Using ChatGPT to design a light

MoshJosh

2500 Club Member
View Badges
Joined
Jan 27, 2022
Messages
3,528
Reaction score
3,897
Location
Grand Junction
Rating - 100%
1   0   0
For fun I asked ChatGPT to help me choose what ratios of what spectrums to use for growing corals.

Prompt was a 12 diode light.

With my "help" it came up with:
  • Blue: 3 LEDs (25%)
  • Violet: 1 LED (8-9%)
  • Royal Blue: 3 LEDs (25%)
  • Cyan: 2 LEDs (16-17%)
  • Red: 1 LED (8-9%)
  • Cool White: 2 LEDs (16-17%)
But what SURPRISED me was chatGPT's suggesting for a "better" set up/ration. . . it suggested:

  • Blue: 4 LEDs (33%)
  • Royal Blue: 3 LEDs (25%)
  • Cyan: 2 LEDs (16-17%)
  • Violet: 1 LED (8-9%)
  • Deep Red: 1 LED (8-9%)
  • Warm White: 1 LED (8-9%)
ChatGPT "thinks" the warm white and deep red will:

"
Warm white light, which includes a mix of wavelengths across the visible spectrum, can help provide a more natural lighting appearance and balance the overall color rendition in the tank. It can also enhance visual aesthetics and promote the growth of symbiotic algae (zooxanthellae) that live within the coral.

Deep red light in the range of 660-680nm has been shown to influence coral behavior, such as polyp expansion and coral fluorescence. While it may not directly contribute to photosynthesis, it can still have positive effects on certain coral species by stimulating specific pigments and enhancing their coloration."

what is everyones thoughts?
 

A_Blind_Reefer

Valuable Member
View Badges
Joined
Aug 13, 2019
Messages
1,783
Reaction score
2,389
Rating - 0%
0   0   0
1688774007224.gif
 

taricha

5000 Club Member
View Badges
Joined
May 22, 2016
Messages
6,547
Reaction score
10,108
Rating - 0%
0   0   0
Deep red light in the range of 660-680nm has been shown to influence coral behavior, such as polyp expansion and coral fluorescence.

That's not accurate. To excite fluorescence, the light needs to be shorter wavelength - higher energy than the fluorescent light emission. If
660-680nm excited any fluorescent pigments, the emitted light would be longer wavelength and thus too red to see.
 

flagg37

Custom stair builder - TreeofLifeStairs.com
View Badges
Joined
May 28, 2016
Messages
950
Reaction score
1,119
Location
Denver area
Rating - 0%
0   0   0
That's not accurate. To excite fluorescence, the light needs to be shorter wavelength - higher energy than the fluorescent light emission. If
660-680nm excited any fluorescent pigments, the emitted light would be longer wavelength and thus too red to see.
I seem to remember a study done with pocillopora that showed that red light inhibited growth.
 

oreo54

5000 Club Member
View Badges
Joined
Sep 18, 2017
Messages
5,603
Reaction score
3,449
Rating - 0%
0   0   0
I seem to remember a study done with pocillopora that showed that red light inhibited growth.

Correct but that was high par red and
" alone" or with a minimal " other" spectrum.
Sooo don't run 100% red light
 

Jekyl

GSP is the devil and clowns are bad pets
View Badges
Joined
Jan 15, 2019
Messages
11,539
Reaction score
15,879
Location
Michigan
Rating - 0%
0   0   0
That program doesn't think for itself. It scours the internet for articles and opinions. So maybe it's right or maybe it's some random guys opinion. Any second hand information is never going to be as good as first hand knowledge.
 

flagg37

Custom stair builder - TreeofLifeStairs.com
View Badges
Joined
May 28, 2016
Messages
950
Reaction score
1,119
Location
Denver area
Rating - 0%
0   0   0
GraphSpectrums_About-Photosynthesis_Chlorophyll-white.png


Around what wavelength does that big peak on the right of the chart look like to you?
It’s been a long time since I researched this topic. I first looked into it because I just gotten the newly released ap700 and wanted to simulate a sunrise/sunset. I programmed them to start and end with full reds.

What I think I remember was the case was that Zooxanthellae used chlorophyll A and C but not B. This is why the reds played no role in photosynthesis for this organism. I could be wrong though. Maybe @Dana Riddle has some
insight on this.
 

Hooz

Valuable Member
View Badges
Joined
Aug 16, 2020
Messages
1,468
Reaction score
1,541
Location
Heath, OH
Rating - 0%
0   0   0
It’s been a long time since I researched this topic. I first looked into it because I just gotten the newly released ap700 and wanted to simulate a sunrise/sunset. I programmed them to start and end with full reds.

What I think I remember was the case was that Zooxanthellae used chlorophyll A and C but not B. This is why the reds played no role in photosynthesis for this organism. I could be wrong though. Maybe @Dana Riddle has some
insight on this.

The magenta line on the graph is for Chlorophyl A. Big spike around 450nm, and another big spike up in the 670-680nm range.
 

oreo54

5000 Club Member
View Badges
Joined
Sep 18, 2017
Messages
5,603
Reaction score
3,449
Rating - 0%
0   0   0
It’s been a long time since I researched this topic. I first looked into it because I just gotten the newly released ap700 and wanted to simulate a sunrise/sunset. I programmed them to start and end with full reds.

What I think I remember was the case was that Zooxanthellae used chlorophyll A and C but not B. This is why the reds played no role in photosynthesis for this organism. I could be wrong though. Maybe @Dana Riddle has some
insight on this.


Perhaps the real story here is the photo-protective responses (the xanthophyll cycle) of zooxanthellae to different light sources, especially red light. Why is there little, if any, protective cycling of xanthophylls when strong red light is used? Strong doses of red light can regulate zooxanthellae densities even to the point of bleaching (Kinzie et al., 1984). What are the long term effects of red light on captive corals? This deserves investigation, and will be the subject of a future report.
 
Last edited:

oreo54

5000 Club Member
View Badges
Joined
Sep 18, 2017
Messages
5,603
Reaction score
3,449
Rating - 0%
0   0   0
Ignore,
Wrong place.
 
Last edited:

Lebowski_

Valuable Member
View Badges
Joined
Apr 22, 2023
Messages
1,339
Reaction score
2,440
Location
.
Rating - 0%
0   0   0
This is why when I see people call my tank "windex" on these online forums and socials, instead of being annoyed, I embrace it. My coral seem quite happy in all the blue, especially LPS.
 

Reefing threads: Do you wear gear from reef brands?

  • I wear reef gear everywhere.

    Votes: 30 15.8%
  • I wear reef gear primarily at fish events and my LFS.

    Votes: 11 5.8%
  • I wear reef gear primarily for water changes and tank maintenance.

    Votes: 1 0.5%
  • I wear reef gear primarily to relax where I live.

    Votes: 25 13.2%
  • I don’t wear gear from reef brands.

    Votes: 111 58.4%
  • Other.

    Votes: 12 6.3%
Back
Top