SODA - (Super Original Death Arena)
Hey guys, it's been a while since I've posted a project and I'm excited to showcase this project to the community and get some feedback :)
The project is called SODA (Super Original Death Arena) and is a FPS Rogue-like game with a multiplayer component.
This project grew out of the desire to see how rogue-like gaming principles could be applied to the FPS genre with the hopes of creating some interesting gameplay.
All of the levels in SODA are randomly generated, here is a pic showcasing top down and in game views of a dungeon.
The premise of the game is pretty simple, you enter into a tournament hosted by the gods of this realm where they construct unique Death Arenas designed to challenge champions of various species from around the universe. You are a species of people who happen to posses Megaman style arm cannons (which are pretty useful in Death Arenas).
As you progress through the game you will battle through various randomly generated levels each which have an end portal which you have to get to, once you get to the portal you will travel to the room of rest where you are able to save or move to the next dungeon. You earn experience points from killing enemies and will level up once your exp bar is full, levelling up allows you to upgrade your stats and adjust your arm cannon.
Inside the dungeons are various religious alters which the player can worship in order to receive different bullet types and power ups, each of these gods relate to the group which host the tournament who each control their own dungeon which the player will play through.
The game is currently still very early in development and doesn't have much art to speak of however we are currently also looking for graphics designers/3D modellers who might be interested in this project.
The graphics design job would entail creating and texturing enemies and general aesthetic management of most of the game elements, this will be a paid position of course.
Version 0.03
==========
You can download a Windows build here: https://drive.google.com/file/d/0B4Cx5fJAQK5nWG9SM2dpcWRxWXc/view?usp=sharing
Mac Version: https://drive.google.com/file/d/0B4Cx5fJAQK5nb3RWQmc1MW1aS2s/view?usp=sharing
The builds still have quite a few bugs but get the general idea of what the game will be about across. The multiplayer will work over LAN and can work online (however I'm not making any promises as they online portion hasn't been working reliably lately).
The project is called SODA (Super Original Death Arena) and is a FPS Rogue-like game with a multiplayer component.
This project grew out of the desire to see how rogue-like gaming principles could be applied to the FPS genre with the hopes of creating some interesting gameplay.
All of the levels in SODA are randomly generated, here is a pic showcasing top down and in game views of a dungeon.
The premise of the game is pretty simple, you enter into a tournament hosted by the gods of this realm where they construct unique Death Arenas designed to challenge champions of various species from around the universe. You are a species of people who happen to posses Megaman style arm cannons (which are pretty useful in Death Arenas).
As you progress through the game you will battle through various randomly generated levels each which have an end portal which you have to get to, once you get to the portal you will travel to the room of rest where you are able to save or move to the next dungeon. You earn experience points from killing enemies and will level up once your exp bar is full, levelling up allows you to upgrade your stats and adjust your arm cannon.
Inside the dungeons are various religious alters which the player can worship in order to receive different bullet types and power ups, each of these gods relate to the group which host the tournament who each control their own dungeon which the player will play through.
The game is currently still very early in development and doesn't have much art to speak of however we are currently also looking for graphics designers/3D modellers who might be interested in this project.
The graphics design job would entail creating and texturing enemies and general aesthetic management of most of the game elements, this will be a paid position of course.
Version 0.03
==========
You can download a Windows build here: https://drive.google.com/file/d/0B4Cx5fJAQK5nWG9SM2dpcWRxWXc/view?usp=sharing
Mac Version: https://drive.google.com/file/d/0B4Cx5fJAQK5nb3RWQmc1MW1aS2s/view?usp=sharing
The builds still have quite a few bugs but get the general idea of what the game will be about across. The multiplayer will work over LAN and can work online (however I'm not making any promises as they online portion hasn't been working reliably lately).
gren2.jpg
928 x 807 - 107K
randomGen.jpg
1642 x 712 - 189K
Soda screen.jpg
1909 x 843 - 129K
SODA titleSCreen.jpg
1946 x 1100 - 265K
Comments
The speed of your character's bullets are way to slow given how fast the flying orbs move. It makes it very difficult to hit them unless you're very close to them. Even after putting three whole levels worth of upgrade points into bullet speed, I still struggled to hit them. This just led me to sit out of range and snipe them while they were inactive on the ground.
Stage 2 killed me the second I set foot into it. Orbs just shot way to many bullets for me to handle.
I would really like to play it some more, but would only be able to give you meaningful feedback once I can invert the mouse y axis, even if it's a text file I need to edit.
