Using Quanta Index to determine the optimal number of Pillars in a deck

[Essence]

Good idea, Vyrys.  An excellent post given that it's among your first 4. 8)

[RootRanger]

I find that while large decks usually have a higher QI, deck size is not the factor that causes larger decks to want a QI. It is instead, it is the deck's amount of cards it wants to play as soon as they are drawn. If your deck is a rush with a lot of creatures, it will want a lower QI than average so it can play its cards as soon as they are drawn. However, a stall will often not want to play it's cards as soon as they are drawn. Thus they will wait longer to want to play the card, perhaps when they opponent has a creature for them to kill. Which means they will get more quantum to play that card during the time they wait. Rushes need a low QI; stalls need a high QI. So what does this have to do with deck size? Rushes are usually 30 cards; stalls are often larger. QI should be balanced based on speed instead of size

[UTAlan]

Has anyone started working on a tool to calculate QI? I'd like to, but I don't want to waste my time if someone else is already doing so. I skimmed this thread and didn't see anyone talking about one. I'll go back and read it more carefully, but thought I'd ask in the mean time. 

[Xinef]

It seems that all publicly declared attempts to write a tool are currently on hold. Anyway, you might be interested in this thread:
http://elementscommunity.org/forum/index.php/topic,5672.0.html

If I remember correctly, I declared my attempts too at some point, but a higher priority project appeared soon enough so I had to stop pretty much as I was still designing the GUI. Puppychow's attempts are probably more advanced, but I haven't seen him for quite a while.

[UTAlan]

I may need to start my own thread for this, but for now this seems an appropriate place to put it:

Quantum-Index.com (http://Quantum-Index.com).

I'm open to constructive criticism, so let me know what you all think. I still have yet to read this thread in it's entirety, so there is certainly room for improvement, but this is a good starting point. As of now, all cards have been added and I did my best to accurately measure the quanta cost/gain per card. I'm hoping to put this info on the site soon, but for now, I put a spreadsheet online (https://spreadsheets.google.com/ccc?key=0AtIJScFz1sS7dEtDbnpacjFFTDZCZ2NwMW1zQzBxaVE&hl=en).

[pikachufan2164]

I may need to start my own thread for this, but for now this seems an appropriate place to put it:

Quantum-Index.com (http://Quantum-Index.com).

I'm open to constructive criticism, so let me know what you all think. I still have yet to read this thread in it's entirety, so there is certainly room for improvement, but this is a good starting point. As of now, all cards have been added and I did my best to accurately measure the quanta cost/gain per card. I'm hoping to put this info on the site soon, but for now, I put a spreadsheet online (https://spreadsheets.google.com/ccc?key=0AtIJScFz1sS7dEtDbnpacjFFTDZCZ2NwMW1zQzBxaVE&hl=en).

Very nice

There are a few things that aren't quite correct with the calculation, though.

1. The mark is not factored into the calculation. It should count as a pillar for the element in question (the lack of a mark consideration messes up calculations involving pendulums).

2. Elements are not QI'ed separately. Each element should have its own QI analysis done (i.e. if I had a / deck, then I would need to QI the portion, and then do a QI on the portion).

3. Ability activation costs are not done properly. They should only be accounted for if that element is actually usable in the deck. Also, there may be a need to count ability costs once (if it's situational), twice (if it's advantageous to use every turn), or none at all (if the ability isn't meant to be used for the deck in question).

I'm not entirely sure on the QI calculation of cards like Quantum Tower, Nova, Supernova, Fractal, Miracle, and Sky Blitz -- some thoughts on how to calculate those should hopefully be in this thread somewhere.

[UTAlan]

1. The mark is not factored into the calculation. It should count as a pillar for the element in question (the lack of a mark consideration messes up calculations involving pendulums).

This has been added. I meant to do this before posting here, but forgot to go back to it. 

2. Elements are not QI'ed separately. Each element should have its own QI analysis done (i.e. if I had a / deck, then I would need to QI the portion, and then do a QI on the portion).

This is something I intend on doing soon. I built my data around that concept, as I am already storing the quanta cost/mark to play the card, as well as the cost/mark for the special ability. This is fairly trivial for cards like Lava Destroyers, but gets more complicated for cards like Leaf Dragons.

3. Ability activation costs are not done properly. They should only be accounted for if that element is actually usable in the deck. Also, there may be a need to count ability costs once (if it's situational), twice (if it's advantageous to use every turn), or none at all (if the ability isn't meant to be used for the deck in question).

This is a little daunting, but it can be done. (If quanta generation for a particular mark is 0, ignore its costs.) It'll probably look more feasible once (2) is done.


Thanks for the feedback!

[RootRanger]

I've always counted the mark as 2 pillars because it generate about as much quantum as 2 random pillars. 3 if you have a low TTW.

[Crimsonthorn]

Cards: 37
Total Cost of Cards: 110
Total Cost of Abilities: -25
Number of Pillars: 2.0
QI

Your QI is 42.5, which means you have too few pillars.

6 RoL, 6 Cremate etc

[Jangoo]

I may need to start my own thread for this, but for now this seems an appropriate place to put it:

Quantum-Index.com (http://Quantum-Index.com).

