Deck Builder Feedback✨

[whole phoenix]

We’ve pushed a new build with some bug fixes and optimisations, but we’d love to hear how the Deck Builder’s working for you.

Let us know what you think and how we can keep improving it - drop your feedback below. Thanks!

[floral ore]

🦜

anyways i think the deck builder is great for newer players who want a decent deck to start off with

[slow wagon]

Improve its performance, it's somehow laggier than round 300 BTD6 with multiple paragons and a max vtsg

[viscid otter]

I still think decks need longer names, deck builder could also do with a naming system of its own that names the deck for you in refrence to the cards in the deck

[shy rampart]

make it work for cards you dont own, currently doesnt benefit newer players at all who need the most help with deck building

[slow wagon]

This is a nice idea

[dusk sparrow]

I’d love if PPE was a little more accurate, like max PPE, min PPE for conditional things. Also, can we please have sorting options, like sort by PPE, sort by interactions, etc. Lastly, I’d love if it worked for cards you don’t own too

[worthy basalt]

I dk why but I think it's a bit unfair

[tender turret]

I agree with this one in particular. Sometimes the game just says I don't have any cards (and that's actually impressive!). And it happened quite often

[slow wagon]

Yeah, that happens often to be too

[smoky finch]

I'm new to the game, I'd say it helps a lot in seeing which cards complement eachother, but I've seen for example when mixing decks with cards of different categories, it mostly offers cards from the ones you have the most, I think it would be nice to have suggestions to create sinergy between different types of cards, but idk if its just me 😅
Also sometimes the deck builder freezes loading cards and I have to exit and enter again, specially right after selecting a card

[coarse tinsel]

I'd have to agree with this, as well. I put Eternal Flame in there and was hoping to get burn/lock suggestions, but it was mostly science boosters.

Perhaps rank could be somehow determined by winrate of certain card presences in the game? Like if out of all Eternal Flame Decks, Loch Ness has the highest win rate, that's card #1. Or at the very least, emphasis on stacking certain mechanics. Also, I tried using deck builder for Harper Lee (finally decided to try Events) and... it... was not building anything decent... i ended up lumping her in with sea birds and just running a 17-card deck for five games.

Also in favor of unowned cards being shown.

imo a key distinction with permitting unowned cards is also being able to manually disable it and show only owned cards

[ruby falcon]

Hi, first things first, need to give props where it is needed. The new tool is really nice and works much better than the old one the game had(autofill).
I did a few tests with the tool yesterday and tdy (since there was an update). While from the looks of it, the tool got better, it still faces one issue that I saw. While recommending cards the tool doesn’t really update the logic until the deck is builded in certain scenarios. The example that I tested in many scenarios is with the card “paper”. I basically choose the card and used the assist mode to pick the 17 other and while I selected the cards the game makes some mistaken suggestion: like recommending different collection based cards, in my case it recommended Bothriolepis and 4 cards later it would say that the next possible best card was Sea People. Once the deck is fully builded with 18 cards if I do a card by card review it will recognize that sea people/ bothri are some of the worst cards possible, but before completing it will recommend both as they are the best together.
So for feedback there could maybe exist a clause where once a collection card is placed in deck, other collection cards with the lower or exact values that don’t match be harshly evaluated.

[coarse tinsel]

yeah! this guy^ just reminded me of something i had wanted to say. last night i was thinking how this seems more like a deck finishing tool than a deck building one. it's plugged as "put in a card and assist away" but it really does require you to start the deck, yourself.

just having some kind of idea of what makes the cards work would make the deck builder work better; like, instead of being like "oh, this card's a legendary, let's recommend a card that boosts legendaries," if the builder could envision potential decks - for example, on lichess, there is a thing called the Opening Explorer. When you choose a move, it has other suggestions and tells you what opening you're building. If here it was like "hey, you chose Pumpkin Toadlet, your best bet is probably either tiny deck or Amphibians," that would be dope.

(wanted to add, i am noticing others mentioning an update, not sure I've had a chance to experience that)

[worthy basalt]

I didn't try it yet, I deleted the game just until I done with high school
So I can't judge at all
I think it'll help me cause am terrible at this side but I still think it's unfair

[ruby falcon]

Just for a more detailed view, this is what the deck ends up being with the order of suggestion not changes (it is already much better than yesterday and much more logic, but still doesn’t identify the collection issue until the deck is fully builded).

[lilac urchin]

Id appreciate deck builder mode defaulting to being off whenever you are building a new deck. It is a little jarring whenever I'm building a new one by collection, select my first card, then having to backtrack when the first card triggers assist mode because I left it on last time.

