Jump to content
Pod Your Reef

Cancelled Comparison


HarryPotter

Recommended Posts

HarryPotter

Head to Head: NanoBox Duo vs Radion

 

Same tank. Same parameters. Two Lights.

 

The Question: Which light grows an identical sized (and origin) piece of Montipora cap faster and with better coloration?

 

I will be putting the Radion on one side of the tank and NanoBox on the other. A small frag of coral will be placed on the sand bed (Slightly raised) for a month or so to compare growth

 

The Lights:

 

NanoBox Duo with V3 Pucks (Newest, includes lime)

 

 

 

Nano box Duo par readings (chart is mislabeled, this IS duo)

 

 

 

Radion XR30w Pro Gen 2 (Not newest model)

 

[Details below are for XR30w Pro Gen 3. Contacted Ecotech for Gen 2 chart]

Link to comment
HarryPotter

Same PAR?

In the process of getting par charts. Nano box had the quad listed on the duo page, and I need to get the Gen2 Pro par from Ecotech

Link to comment

I would suggest putting an opaque partition between the two sections of the tank (above and below water) so that both corals are isolated from the others light source. It will take away any variance that may occur from light bleeding over to the other side of the tank.

Link to comment
HarryPotter

I would suggest putting an opaque partition between the two sections of the tank (above and below water) so that both corals are isolated from the others light source. It will take away any variance that may occur from light bleeding over to the other side of the tank.

Hmm. I see what you're saying but this IS my display tank and I only have one Vortech MP40.

 

There will be a slight change: the NanoBox is going to be a Trio. I will update the information above accordingly

Link to comment

Hmm. I see what you're saying but this IS my display tank and I only have one Vortech MP40.

 

There will be a slight change: the NanoBox is going to be a Trio. I will update the information above accordingly

 

This would be nice but I'm thinking they are going to interfere with each other as far as the spread goes no?

Link to comment

why not use one light for a month and than switch to the other light for a month? and not use them simultaneously. that seems like it would be the most accurate test.

Link to comment
masterbuilder

I look forward to what happens. If I were a betting man I know which one I would put my money on.

 

p.s. The bigger mystery will be what the post count is on how to run your test. B)

Link to comment
12_egg_Omelette

why not use one light for a month and than switch to the other light for a month? and not use them simultaneously. that seems like it would be the most accurate test.

Negative ghost rider.

Link to comment

Hmm. I see what you're saying but this IS my display tank and I only have one Vortech MP40.

 

There will be a slight change: the NanoBox is going to be a Trio. I will update the information above accordingly

Then just place the separator above the water line to minimize any light spill. If you want to make this an accurate test, you have to have some kind of separation.

 

why not use one light for a month and than switch to the other light for a month? and not use them simultaneously. that seems like it would be the most accurate test.

Not really. You influence the growth and coral with one light, then change to another light. You can't see how well the coral reacts from that point because you have changed the way the coral grew from the original sample.

Link to comment
12_egg_Omelette

In reality if you want to have the most reliable study, you need to have more then one replicate to control for erroneous data from maybe dips in trace elements, high temperate or any other issues. What you're really wanting to do is an ANOVA with a Dunnett's post hoc test set to a control.

 

I agree with Evil, on his explanation of why switching would not be ideal. Also I would look at recording pre circumference and heigh, post circumference, and delta (post-pre).

 

The trouble comes into how you want to set up the study design, I think the issue comes into low replicates so a T-Test with no control might be the better route when you look into that, also if you have unequal samples you'll need to do a non-parametic test, you also have to look into equal distribution or goodness of fit for the data.

Link to comment

You won't be able to do any statistics that are meaningful on a single replicate of a single sample. You also have no controls to compare to.

 

I like your idea, but unless you are going to go all the way, I'd say just see which light ends up yielding happier coral. It's not definitive and there are LOTS of variables that could sway your data without you thinking about it.

Link to comment
12_egg_Omelette

You won't be able to do any statistics that are meaningful on a single replicate of a single sample.

 

If you can repeat the experiment 4-6 times it would be meaningful, especially since the variance between sample is low since it's a clone. Research in inbred mice can have an N=6 with a power analysis.

Link to comment

 

If you can repeat the experiment 4-6 times it would be meaningful, especially since the variance between sample is low since it's a clone. Research in inbred mice can have an N=6 with a power analysis.

Right, but that's multiple replicates. Has harry proposed to do multiple replicates for months each? I didn't think so?