I've been struggling with the bullet issue myself, I want the starting bullet speed to feel unsatisfyingly slow so that once the player starts upgrading it that they feel rewarded. I think this issue might be solved by giving more bullet speed per level up, what are your suggestions? Currently I believe you start at 25 and go up in 2's maybe I should change it so that BS goes up in 5's or maybe even 10's?
I haven't properly balanced stage 2 yet, but will get on that asap :)
Also those that stick it out and upgrade their weapon through multiple levels and finally get something that fires nice will eventually die (since this is a rogue-like) and have to go through that whole process again just to have a weapon that's satisfying to fire. Dying is bad enough. Being then handed a weapon that fires marshmallows makes it worse.
I'm not saying make the first weapon top tier, but it should at the very least be satisfying to fire. There are many other things you can reward the player with ;)
I had the problem that my mouse kept crossing over to my other monitor, and when I fired it minimized the application (but carried on playing so that I could die).
There isn't really a lot here to comment on, everything feels like a placeholder. So I assume that most of what I'm seeing is going to improve, but here is my feedback in any case: The bushes that you can see past but not shoot past and can absorb infinite amounts of damage are probably don't have a place in this style of game. The level layouts aren't well suited to this sort of game, there's lot of cliffs which allow enemies to shoot you from places you can't see, or fail to shoot you when they should be shooting you. I'm not particularly fond of shooting robots to begin with, but, frankly, most roguelike FPS's haven't done much better.
It seems like you're focusing on getting some systems together first (like the dynamic music, and going from level to level, and leveling up). These systems are technically working, but the upgrades as yet don't provide much room for experimentation and each run is likely to result in mostly the same upgrades chosen which won't result in the kind of replayability that you're surely aiming for.
I can see a player enjoying becoming more powerful, and I think FPS roguelikes are an interesting space, and adding multiplayer certainly adds some potential. I don't think anyone has nailed an FPS roguelike yet, let alone a multiplayer-FPS-roguelike, though games like Tower of Guns and Ziggurat have come closer than others.
Good luck! Thanks for sharing so early on!
@BlackShipsFilltheSky I'm actually taking a page out of your book and wanted to get the game out to the community for feedback as soon in development as possible. I'm hoping to keep updating the game and this thread as regularly as possible. All of the assets are currently placeholder (including the silly name ;), on that note I would love suggestions for alternative names, thematically the most important characters in the game are the Gods which create the various dungeons). Also the enemies aren't going to robots in the final version (as I would like the enemies in game to feel like monsters created by evil gods rather than robots) the reasons they look like they do is because I made both enemies out of Unity 3D primitives just to have some placeholder figures. The only assets which aren't Unity primitives are the 2 tree types :P
I've just uploaded a new version of the game, I have temporarily removed stage 2 as it has a few problems I need to fix (instant death and such).
Change log:
+ Added screen shake and chromatic aberration effect for when the player gets hit.
+ Added 7 Deities boons in game and ability to worship or desecrate other alters. Players will find these alters randomly in dungeons. The area around an alter is Holy Ground where enemies cannot enter or damage the player. Players can worship gods to receive boons (power ups essentially) but if they do they have to partake in a religion with that god, each deity has its own religion and punishes those who don't adhere to it. Desecrating alters will allow you to gain favour from your god whilst being punished by the opposing god, (essentially this will upgrade the different power ups however the power up upgrades aren't in game yet).
+ Cheats are now enabled, pressing C will give you bonus ExP so you can level up indefinitely to try out a few different skill builds.
1) Wrath the god of war grants the player with multiple bullets per shot and requires that the player kills all enemies in a level or they will be punished. His punishment halves the players HP permanently and only happens when they transition to the next stage.
2) Chaos grants the player the ability to prey to him to double their damage temporarily, the double damage effects stacks infinitely meaning players can continue to pray to keep doubling their damage. However the doubled damage lasts for only 1 shot and is reset if the player is damaged or misses. Chaos requires that you desecrate all other alters or he will punish you. His punishment removes 5-50% of your current damage.
3) Tsuna Goddess of Spirits grants the player heat seeking bullets which track towards the nearest enemy. If you desecrate her alter she will remove bullet speed. I need to find a religion which suits her ability an play style, please let me know if you guys have any ideas.
4) Raika God of Lightning grants the player infinite bulletspeed but requires that the player never misses a shot, if they miss he removes some atk speed.
5) Zephyr God of Pride removes your gun and gives you a melee attack (I don't have an animation for the melee attack atm so it;s a bit hard to use). He also requires that you never stop moving, if you stand still for more than 2 seconds he will remove movement speed. The melee attack does do significantly more damage.