[lyric isle]

It seems to prioritise number of affected cards over power of the effect. This is not bad in theory, but in practice it loves to add weak stuff like principia mathematica and hope diamond. Also for some reason it considers autotune terrible. And it likes to put in both statue of liberty and jargon file, even when they don't make sense together.
More generally, it would be nice if it knew which cards were added manually, as currently if you don't start with enough cards it will often disregard them entirely and spit out some generic rainbow pile.

[red copper]

Hi, I think that the new building assistant is overall quite nice, and much better than the previous autofill. But from what I’ve read so far, most people are still trying to use this tool as an autofiller and build decks solely with those cards. Instead I find that the tool helps much more once the deck is near completion or after one has tried their deck and feels that one or two cards need swapping because they’re underperforming. In this scenario the recommendations are much better, and one is supposed to browse through those because they’re still not perfect or smart at all times, and for this reasons there are many different ones and one shouldn’t just pick the top one.

And I do agree that sometimes it struggles with albums-collections and makes the game laggy, especially in battery saver mode. And that it should recommend not owned cards, especially for newer players who are not familiar with the near 6k cards.

What instead I wish it could be implemented is an additional tool where one can specify the type of deck that they’re aiming to build (like lock-burn, vitcon, perm scaling, sharks, mono space etc…). This way, the suggestion can be much more tailored to one’s liking and intentions and can overall create a much better performing assistant

[viscid otter]

Devs I need my deck to be named "immortal ten direction realm ancient true Dao cultivation reversal supreme true loong technique"

[ashen harness]

same thing i did with harper and the birds, it doesnt feel like theres even a real reason or mechanic its attempting to teach you with this adventure.

[coarse tinsel]

i really like that last paragraph's suggestion.

[coarse tinsel]

worth noting there is one other write stuff card which addresses penguins. perhaps the purpose is to teach new players that just because a deck has cooperation, doesn't mean that it's going to work/is good.

[slender pebble]

I will mention some negatives since it will act as constructive feedback:

1. Cards with generic abilities which might enhance a certain type of deck are not ranked high

Example: Goatman improves decks where an album has exactly 9 cards. So ideally this should be ranked high when I already have 9 cards from one album, 8 cards from the album Goatman belongs to. However, I found it was ranked 1400 out of 1800 cards I have.

2. Maximum impact possible should be displayed and it should affect recommendations.

For example, some card which is enhancing Chinese Folklore cards by a combined of 50 points across all cards in the deck is being ranked higher than Nazca Lines which has a higher point impact on the entire deck for the same deck. Just because it does not primarily pertain to the album or collections I am using, it should not be punished.

So while impact usually stacks up for similar albums or collections, it is better to know the maximum possible impact from a card so that we can create decks where we need at least 1 card from a certain collection or album which might help our deck overall but may not have stacked impact since only 1 card of that album or collection is present in our deck

[slender pebble]

Hey so actually to execute the suggestion that you just mentioned, try using the card filters along with the auto assist. So let's say I have a certain type of deck. I only want to add 3 History cards. So after I put the auto assist as on, I use the card filters and only the History related suggestions are present. You need not necessarily rely on the rankings there but this way, you can use your knowledge or requirements and get the job done

[red copper]

I agree that the filters can offer a great help, but what I’ve found out experimenting is that certain types of decks, such as ppt, lock, burn, and other more complex ones, cannot rely solely on the filters but require external cards like SOL, Jargon, and overall strong standalone cards that will complement the aspects that the deck would normally lack (like any strong burst in ppt, or “combo” cards like mark of zorro to pair with lock)

Even more importantly, one should be able to specify the energy league requirements for the deck, since it never suggests ept cards to match with more expensive decks

[simple forge]

Great tool. My overall feeling is that it favors “number of cards affected” or “affected by” metrics. I have a sharks deck and I’itoi, which has a great impact over several cards, but is ranked below Pythagoras of Samos.

Doesn’t make much sense to me

[covert chasm]

i feel like the deck builder should a toggle to include cards not currently in ur collection. right now its annoying to plan a deck for an unowned collection

[wicked talon]

If you turn on the assist mode it download really slow

[lofty ermine]

we went from paper to game dev, what happened

[lofty ermine]

this, it would be helpful if after the first 3 cards, the deck builder would actually form a deck archetype

[solemn grove]

Little bug with this feature: if you build a new deck and use an already existing deck for it where you just removed all the cards (and not as in when you click on new deck and start with an empty deck), all the removed cards are still taken into consideration.
See the screenshot: it says on Cosmic String that I have 21 a&c cards which is of cause impossible, but that is because I had a full a&c deck prior to it that I removed all cards from.
Hope that it's somewhat understandable.

[viscid otter]

Yo? John Henry is the guy from teen Titans Go, no?

[coarse tinsel]

no he's like a hammer guy

tunnelmaker