What needs to be improved?

[mtgrares]

This is a dangerous question to ask but what are the worse parts of Forge?

And I know that everything could be done better from state effects to dealing damage. In my puny version 2 code I modified Input to it could create a dialog (pop-up) box if needed.

[slowe]

Ooh, you've asked for it...

Really though, Forge is pretty balanced in terms of its weaknesses -- I feel that if something is glaringly bad it gets worked on first. So I'm having a bit of trouble singling anything out (or maybe I've just gotten used to Forge's unintended features ). A few things I did come up with:
1) Mana pool issues. If you can't cast something, it's really annoying, and if you can cheat extra mana it's broken. Right now you can't pay for Flame Javelin or Spectral Procession from the mana pool, which occasionally makes them unplayable. You can also essentially double your colorless mana using snow mana.
2) Including cards that aren't entirely implemented. It's really frustrating to see cards that have madness or convoke or some other ability that isn't actually implemented.
3) Not properly and explicitly implementing steps/phases and priority. Mickey mousing in before attackers and before blockers steps helped alleviate a lot of the complaints related to this, but fixing it in a general way takes care of all those complaints and is better in the long run.

Hmm, guess I did have a few things to say. Still, Forge has come a long way and most of it works quite admirably. Keep up the good work, and good luck with version 2!

[Chris H.]

It would be nice if the computer had a mana pool of sorts and could then use sources that provide more than just a single mana, like a Black Lotus.

It would also be nice if the computer mana system could sort the list of mana sources, place non creature based mana first in the list. Creature based mana sources would come afterwards. This way there would be some hope to having the computer animate man-lands and the mana needed to do so would come from non-creature based sources.

@ Slowe - We used to have far more cards that are not fully implemented. At one time, a substantial portion of all cards were not complete. The number of these cards has dwindled and is now fairly small.

The steps/phases work reasonably well at this time … Thank you Dennis. Having the computer respond to our actions might be more welcome by the user base. Imagine the AI casting a Giant Growth after blockers are declared rather than before it attacks.

[slowe]

The steps/phases work reasonably well at this time … Thank you Dennis. Having the computer respond to our actions might be more welcome by the user base. Imagine the AI casting a Giant Growth after blockers are declared rather than before it attacks.

That's what I meant by "features" of Forge I've gotten used to. Having the AI be a bit more interactive at instant speed would be great. I always feel a little bad killing the AI's pingers or other creatures with relevant abilities knowing it won't activate them in response...

[mtgrares]

Forge is basically one big prototype that happens to still work. I know that everything in Forge could be improved. Not having the other phases never really annoyed me. I haven't gotten any phase complaints since Dennis or somebody added the After Attackers phase stop.

And yes Forge has made tremedous progress. 3,000+ cards is insane, Forge is easily playable for months.

Thanks for your comments.

[Huggybaby]

Just squash bugs, before adding any new cards or features.

[Sloth]

Just squash bugs, before adding any new cards or features.

This may be what the user wants the most, but the developers need some greater goals, other than bug fixing. They do this for fun not for money.

My suggestion is:

AI improvements concerning the use of instants and cards with flash.

[Huggybaby]

Just squash bugs, before adding any new cards or features.

but the developers need some greater goals, other than bug fixing. They do this for fun not for money.

Perhaps. But there's a difference between a developer and a card adder. Anyone can add a damn card, but it takes a real man to make sure it works properly with every other card. That is the prime directive in fact. When rares started MTGForge, he only started with a handful of cards until he got them working.

If you only add cards and don't focus on bugs or usability, then you end up with something no one wants to use. Personally, I won't use buggy software. I don't have the time or patience. Even one small bug will tell me that the program isn't ready for my time and I will find something else to do.

What's better, 2,500 cards, all of which work, or 3,000, 200 of which don't work? The new DotP has less than 1,000, but they realized that anything less than a good interface and cards that work would never sell. I know, Forge is free, but even the most independent person wants others to enjoy his hard work, otherwise he'd never post changes here or commit to SVN.

[Chris H.]

It can be easy for us to paint pictures using only black and white, the two extremes. The Forge "world" is many shades of gray. Every beta release comes with a long list of bug fixes. We can't force people to read or to remember them.

I am still very new at java and programming and yet even I have squished a few of these insects. Most of the bug reports that come in are above my skill level. When I fail at fixing a bug or fail at creating some new code, I view this as a learning experience, and I grow with it.

