Update 0.21

Hello everyone!

Well Friday was supposed to be the launch of the Web-Release to various web portals but I've decided to postpone this for a week in order to do some final balancing and polishing. Since I was really pushing to release yesterday this update is therefore a pretty massive one that finalizes a lot of design decisions. The next week should mostly be minor changes, fixes and cleanups leading up to the release.

CUTTING THE MESS:
The main thing I'm focusing on in this update is just stripping away any ideas, designs or elements that feel kind of half baked and that I feel need more time during EA to fully develop. Since the Web-Release is going to have to stand on its own with minimal updates after release, I really don't want to put anything into it that I'm not totally confident about. Similar to cutting out the races, this doesn't mean in any way that these things are gone for good, rather I'm simply saying "I'll deal with this when I have more time". I'd rather have the Web-Release be smaller and tighter rather than larger but with a lot of mess that will never actually get cleaned up.

REDUCING DIFFICULTY:
The second thing I'm focusing on is a continued reduction in the overall difficulty of the game, especially in the earlier sections. I've said before that the difficulty level I'm aimming for in the base game is one in which any experienced player should find it a pretty easy experience. The base game should require focus and attention but should otherwise be a fairly casual experience for anyone with more than 10-20 hours of play time. With a tiered ranked/challenge mode coming in EA there is really no reason to not treat the base game as the introductory 'easy mode'. Furthermore from the perspective of the Web-Release acting as the introduction to the game itself for a large number of players I think its better to lean on the side of easier to help ease the learning experience for players that are likely coming into it with minimal experience in traditional rogue-likes.

THE TALENT SYSTEM:
The biggest change here is with the games talent and character progression system. Throughout Alpha I've being adding new talents without a real overall plan, simply to test out how different talents and abilities work with the more advanced gameplay that I'm envisioning. Some of these talents have worked out great while others have not. A lot of them, in my opinion, simply require a lot more work to get 'right'. In the last few weeks, in an attempt to balance the number of talents per class I've added a bunch more simply to 'fill holes' that are probably not really necessary. So there were a lot of talents in the game that I really think need more work and so have been 'removed' for the Web-Release with the intention of revisiting them in EA.

Without any real consideration to the overall character development system, the number of talents per class grew from 6-8. Due to the changes made in number of attribute points the number of guaranteed talent points also grew from 12-16. So in both cases, changes made for other reasons ended up quite drastically changing the overall balance of the talent system without really intending to. I've since become quite uncomfortable with the sheer number of talents and talent points that are accumulated throughout the game. Not only is it overly complex and messy in what is supposed to be a fairly streamlined game, but I feel that its removing a lot of the hard choices that were necessary in RFIII due to the combination of a smaller starting talent pool and less overall talent points. So this to has been reduced with the intention of revisiting in EA.

To summarize the changes I've reverted the game back to RFIII's numbers in which classes start with 6 talents and the player gains a guaranteed 12 talent points over the course of the game. This was done in the first case by removing talents that I felt are just not working yet and in terms of talent points, the Tomes-of-Knowledge will no longer grant the talent points along with the talent. From my own quite extensive testing with this system I think its working much better than what we had before. The question of what we end up with in 3 years time is almost certainly 'something quite different', but for the Web-Release and keeping things streamlined and high quality I think its working very well. From my own notes:

