical, electrical, enclosure, color, temporal-light, and safety evidence. For a color-changing system, include the channel map and scene-file ownership. Finish with a representative mock-up, commissioned acceptance record, deviation process, and maintenance revalidation trigger.

Turn the concept into an evidence-ready inquiry

Send the project elevation or site photos, mounting and surface geometry, viewing goals, critical scenes, control architecture, electrical conditions, market, receptor constraints, and required evidence. Ask for gaps and deviations to be identified before a sample or quotation is approved.

Discuss Your Colored Flood-Lighting Project

Frequently Asked Questions

These short answers preserve the configuration and project boundaries used throughout the guide.

What kind of flood lights can change color?

Flood lights change color when the exact fixture system provides multiple controllable channels, commonly RGB or RGBW, with a compatible driver or decoder and controller. Confirm the model, channel map, personality, topology, scenes, dimming behavior, and optical configuration. Then test the intended colors at representative mounting, throw, surface, and viewing conditions. A product-family label or rendered image cannot prove the installed result.

Is RGBW better than RGB for facade lighting?

Not automatically. RGBW adds a dedicated white channel, but that does not prove better white, smoother dimming, higher consistency, clean optical mixing, or compatibility. Define why the white channel exists, how it is mapped and blended, and which scenes use it. Compare exact configuration data and run a representative mock-up. RGBW is useful only when the extra channel supports a documented project requirement.

What must be tested before approving colored flood lights?

Test critical scenes, not only full-output red, green, and blue. Record exact fixture and control configuration, channel and dimming states, mounting and aiming, facade surface, throw, viewing positions, ambient conditions, color-uniformity observations, spill, relevant temporal-light behavior, and signal-loss response. Tie results to criteria, evidence files, deviations, and accountable sign-off. A sample checks the product; a representative mock-up checks the proposed system in context.

Does DMX512 compatibility guarantee that every component will work together?

No. ANSI E1.11 provides a communication framework, but project operation still depends on personality, channel mapping, addressing, topology, equipment limits, supported features, power, scene data, and loss behavior. Require an interface schedule and witnessed tests for the exact hardware and scene-file revision. “DMX compatible” is the start of an integration question, not plug-and-play proof.

Does IP66 mean the installed lighting system is waterproof?

No. An IP code classifies enclosure protection under defined IEC 60529 tests. It does not by itself validate the completed installation, connectors and cable glands, orientation, drainage, condensation control, cleaning exposure, corrosion environment, workmanship, or maintenance. Check exact-model evidence and instructions, then review site water paths and environmental conditions separately. Record who supplies and seals every connector, how entries are oriented, where water drains, which cleaning methods are permitted, and how seals will be inspected. Avoid transferring one enclosure rating to the whole assembly or using “waterproof” as a blanket system claim.

What color light is best for outdoor lighting?

There is no universal best color. Start with purpose, local rules, surface response, nearby people or habitat, and operating hours. Justify dynamic colors scene by scene, then verify aiming, dimming, schedule, spill, and affected receptors.

References & Sources

  1. CIE S 025/E:2015 — Test Method for LED Lamps, LED Luminaires and LED Modules — International Commission on Illumination.
  2. IP ratings — enclosure protection under IEC 60529 — International Electrotechnical Commission.
  3. IEC 62471-7:2023 — Photobiological safety of light sources and luminaires — International Electrotechnical Commission.
  4. DMX512-A has been revised — Entertainment Services and Technology Association.
  5. Flicker Research — United States Department of Energy.
  6. Consolidated Commission Regulation (EU) 2019/2020 — EUR-Lex.
  7. Lighting and Power Upgrade Recommendations for U.S. National Park Service Caribbean Units — Pacific Northwest National Laboratory.
  8. Complex-facade lighting design case — Buildings, peer-reviewed research.
  9. Temperature Compensation for Chromatic Stability of RGBW LEDs in Automotive Interior Lighting — Electronics, peer-reviewed research.
  10. Five Principles for Responsible Outdoor Lighting — DarkSky International and the Illuminating Engineering Society.
  11. Update on Stress-Testing Results for Multi-Source LED Lighting — United States Department of Energy.