6) Gaia Goddess of nature gives you a shield that you activate with right click. The shield stops all damage and the player cant shoot while using it, it last 1 second and takes 1 second to recharge. If you get hit gaia will punish you by removing a small amount of experience.
7) The truth grants you with no ability but assures you that it is the one true god. Eventually I will have this god clear one of the later game dungeons for you if you worship him exclusively for a play through as he just provides an interesting handicap.
New download links are in the original post.
I don't think the FPS gameplay is strong enough to try and differentiate your design around yet.
Basically, you're trying to add stats and these granular systems that are supposed to provide player progression and player satisfaction when the core moment to moment play of being a thing moving around in a world and getting shot at isn't really very strong. I feel like you're trying to build a generic shooter space with a bunch of player-exposed numbers in it and then make it the player's job to turn those numbers into fun... Yes, that probably sounds like it's working when someone tells you that they pumped all their level ups into damage or fire rate or bullet speed, but all that's really telling you is players wanted to damage things faster so they could move on.
I feel like you need to take a step back and really explore what FPS differentiation options are available to you. Why does a Quake game feel different to an Unreal? Why do both of those feel very different to COD? How does Halo approach movement? Then experiment with those factors and really dive deep into how you can build enemy interactions that make things interesting for players. Right now the enemies are pretty boring: The tank things running at me and firing doesn't feel good at all, melee enemies running at me gives me a time limit to deal with them, but fast shooting attacks getting closer to me as a player feel icky - I just take a bunch of damage while I'm trying to shoot them, so either I game the terrain/AI ranges or I up my damage to take them out faster. The floating enemies are interesting, but those round bullets are really hard to tell directionality from (in a single static screenshot, which way would you dodge?) and spewing attacks everywhere feels odd when there's no pressure created by that on me.
So, expand your movement and enemy design in these really large chunks first: Like, is a super fast jumpy enemy fun? What about a snipey cannon that charges up and blasts me if I don't move? What about a flying hover-thing that tries to stay away from me all the time? What about waves of enemies? What about enemies that have ground-following attacks I can jump over? What about having ground-slamming things if I get too close? Then do large chunk weapon and movement stuff too: I'd suggest leaving stats out of things completely, at least as player-tweakable things. It's much more interesting as a player to decide between discrete upgrades that are powers, like Nuclear Throne or Downwell does it: Choosing between Vampirism, Ground Pound Boots and a Deployable Shield (when I can only get one of them this level up and I might never see the others again!) is way more fun than making a number go up slowly that only marginally impacts my play experience (I'll always feel like it's not making that much of a difference and I should have picked some other number, no matter what number I actually picked).
So, yeah. The map having maze-walls also seems like it makes things less cool - sure that might be an interesting environment at some point (maybe with some sort of positional peril to not fall into, but don't overdo it, jumping puzzles in 3D suck), but right now it just makes me want to game the AI ranges even more.
(EDIT) Another possibility would be to still have exp and a progression system but have level ups give the player perks as opposed to stat points and the perks could be designed around helping the play feel that there are multiple options moving forward when leveling.
In terms of the maze like structure, I enjoy how the cliffs create various levels of height for the player to play in and I do want players to use the environment to beat enemies which they find more difficult. I don't want playing cautiously and sniping enemies to feel cheap and I also don't want running in with guns blazing to feel to difficult however I would like to punish the player for more rash behaviour. I also want players to jump onto the walls and use the height, I plan to create enemies which will make this strategy more difficult. Right now falling off the side of the map only has you fall back into the map however this is supposed to kill the player which adds a higher level of risk to using the walls as falling off ends a play-through.
The reason I've built the systems horizontally so far is so that I can start experimenting with AI and enemy types as I want this aspect of the game to be most interesting.
Prototyping horizontally isn't an approach I've ever taken, but I think if I needed to make a game with very strong progression elements it's something I'd try.
The downside of the horizontal approach is that the game kind of feels pants for a very long time (in terms of game feel), but in theory it all starts to come together eventually.
FPS's (and all games really) obviously need to feel good. And getting a good FPS feel is non-trivial.
The point I'm getting at is that, in theory anyway, you could also get very far into this project and then find out the assumptions you made about how the game should feel are all wrong, or you just can't get the game to feel good at all. Of course it might all come together, but there's no way of knowing while prototyping horizontally.
Which also makes it hard for us to give feedback. I know you say you're taking a page out of my book... and I'm flattered by that... but the way I usually go about prototyping, in vertical slices, is best suited to getting feedback at every step, where horizontal prototyping has other kinds of advantages/disadvantages... (and it's unknown territory for me).
