Collectible Card Game Headquarters

Hello, everyone.

I'm just dropping by to leave an updated booster-prices.txt list for the next Forge release. I checked it with the most safe / trustable sources available. It includes updated prices for packs, boxes, starter & tournament decks for all editions up to Khans.

I hope it is of good use.

v2 of the updated booster prices file. Updated some rare boxes prices to be better adapted into the Forge scenario (ie, sealed Unlimited booster boxes are significantly more valuable than 36 individual packs due to their rarity, but that doesn't really make sense in Forge since boxes are "replenished" -- so I adjusted the booster pack prices accordingly), and corrected some things that were just downright wrong (updated the list @3AM, etc).

EDIT: Fat Pack prices definitely need adjusts now with the new booster prices, and I'm not sure how to do that. I suppose an automated re-pricing formulae using the price of individual booster packs would work?

Just to let everyone know what I'm doing: I've been working on fixing attack declarations to match the actual game rules (including implementing the missing cards in that category, like Okk), which is a LOT more difficult than it sounds. I've spent all my Forge time since my last commit (three weeks or so) on this, and have gotten a lot further, but it might still be a few weeks before I have a working implementation.

My plan is to create one algorithm that's able to determine, in O(n) or O(n log n) time – n the number of possible attackers – a legal attack declaration, ie. one that fulfils the most requirements without violating any restrictions. This can then be used to check if a human player's attack is legal, which it is if it doesn't fulfill more requirements than the attack chosen by the player; and to give the AI at least one legal attacking solution if it tried a couple (ten or so) which are all illegal.

elcnesh wrote:Just to let everyone know what I'm doing: I've been working on fixing attack declarations to match the actual game rules (including implementing the missing cards in that category, like Okk), which is a LOT more difficult than it sounds. I've spent all my Forge time since my last commit (three weeks or so) on this, and have gotten a lot further, but it might still be a few weeks before I have a working implementation.

Be careful with both Banding and Propaganda effects. But yea, I've done a decent amount of work in that combat code, it's definitely not straightforward.

elcnesh wrote:My plan is to create one algorithm that's able to determine, in O(n) or O(n log n) time – n the number of possible attackers – a legal attack declaration, ie. one that fulfils the most requirements without violating any restrictions. This can then be used to check if a human player's attack is legal, which it is if it doesn't fulfill more requirements than the attack chosen by the player; and to give the AI at least one legal attacking solution if it tried a couple (ten or so) which are all illegal.

How would the time be affected by possible "goals" for attack, i.e. players and planeswalkers? With cards like Ruhan of the Fomori, Alluring Siren and Gideon Jura, that too has to be taken into account.

And will the algorithm be extendable to blocking as well, later?

So while I'm busy making improvements to the match layout, I decided to improve how zones are displayed when you click on the icons by your avatar. Rather than displaying a modal list view, it will now pop open a resizable modeless zone window that displays actual cards. You can even click these cards to activate them, such as playing a flashback card from your graveyard.

I still need to do a bit of work to make it so these floating zone windows are updated as cards move between zones, but otherwise I think they should prove to be a nice improvement over the old system.


EDIT: The zones windows now update as cards move between zones. I've also made it so if you close them, they'll appear in the same location and size the next time they're opened (though at this time that does not carry over if you close and reopen Forge).

I just committed a fix for a big pet-peeve of mine, which was Forge picking an edition of a card that you don't have the image for when there's other editions of the card you do have the image for. This would commonly happen when cards were reprinted in newer sets that don't yet have images for download, and there really wasn't a good way around it unless you built all the decks manually and selected specific versions for each card (meaning you couldn't have Unique Cards Only checked in the deck editor). Otherwise, you'd end up with placeholder art for cards that have perfectly good images that could be used, which was especially problematic for the desktop version where I haven't been able to support custom rendered images like I have for the Android app.

Please let me know if you have any questions or concerns regarding this change.

Thanks.
-Dan

We're definitely not ready for a release tomorrow. I guess we'll see how things look next week.

drdev wrote:
Please let me know if you have any questions or concerns regarding this change.

Which picture gets chosen now?

lujo wrote:

drdev wrote:
Please let me know if you have any questions or concerns regarding this change.

Which picture gets chosen now?

For randomly generated decks, if you have the Random Art setting on, then it will pick a random edition you have the image for. If that setting is off, it will use the most recent edition you have the image for.

For the deck editor, it will always use the most recent edition you have the image for if you have Unique Cards Only checked. If you uncheck that, then you'll see all editions, including those you don't have the image for.

Note that if you have a specific edition/art of a card in a deck you've imported or built, it will respect that regardless of whether you have an image for that version or not.

Is it possible to have an "oldest" or a way to choose? I bet there's people out there who prefer older art and frames (the whole reason I downloaded all the set pictures was so that I could avoid being stuck with new frames or art from "cheap bulk art" like Onslaught or Invasion).

Just asking ofc.

lujo wrote:Is it possible to have an "oldest" or a way to choose? I bet there's people out there who prefer older art and frames (the whole reason I downloaded all the set pictures was so that I could avoid being stuck with new frames or art from "cheap bulk art" like Onslaught or Invasion).

Just asking ofc.

I actually implemented a "Preferred Art" picker for the Android app. I could certainly implement similar support for the desktop version.

In fact, the backend code for it should already work, it's just the GUI support that needs to be added. So in the meantime, if you manually edit (or create) card.preferences in your C:\Users\<username>\AppData\Roaming\Forge\preferences folder to look like this, it should work for you.

Code: Select all

<preferences type="cards">
    <card art="RTR|1" name="Seek the Horizon"/>
    <card art="M14|1" name="Jace's Mindseeker"/>
    <card art="M12|1" name="AEther Adept"/>
</preferences>


The "art" attribute is simply the set code followed by "|" followed by the art index, which is usually 1 except for cards that have more than one art for the same set (e.g. basic lands).

Awesome, thank you very much. Can't wait to see all the mobile stuff improvements applied in the pc version eventually.

I admit I love the new visual zone preview (for library, graveyard, etc.). Great job!
Do you think it's possible/viable to implement the same kind of visualization (except in a "modal" version) for when the player searches the library, picks a card from the library, etc.?

- Agetian

Implementing some sort of visual improvement, or at least user friendly way of browsing your sideboard (or simply making it more like the list in the deck editor, with lands stacked), would improve the draft experience by a lot. Also playing with older sets / full pool, as I bet many people don't know each card by name. In the current system you can also mess up by pressing space and not sideboarding anything, somehow.