• In the previous system the player often felt like he was gaining talent points faster than attribute points and so was often waiting around and spending talent points on sub-optimal choices. The new systems seems to line up much better and I'm finding that I don't even unlock everything I want from the current tier before unlocking the next. This leads to a much smoother feeling progression and more interesting choices.
• A consequence of this is that I've found I suddenly actually need to specialize a bit since I can't just get everything I want and especially as new talent tiers unlock I find myself suddenly with pretty difficult choices to make which is exactly what a good character development system should do in my opinion.
• In the older system it was often the case that the last 3-4 talent points gained would be sort of throw aways. I'd already gotten all the core talents I wanted and was essentially just dumping points in things that were much less important to me. This greatly reduced the reward of gaining new talent points. With the new system I'm finding myself desperate for every point I can get right up to the last one.
• The previous system with 8 Class-Talents and 4 Wild-Talents tipped the balance very heavily in favor of Class-Talents and so it was the case that most characters of a given class ended up looking very similar simply due to the fact that so many of their points could be spent in the same 8 talents. With a 6/4 ratio however much more of a characters final build is determined by the Wild-Talents found in each run.
• The other issue with the 8 Class-Talents was that they simply tended to cover to much ground and give classes to many answers to the games challenges. There wasn't really a strong incentive to branch out since you could easily handle the majority of the game with just those talents found in your starting kit. Most of the talents removed to go from 8 => 6 were honestly kind of crappy but this has still had the effect of 'tightening up' the starting kits so that they provide less solutions and so consequently there's a bit more of a pressure to make use of Wild-Talents to cover those holes.
• There was also the issue in which classes talents tended to overlap with each other a lot which is no longer the case. The current talent list actually has very little overlap and there are some seriously useful talents that were previously part of a classes base kit that are suddenly going to be extremely rewarding when found on a tome or in the library.
• The biggest causality in terms of talents were the so called 'Charged Talents' on which I could write a whole essay about my feelings. In summary though I'm not really sure if these are necessary, how they should work, what their purpose is in the overall game design etc. So in the interest of making the Web-Release tight and refined, many of these have been removed and will be readdress in EA.
• Generally speaking the classes feel much more cohesive and 1-dimensional which both makes them play more uniquely and also creates a big incentive to broaden your build based on resources found in the dungeon.
• A final consideration is that in EA I'm planning on a system where each talent has multiple possible upgrades and so reducing the number of talents per class makes even more sense since otherwise we would end up with something like 24-36 'slots' for points in the base class alone.

FINAL TALENT LISTS:

• Every talent should be very useful. I'm trying to remove as many of the 'dump talents' as possible so that the player has to make difficult decisions and every talent gained feels powerful and meaningful.
• Every class should feel cohesive with definite strengths and weaknesses.
• Every class should feel lacking in some pretty fundamental areas in order to encourage branching out with your build. This also makes many talents very rewarding if found.

FIRE MAGE:
The Fire Mage has almost no utility or defense with his kit and is focused almost exclusively on dealing damage. He has a decent mix of damage ranges and targeting types (single vs AoE).

• 1) Fire-Ball
• 1) Shield-of-Flames
• 1) Mental-Clarity
• 2) Burst-of-Flame
• 2) Flame-Portal
• 3) Flaming-Battle-Sphere

STORM MAGE:
The Storm Mage has a blend of damage and knock back abilities that tend to make him more focused on dealing with groups.

• 1) Lightning Bolt
• 1) Shroud-of-Wind
• 1) Mental-Clarity
• 2) Shock
• 2) Burst-of-Wind
• 3) Summon-Tornado

ICE MAGE:
The Ice Mage is quite defensive and though he has powerful offense it is all very close range.

•  1) Cone-of-Cold
• 1) Shield-of-Ice
  • Note that this is the regenerating HP version from Tier-2
• 1) Mental-Clarity
• 2) Freezing-Cloud 
• 2) Ice-Lance
• 3) Freeze

NECROMANCER:
The Necromancer is all about sustain, able to keep himself alive and dealing damage longer than almost all other classes. He lacks any real source of burst damage.

• 1) Life-Spike
• 1) Aura-of-Death (this one still feels a bit meh)
• 1) Mental-Clarity
• 2) Poison-Cloud 
• 2) Blood-Portal
•  3) Summon-Skeleton

ENCHANTER:
The Enchanters abilities are all extremely powerful but his kit is completely lacking in any way to actually directly attack enemies.

• 1) Discord
• 1) Staff-Attunement
• 1) Mental-Clarity
• 2) Confusion
• 2) Mirror-Image (this one still feels a bit meh)
• 3) Dominate

WARRIOR:
The Warrior is completely defensive with Power Strike being the sole (though powerful) exception. He completely lacks mobility or utility of any kind but is the most survivable of all classes.

• 1) Shields-Up
• 1) Fortitude
• 1) Shield-Block
• 2) Power-Strike
• 2) Recovery
• 3) Shield-Wall

BARBARIAN:
The Barbarian is completely offensive and with some of the changes made to his talents he now lacks any real inherent defense or sustain. He's a bit more mobile than the Warrior but will probably still be very happy to find additional mobility talents.

• 1) Charge
• 1) Weapon-Mastery
• 1) Killing-Strikes (this one still feels a bit meh)
• 2) Cyclone-Strike
• 2) Blood-Lust
• 3) Berserk

DUELIST:
The master of mobility and with a very high melee damage potential. Otherwise very fragile and lacking sustain.