What I think you could do though, when you feel ready, is produce a couple small parallel prototypes that prototype game feel. Just do a scene with a group of enemies and try make that fun/interesting on its own by changing the behaviours and layouts.
I think quite often FPS's start like this, with just iterating as rapidly as possible on the game feel in a small and controlled environment (that's certainly how I've always started FPS's), and I see no reason why you couldn't do some prototypes like this after you've got the broader systems to a place and you find you need to commit to a feel for your FPS before continuing.
The game feel is going to inform a lot of the level design (in particular), so I'd expect it's possible to go too far horizontally and find yourself in a situation where you're either undoing the systems you've built in order to accommodate the game feel, or compromising the game feel to maintain the systems you've designed.
Sorry if I'm overexplaining this. I think this might all be common sense, I'm not that familiar with horizontal prototyping so I don't really know the most robust way to proceed. It's quite likely you'll get a lot of "I see where you're going, but this doesn't feel good" type comments until you get to the point where you're working on the game feel. As a player it's really hard to look past the feel, and usually it's a bad idea to look past it.
[Edit] Just regarding methods of leveling up/ gaining perks etc. There's so many ways to do this. Ziggurat is one of the worst at this, with mostly just forms of stat increases that make you more powerful. Tower of Guns is just about as bad (though better).
Where Binding of Isaac is excellent, and while maybe not better than Binding of Isaac, Risk of Rain fits together phenomenally...
The value judgments I'm making here is this: Roguelike upgrade systems where the perks/levels you gain change the way you play are better than ones where you only become more powerful.
Obviously, while not a Roguelike, Broforce has benefited immensely from getting players to change their tactics during gameplay. It keeps people engaged when they have to rethink their approach, even if it's a slight change. Becoming more powerful is satisfying, but if the advantage applies to all situations, like adding more hitpoints, it doesn't affect tactics.
The accruing-new-effect style upgrades of Binding of Isaac and Risk of Rain both give the player new case-specific advantages that they then need to consider...
You should be able to pick Risk of Rain up pretty cheap at this point. It's worth a play and is a pretty great stripped-down example of a decent Roguelike upgrade system. There's some other good things about it obviously, but it illustrates what I'm saying about upgrades that give the player case-specific advantages are better than upgrades that give the player advantages in every situation.
In terms of the horizontal prototyping the only reason I had used this approach so far is because this project started out as a game which I was developing for my final project for Uni, the Uni had various requirements for us (such as level progression, levelling systems and such) which I had to put in the game. I completely agree with you that I need to get game feel right first which is why I think I'm going to move forward with trying to rapidly produce a parallel prototype where the progression is more tied to items.
The idea that I had was that each Deity gives you a different weapon, these weapons would then have preset stats (so that I could control game-feel more finely). The player would also have a set movement speed (also allowing me to control the feel of the environment and player movement better) and a basic weapon. The deities would then require the player to perform a "religion" based on the weapon they had and if the player doesn't consistently do this task the weapon is taken away from them. An example of this would be the God of destruction giving a shotgun style weapon to the player but requires that the player kill an enemy every 5 seconds, The shotgun weapon would be much better than the starting gun but would require the player to put themselves in more danger in order to keep it (and the other deities would work similarly but with different guns and requirements).
I'll definitely will check out Risk of Rain, I'm currently more familiar with binding of Isaac (which I've sunk considerable hours into). Do you think an FPS rogue-like would benefit from many many weapons each with unique trade offs (like BOI) or a few specific weapons with direct consequences to play-style (like the above example) or maybe a combination of Both where basic weapons which can be found are weaker but the diety guns are much stronger but have much more visceral trade offs?
I think the way you receive the weapon is important. There's a big gambling element in BOI which doesn't seem to be present in the way you've described the deity system.
I think you need to figure out what the values you want are and then explore the space (as you're intending to do). How much variety do you want from run to run Versus how much control players have? Do you feel it's okay to be locked into certain choices from near the start, Or do you want the player's path to potentially change late in a run? How much gambling do you want the player to do Versus how much control players get? How big and difficult do you want the decisions to be Versus do you want lots of decisions and times to adapt to new situations within a run? How persistent do you want decisions to be Versus how much can change at any upgrade decision? How much fantasy/humour/storytelling do you want you upgrades to relate Versus how much freedom do you want to add upgrades that make sense from a pure gameplay perspective.
As an aside, Unlike Binding of Isaac, FPS's don't work well with constraints on shooting (like the four directional shooting in BOI). So FPS Roguelikes necessarily involve a lot more player input in the attacks, and so upgrades have less impact. Which is part of why FPS Roguelikes haven't had success like Binding of Isaac or Risk of Rain. I really don't know how this can be solved.