 

There are some great statistics OP could use if he were to carry out the experiment fully, but there are a lot of variables that need to be taken into account for that, and I doubt OP wants to do that (or has the time/money).

Link to comment
12_egg_Omelette

Right, but that's multiple replicates. Has harry proposed to do multiple replicates for months each? I didn't think so?

 

There are some great statistics OP could use if he were to carry out the experiment fully, but there are a lot of variables that need to be taken into account for that, and I doubt OP wants to do that (or has the time/money).

 

Well Harry needs to decide if he wants to be a man of science also known as the keepers or truth ;) The bio-statistician in me actually started writing out a model (FML right).

Link to comment

 

Well Harry needs to decide if he wants to be a man of science also known as the keepers or truth ;) The bio-statistician in me actually started writing out a model (FML right).

I was initially writing a post about how to design multiple tanks with all the correct controls and such, lol.

 

Biostats! Awesome - random, but do you recommend any good books for that? I took a biostats class in grad school but it was abysmal and I learned virtually nothing. I'd love to be better versed in biostats for work.

Link to comment

Unless there's total separation of light both frags will be influenced in a similar way as using one light for a month and than the other. Basically not accurate.

Harry just setup another tank. Plumb them together. Problem solved.

Link to comment

Unless there's total separation of light both frags will be influenced in a similar way as using one light for a month and than the other. Basically not accurate.

Harry just setup another tank. Plumb them together. Problem solved.

 

Thats what I was thinking

Link to comment
HarryPotter

I would select the same kelvin temperature on each

Probably, and get as similar intensity as possible

 

I look forward to what happens. If I were a betting man I know which one I would put my money on.

 

p.s. The bigger mystery will be what the post count is on how to run your test. B)

 

I dunno which honestly. I love my Radion, but I've heard a lot from NanoBots.

 

Then just place the separator above the water line to minimize any light spill. If you want to make this an accurate test, you have to have some kind of separation.

 

 

Not really. You influence the growth and coral with one light, then change to another light. You can't see how well the coral reacts from that point because you have changed the way the coral grew from the original sample.

 

I can make a separation that goes only 1" into the waterline and up to the lights- will that work?

 

Right- that's why half and half with 2 identical samples is my idea

 

 

In reality if you want to have the most reliable study, you need to have more then one replicate to control for erroneous data from maybe dips in trace elements, high temperate or any other issues. What you're really wanting to do is an ANOVA with a Dunnett's post hoc test set to a control.

 

I agree with Evil, on his explanation of why switching would not be ideal. Also I would look at recording pre circumference and heigh, post circumference, and delta (post-pre).

 

The trouble comes into how you want to set up the study design, I think the issue comes into low replicates so a T-Test with no control might be the better route when you look into that, also if you have unequal samples you'll need to do a non-parametic test, you also have to look into equal distribution or goodness of fit for the data.

 

Eek I didn't realize it would be so complicated. I'm thinking of the parameters of the water are the same for both samples, the lighting would be the only variable. I can have a few samples though- like 3 of each coral on each side?

 

You won't be able to do any statistics that are meaningful on a single replicate of a single sample. You also have no controls to compare to.

 

I like your idea, but unless you are going to go all the way, I'd say just see which light ends up yielding happier coral. It's not definitive and there are LOTS of variables that could sway your data without you thinking about it.

Sure- that's my idea. Which light grows coral "better" (faster/more colorful) in the same parameters

 

 

Right, but that's multiple replicates. Has harry proposed to do multiple replicates for months each? I didn't think so?

 

There are some great statistics OP could use if he were to carry out the experiment fully, but there are a lot of variables that need to be taken into account for that, and I doubt OP wants to do that (or has the time/money).

Well how about using 3 of each coral on each side?

 

Didn't mean for it to be a very long term thing, just a month or two. I don't have the $$ to setup a more intensive study

 

 

Well Harry needs to decide if he wants to be a man of science also known as the keepers or truth ;) The bio-statistician in me actually started writing out a model (FML right).

 

Haha I'd like to, but I can't plumb in another system or completely seperate the two areas. Someone PMd me about letting me borrow their PAR meter- that might be interesting as well.

 

Unless there's total separation of light both frags will be influenced in a similar way as using one light for a month and than the other. Basically not accurate.

Harry just setup another tank. Plumb them together. Problem solved.

Haha I wish! No way would my parents let me plumb in another. They are unhappy enough already with the half hour a day I spend looking at or fiddling with the tank

Link to comment

Archived

This topic is now archived and is closed to further replies.

  • Recommended Discussions

×
×
  • Create New...