• 1) Disengage
• 1) Jump
• 1) Precision
• 2) Lunge
• 2) Strafe Attack
• 3) Dash Attack

RANGER:
The master of mobility and range attacks but very fragile and lacking sustain.

• 1) Power-Shot
• 1) Sprint
• 1) Perfect-Aim
• 2) Tunnel-Shot
• 2) Strafe-Attack
• 3) Storm-Shot

ROGUE:
The master of dirty tricks and exploiting enemy AI. The rogue otherwise lacks defense and even inherent offense.

• 1) Sleep-Bomb
• 1) Stealth-Mastery
• 1) Sneak-Attack
• 2) Bear-Trap
• 2) Shadow-Step
• 3) Vanish

PLAYER BALANCE:

• Talents per class reduced 8 => 6 as above and Tome-of-Knowledge no longer grants a talent point along with the talent. Total talent points is therefore reduced 16 => 12. This is exactly the same numbers that RFIII used though with a slightly different implementation.
• Tome-of-Knowledge offers 5 talent choices (removed the free talent point).
• Tome-of-Knowledge, Tomes and Librarian offer some of the talents that were removed like Meditate, Enchant-Item and Evasion. I think in EA there will likely be a lot more of these 'Wild-Talents' that are not connected to any particular class.
• The system that Tome-of-Knowledge uses to offer a balanced set of talents between attributes and tiers is now also used by the Librarian i.e. she should offer a more comprehensive and balanced list.
• Soul-Reap, Fire-Storm, Lightning-Storm and Freeze have all been moved onto the associated elemental wands and their charges set to 3/3. This makes these previously kind of pathetic items into serious power houses and puts these abilities where they belong imo i.e a powerful but limited resource.
• Reduced the EXP requirements to attain level 3 quite a bit. I think a big part of making the early game easier is to simply let the player level up a bit faster. The major danger in the very early game is simply that the player has so few tools and resources. So speeding up this process should help significantly with early game survivability.
• All Tier-1 spells have had their MP cost reduced 4 => 3. The casters were feeling quite weak compared to the other classes and these basic abilities, which are really like a secondary fire mode should be very cheap to use.
• Energy Shrooms adjusted from 1 + 25%MP => 2 + 25%MP. This makes them more powerful in the early game without making them overpowered later on.
• Healing Shrooms adjusted from 30%HP => 5 + 25%HP. This makes them MUCH more powerful in the early game and for low HP characters while maintaining balance later on. The math works out to a huge effect at low HP, a slight buff for HP < 100 and a slight nerf to HP > 100 which is totally fine imo.
• Poison Damage has been slowed down quite a bit to give the player more time to find and use healing shrooms and is capped at 5DMG per turn so that even massive amounts of poison damage don't kill you in a few turns.

TALENT BALANCE:

• Berserk: Removed the Life-Tap component in order to streamline the ability and remove some of the Barbarians inherent sustain.
• Berserk: Duration 5,5 => 8,8
• Charge: costs 1SP
  • A pretty substantial nerf to the Barbarian early on that has caused me to take DEX much earlier than I normally would. Doesn't make a huge difference in the later levels once you have more than 2-3 SP.
  • This is both a matter of consistency with all other movement abilities and a matter of balance since charge is actually a pretty massive movement ability (comparable to Shadow-Step). It also opens up the possibilities of STR vs DEX focused Barbarian builds.
• Charge: Crunch damage 150% => 200% partially to compensate for above changes partially because this is somewhat difficult to pull off and should be rewarded. Its also just cool to smash enemies into walls for massive damage :P
• Blood-Lust Duration: 4 => 5
• Blood-List only needs to be adjacent to blood. These 2 combine to make the ability much more usable.
• Burst-of-Flame only conserves flame at Rank-2 for 50%.

ITEM BALANCE:

• Bear Hide Cloak: 8HP => 20% Cold Resistance. There were just way to many HP items in the early game which was causing practically all characters to end up with inflated HP pools at little cost.
• Goblin Battle Staff: removed the MP as the heal effect alone makes this an extremely powerful early game weapon.
• Potion-of-Power 10DMG, 100%AP => 5DMG, 50% AP. The power effect was just waaay to powerful especially with abilities that use a weapons damage. This change also applies to Power-Shrooms. If needed the duration could be increased a bit.
• Staff damage has been reduced a bit for the middle tier staves though they still build towards the same final damage at Tier-3. This mostly just smooths out the damage progression curve.