I'm open to constructive criticism, so let me know what you all think. I still have yet to read this thread in it's entirety, so there is certainly room for improvement, but this is a good starting point. As of now, all cards have been added and I did my best to accurately measure the quanta cost/gain per card. I'm hoping to put this info on the site soon, but for now, I put a spreadsheet online (https://spreadsheets.google.com/ccc?key=0AtIJScFz1sS7dEtDbnpacjFFTDZCZ2NwMW1zQzBxaVE&hl=en).

Please do start your own thread for this ... otherwise it will be lost instantly.

Great work so far!

It does seem to have trouble recognizing what the deck actually is in terms of main-quanta and mark-quanta:
I constantly get different arrangements for duos and trios ... sometimes the actual mark is right in front, sometimes
in the back, sometimes a third "mark" is displayed which plays absolutely no part in the deck. Then, sometimes
mark quanta (for the actual mark) are calculated as 5 and sometimes not.

[YoungSot]

Please do start your own thread for this ... otherwise it will be lost instantly.

Yeah, you should do this when you get a chance. It seems to me that it's come along far enough to be a useful tool, and making it more visible means better feedback and more suggestions for improvement.

[UTAlan]

Quantum Index Tool (http://www.elements.alanbeam.net)

This is a tool I built based on this thread by SG (http://elementscommunity.org/forum/index.php/topic,5676). It is currently a work in progress, so please bear with me as I work out the kinks. I appreciate any and all feedback anybody can provide in order to help improve the tool. Please keep in mind that QI is used as a guide in deck-building. While a QI of 5.0 appears to be ideal in many cases, this does not guarantee your deck will have the ideal amount of quanta every match. I hope it can assist you in your deck-building, but don't rely on it to do all the work for you.

The Formula

QI

=

Deck Cost (Deck Quanta Generation / 5)

QI tells you how good your ratio of non-pillars to pillars is in a given deck.

Assumptions:

• 30 card deck
• A pillar, on average, generates 5 quanta a game
• A Mark behaves identically to a pillar
• There is not currently enough evidence to determine what the ideal QI is in a rainbow deck. (ANYONE?)

Notes:

• QI is calculated independently for each element in use in your deck.
• The observed ideal QI for a deck is 5.0
• QI of greater than 5 is not producing enough quanta, while QI of less than 5 is producing too much quanta.

Each card has 2 values pertinent to this formula:

• Cost
• Quanta Generation

Cost

Each card's cost is a sum of two values:

• The cost to play the card. This will always be a value of 0 or greater.
• The cost to play the card's special ability. This will always be a value of 0 or greater.

Quanta Generation

This is a value representing the average quanta generated in a game by the card.

Pillar
Cost: 0
Quanta Generation: 5 (of respective element)

Tower
Cost: 0
Quanta Generation: 6 (of respective element)

Nova
Cost: 0
Quanta Generation: 1 (for each element)

Supernova
Cost: 2 (Entropy)
Quanta Generation: 2 (for each element)

Immolation
Cost: 0
Quanta Generation: 7 (Fire), 1 (for each non-Fire element)

Cremation
Cost: 0
Quanta Generation: 9 (Fire), 1 (for each non-Fire element)

Devourer
Cost: 2 (Darkness), 1 (Earth)
Quanta Generation: 5 (Darkness)

Example Deck

Code: [Select]

61o 61o 61o 61o 61o 61o 61o 61o 61o 61o 61o 61o 61o 61o 61o 61s 61s 61s 61s 61s 61t 61t 61t 61t 61t 61t 61v 61v 61v 61v
This is The Essential Halfblood Farmer. To calculate the QI, we start by adding up to total cost of all cards in the deck.

Cost = (0 * 15) [Pillars] + (6 * 5) [Immortals] + (6 * 6) [Dim Shields] + (13 * 4) [Dragons] = 118

We then add up the total Quanta Generation of all the cards in the deck.

Quanta Generation = (5 * 15) [Pillars] + (0 * 5) [Immortals] + (0 * 6) [Dim Shields] + (0 * 4) [Dragons] + (5) [Mark] = 80

Throw these values into our formula and you get: 118 / (80 / 5) = 118 / 16 = 7.375

This deck is not producing enough quanta to fuel the deck, according to its QI. You can observe this while playing the deck by seeing that you are usually unable to play all the cards in your hand due to not having enough quanta.

• OTHER cards, such as SoG, have their cost divided up evenly over however many elements in which your deck is generating quanta. For example, in a duo deck with SoGs, the cost of SoG is (3/2) for both of those elements.
• If you have cards in your deck that generate unused quanta, the tool will show 0 cost, X quanta generation. The same goes for cards with costs from elements that are not generating any quanta. If you are okay with not using the extra quanta, or don't need to spend the quanta for that particular card (special ability?), you can ignore these results. Cards like Nova and Immolation will inevitably show every element.
• I documented all cards and their costs/quanta generation in a Google Spreadsheet (https://spreadsheets.google.com/ccc?key=0AtIJScFz1sS7dEtDbnpacjFFTDZCZ2NwMW1zQzBxaVE&hl=en) for those who are interested.

I hope I can continue to improve this tool until it is found useful enough for general use in deck-building of non-rainbow decks.