Decks for Quest mode

[Corwin72]

Sorry Chris and Corwin, you are putting a lot of effort into this, but I think it's really the wrong approach dumping 100+ decks into quest mode and fixing them afterwards.

Quality assurance is much easier, if Corwin (or others) will post only a few decks at a time and everyone can post concerns (about AI usability, balancing, existing similar decks, etc.) or suggestion for improvements.

Maybe we can use thread for new quest decks and have another for improving old decks.

This was originally posted as a testing add on. The decision was made to add it to the SVN so that more people would test it.

What should the policy be for the future when making decks for quest mode?

[Chris H.]

This was originally posted as a testing add on. The decision was made to add it to the SVN so that more people would test it.

What should the policy be for the future when making decks for quest mode?

`
Biggest lesson to be learned is probably to work on only a few decks at a time and to do it slowly.

[lazylockie]

I think you need to be careful with the tier 1 decks. Remember that early quest mode (the first 10-15 games) you'll be basically playing a sealed deck. While the starting Standard pool helps immensely (hello Titans!), it becomes a painful downhill when all the cards are available. You could get lucky and get a Maelstrom Pulse / Vindicate early, but chances are you'll be getting lots of Borrowing the East Wind.

I think as a rule of thumb, tier 1 decks can't have rares/mythics at all. For example, your Atomic Robo 1 has 4 All is Dust and 4 Brittle Effigy. While I agree having a singleton of both would give a fair chance for AI, 4 of each is just unfair for the player. Especially mass removal cards, such as Day of Judgment, Infest or Damnation, where it doesn't involve much thinking for AI.

`
You make a good point.

Would you mind toning down this deck for us. We could use another set of eyes and hands.

`

I'll take a look on the zipped file later. I've been also working on some tier 1 contributions, I'll post them here as well.

[Chris H.]

I'll take a look on the zipped file later. I've been also working on some tier 1 contributions, I'll post them here as well.

`
Thank you Lazylockie.

[Chris H.]

I updated and merged these decks:


Bear 2: AI can not use the mana ability of Nomadic Elf. Replaced them with Elvish Harbinger. Would have used Joiner Adept except that it no longer works for the AI for some reason.

Colbert 3: AI can only use the "Tap: Add 1 to your mana pool." ability of Cascade Bluffs. Not good enough. Replaced it with Volcanic Island.

Deadpool 2 and En Sabah Nur 2: The AI can use Red Sun's Zenith even though it has an X cost. I removed the SVar:RemAIDeck:True.


Does this mean that I should remove the SVar:RemAIDeck:True from the cards that have AF_DealDamage with an X cost?

[Chris H.]

I did not update these decks, a few decisions to make first:


Brood 2: This deck has Signal Pest, I guess that this will be OK if Sloth adds Battle cry to the combat code.

Cable 3: I ran a test and the AI failed to use Tezzeret, Agent of Bolas. I think that we should replace tis card, but with what?

Captain America 3: The AI will not activate the Qasali Pridemage ability to destroy an artifact. Not sure what to do. It is a 2/2 creature with Exalted for a cost of two. Not too bad.

Fin Fang Foom 2: The AI would not cast the Green Sun's Zenith for some reason. AF_ChangeZone may not yet have the code for X costs.

[Corwin72]

I am surprised at Fin. I could have sworn that I have seen it activate for 0 and get the land creature. That is why I left it in.

[Sloth]

Sorry Chris and Corwin, you are putting a lot of effort into this, but I think it's really the wrong approach dumping 100+ decks into quest mode and fixing them afterwards.

Quality assurance is much easier, if Corwin (or others) will post only a few decks at a time and everyone can post concerns (about AI usability, balancing, existing similar decks, etc.) or suggestion for improvements.

Maybe we can use thread for new quest decks and have another for improving old decks.

This was originally posted as a testing add on. The decision was made to add it to the SVN so that more people would test it.

What should the policy be for the future when making decks for quest mode?

I'm not only talking about the future. Please stop this practice now! I don't want to sound rude, but I would be really glad if you (Corwin) would:
1. choose just a couple of your 100+ decks.
2. write a post of what themes you had in mind for each of them.
then we can test and discuss these decks, fix, balance and improve them. Then we can go over the next batch. Decks were you can't say what's special about them have to be removed or replaced with better ideas.
Most of your decks are just an assortment of goodstuff (but mostly not in the hands of the AI). I don't want to comment on your 100+ decks all at once.

[Chris H.]

I think that I should unsticky this topic and lock it. I will then go into the SVN and revert us to an earlier pre-100+ quest deck state.

I will then start-up a new stickied topic and we will try this again but with a different approach.

The new topic will have an archive containing just those decks that Corwin decided not to kill.

We will go through the decks one opponent at a time. Alphabetically starting at the letter A and working our way through the opponents.

This will allow a more organized approach. 100+ decks is too much for me to concentrate on at one time. I can't stay focused. While we have a lot of new cards to consider, there are still a lot of holes in the AF code and not every new card can be effectively used by the AI at this time.