Rares has stated on many occasions that he had to cut many corners. Rares original version(s) had many bugs and incomplete cards. Over the last two years our dev team have succeeded at:

fixing many but not all bugs
completing many but not all of the incomplete cards
improving the AI in some areas
adding new fun cards and desired features
adding a new quest and then an improved quest mode

We move forward one small step at a time. No one is forced to use something that displeases them. And lets not forget how much the tone has improved in our forums these last two years.

There will always be people that are unhappy and spend to much time seeking attention by complaining. These people might complain publicly or they might file their complaints privately with a moderator. I do not envy your position. And I am not pointing the finger at Huggybaby. He can't help it if some people file complaints privately and he realizes that someone needs to stand up for the silent majority.

None of us can make an unhappy person be happy. A hard lesson to learn.

[Resonantg]

Well... I just thought of this so I wonder if it's possible or if someone smarter thought of this earlier.

What about a "skinnable" interface? I have no idea how hard that would be, and it would allow then people to create different skins for their Forge and share those too. Just a thought.

You guys are so far ahead of the curve on so many fronts otherwise, I'm a very happy camper otherwise. I tap 1 colorless and summon 'happiness'.

nobody better counter that.

[Huggybaby]

To change the look, try options->display options->Napkin.

Back on topic, don't misunderstand me, I'm perfectly happy with the way Forge is being developed. Everyone is doing their part, doing what they can, and a very nice balance is being maintained between new features, new cards, AI improvement and bug management. Frankly, I don't think it could be handled any better. If I were in charge of development and was paying people, maybe I would change priorities a little bit but that's a minor thing.

Still, Forge asked 'what needs to be improved' and for me that is and always has been about bugs because 1) there are more cards in Forge than I know what to do with now and 2) Forge won't soon have an interface like Incantus. lol

I know from watching this project and the Shandalar project that adding cards is what people can do, what they like to do, and what they want to do. I accept that reality and wouldn't change it if I could. I think I'm a more casual player than average so if the people who actually use these programs all the time can live with bugs but not without new cards then it's appropriate to emphasize adding cards.

Like Chris said, bugs are constantly being addressed. So, it's all good and we would end up at the same place regardless.

PS No one has ever complained to me privately about forge.

PPS I did just remember one thing that could be changed: I'd like it if the separate parts of new releases were combined in one post. I think right now you have the forge release, then graphics updates in a separate post, and then the rarities file in another post. Why can't they all be together?

[Chris H.]

Dennis and I tried to include the rarity file update with the beta release. It worked for a few versions. One version Dennis had to remove a card that we planned to release from the cards.txt file. A last minute problem came up. We did not realize that the card was still listed in the rarity file that was included. The best laid plans of mice and men. Some bugs are small, smoe will lead to a major crash.

Since we moved to the SVN we decided that it would be best to move all edited/changed cards.txt entries to the top of file. This prevents us from not seeing these types of bug fixes and then unknowingly overwriting them.

Also, when we sort using "Most recently added" we see both the new and the edited cards at the top of the listing in the deck editor. If people sort by rarity after downloading a new version, they can scroll down to the end of the listing and all cards without a rarity value is new.


And for Rares. Helping out with forge 1 is more interesting than trying to continue with version 2 on his own. If he had a new machine he could run eclipse and if he had his own internet account he could join us on the SVN. Now that would be an ideal world.

[mtgrares]

I did just remember one thing that could be changed: I'd like it if the separate parts of new releases were combined in one post. I think right now you have the forge release, then graphics updates in a separate post, and then the rarities file in another post. Why can't they all be together?

The versions that I post on my blog include all of the extra stuff but I only update the "blog version" at the beginning of the month, so it lags behind the versions that Dennis posts.

You can get the latest "blog version" from: http://code.google.com/p/forge-card-game/downloads/list

Since Forge is unlike an official commercial project, different people release different files and users are expected to update their own version of Forge. I think Manalink 2.0 project is similar, users are expected to update the program using many different pieces.

[Huggybaby]

Since Forge is unlike an official commercial project, different people release different files and users are expected to update their own version of Forge. I think Manalink 2.0 project is similar, users are expected to update the program using many different pieces.

Manalink is similar, and the most frequent questions there are how to get the updates to work, precisely because the pieces are spread out, and in spite of having a wiki. No one will update the wiki, and no one will consolidate the pieces, so there is no telling how many people walk away who would otherwise be happy using the updates. It's a shameful situation IMO.
Organizing Forge should be easier since there are less pieces and the core group is more cooperative. Other than inertia I really don't see the holdup. If you want I'll make my own sticky with the links, just PM me when they're released.

[juzamjedi]

I had posted a long response to this thread previously, but the forum keeps timing out before it will post. Grr. So in summary: make the AI better at playing instants. For example I never have to worry about gang blocking a big creature and being blown out by an instant.