DUNGEON BALANCE:

• Reduced the number of monsters per level slightly across most of the game (especially end game). As the player has weakened and the enemies have gained power since RFIII the main balancing factor is supposed to be less enemies per fight and less enemies per level. Levels are supposed to take roughly the same time to complete as RFIII its just a matter of less swarmy spammy fights.
• Reduced the danger level of Upper Dungeon: 1,2,3,3 => 1,1,2,3. The major effect of this is that level 3 enemies will no longer spawn on the second level. This should do a lot towards making the early game easier.
• The chance for a Glyph room to spawn has been reduced from 50% => 30% as these tend to be both very challenging but also very time consuming. In order to keep the pacing up, Glyph rooms should be a somewhat rare but exciting challenge.
• Wandering elites should never spawn prior to danger level 4 (start of the wilderness zone). This was bugged and wandering elites were ONLY spawning in danger levels LESS than 4. Woops!

MONSTER BALANCE:

• An overall slight reduction in enemy HP
• An overall slight reduction in enemy damage
• A relative reduction in the HP of LOW-HP enemies (swarming enemies). This makes them easier to AoE and lets me make the swarms larger and more 'swarmy'.
• A relative reduction in the HP of MLOW-HP enemies (casters, supports, archers, and some dangerous melees). This makes them easier to focus down and lets me make them more dangerous or powerful.
• A relative reduction in the damage MHIGH-DMG  enemies (mostly heavy melee). Some of these numbers were crazy high considering they have no cool-down.
• A relative reduction in the damage of HIGH-DMG enemies (caster nukes). A lot of these had the potential to 1-shot a player not paying careful attention and this should be avoided in the base game imo. Ranked mode will be a whole other story!
• Enemies will no longer cast Inspire on non-melee allies.
• I've made (and will continue to make) a whole bunch of little micro adjustments to stats on enemies across the whole game just based on my own play experience. Most of these are pretty minor so I won't list them all but generally I'm trying to streamline enemies so they have just 1-2 major strengths or threats.

USER INTERFACE:

• All 'Charged' abilities (those that remain) now have proper descriptions.
• Ogre Chef: now shows his damage gain on ally death 'ability'. This is generalized for when we add more abilities like this.
• Fixed Storm Staffs description.
• Force Shield now has a graphic that looks terrible (like Ice-Shield) but is at least more visible to the player. Will do something about these later.
• Ice Bomb targeting will now correctly show that it passes over low obstacles.
• The total Poison Damage is now shown in the status effect.

GRAPHICS:

• Moved mushrooms up a bit so that they show better behind walls and obstacles. 
• Mine Carts will now properly rotate on the tracks when their vault is rotated.

BUGS AND CRASHES:

• Fixed all the Shield-Wall-2 odd interactions with various abilities (there were a bunch of them).
• Shield-Wall-2: Trample will not trigger off of Strafe-Attack.
• Shield-Wall-2: Strafe-Attack will not trigger off of the movement of Trample.
  • The interaction between the last 2 was... crazy.
• Vines from the Vine Bow must match the vines of the zone (since UG for example uses different ones).
• Sealed up all the Glyph rooms so that they no longer generate with a corner missing.
• Fixed enemies spawning in doors and some objects.
• Shadow-Step, Blood-Portal, Flame-Portal and Blink will no longer trigger strafe attack.
• Lunge w/ spear will no longer target through obstacles.
• Aura of Death was not correctly adding any Toxic-Resistance despite its description.
• Fixed that freaking Tunnel-Shot missing bug. Turns out it only happened at max range with a bow and Perfect-Aim-1 (or probably staves with Perfect-Aim-2).

Well that's it for now. I'll be spending the next week doing a ton of testing and adjusting. Trying to catch all the last little bugs and just generally polishing things up. From what I've seen so far the balance, mechanics and character progression are now pretty damn solid, at least as solid as is possible for a project that is essentially only at the 20% complete stage. I'm eager to hear what you all think and what suggestions for tweaks and adjustments you may have. Obviously major changes will have to wait for EA but there's still definitely lots of room for improvement here.