My feeling is that first prize is a system that works similar to Binding of Isaac, or for that matter Risk of Rain (which does have one significant choice of weapon before every run). But I think this is something none of the existing FPS Roguelike's have pulled off (like Tower of Guns or Ziggurat)... There is probably a reason why these games resort to either starting you off with a weapon or adding a small selection of weapons to your arsenal as you play.
In any case, finding a great weapon, as you do in Nuclear Throne, has never felt as good to me as building one over successive upgrades as in Binding of Isaac... Some of this is personal taste.... but I'd expect you'd feel the same given your approach in Pixel Boy.
However given the constraints a Nuclear Throne approach might work well (Nuclear Throne having a sort of hybrid of traditional FPS gun trading and Binding of Isaac perks).
Sorry that's not really an answer to your question, and this has been rambly to boot. I think if you try make some parallel prototypes you can maybe find the thing that feels the best (through testing various options on players from these forums). Obviously just trying to copy Binding of Isaac in FPS form is not what you want to do (like the Paranautical Activity team did), I haven't been able to solve designing upgrades for a FPS myself (and I have tried, and likely will try again).
I've developed a parallel prototype trying a vastly different idea based on what you and @dislekcia said. I've removed the experience system and levelling altogether, I've also adjusted the started character to feel a bit better in the beginning. I've reduced the player health significantly and included health drops to make up for this, however hp drops are seldom in order to make health a resource during play (I also did this because it makes it harder for players to just run to the end of the level as they are more fragile and would die, this way the possibility of health orbs incentivises fighting the enemies). Health drops are green orbs which drop from enemies and will heal the player when they walk over them (unless he is full hp).
Instead of levelling up the player can now find weapon parts across the map, each of these parts gives a bonus with an associated trade off, the players must discover what the trade offs for certain items are. The weapon parts conditions will stack with each other. Also because of minimum and maximum stats the order in which upgrades are used also plays a part for instance:
One upgrade could give + 10 damage and -5 Health, another could give -10 Damage + 5 Health. If the player has already taken a power up which reduced his damage to 1 (the lowest) he could then use the power up which gives -10 damage and + 5 health, which would keep his damage at 1 and give him 5 health, they could then use the +10 damage power up which now has no negative for the player as they gained health for nothing earlier. (these two power ups usually cancel each other out but not in this circumstance).
The reason I thought trade offs could be cool for weapon parts is that it allows me to manipulate the stats in such a way that the player is more guaranteed to notice the upgrade and change on play-style than with levels (eg in a level up 1 point in atk speed doesn't feel like much but an upgrade could give you 10 atk speed but also remove 10 health to balance the advantage.)
Currently there are 12 upgrades in game and they are fairly easy to make and with some play-testing and balancing I feel many more could be included. The dungeon will only spawn 5 choices per dungeon for the player to choose from and all in different locations, the spread out nature of the upgrades will encourage exploration while the choices still allows players to customise the weapon to their tastes without the randomness of a run forcing them down a play style path they don't necessarily enjoy.
Weapon Parts SODA (now with EXTRA Caffeine):
https://drive.google.com/file/d/0B4Cx5fJAQK5nU2lYUkp0dzhib1E/view?usp=sharing
Like, stats are your enemy in the design right now. Listen to how complicated the systems you're coming up with to try and make stats interesting are... That order of operations stuff really doesn't explain itself well - nor does it actually let you do neat things later in the game: You can essentially cheat 1 or two stats up beyond your power curve early by dumping them below their minimum, but after that it's just tradeoffs all the way, no spikes. So again, you're sitting with a design that feels less impactful the further a player gets in it.
If I may make some suggestions to add more variety to game-play and perhaps make it more interesting:
1) Adding more threat source types. Currently an enemy is the only threat. If you can add environmental damage of a non bullet type, like spikes, shock panel, moving blades, a pulsing lazer "gate" to time correctly and pass thru etc could make traversing the level more interesting.
2) More movement options. A trampoline device to catapult one upwards. Currently I could jump on trees and then on top of walls, but propulsion other than jump could be interesting. You could add spinning blades like an awkward placed ceiling fan in the air to be avoided perhaps.
3) More aggressive enemy movement. Currently they are fairly static and I feel safe as I cautiously move through the level and snipe from afar. Having the enemy patrol and potentially unexpectedly stumbling upon me could alleviate this feeling of safety and could increase game pace,
4) Adding a throw-able weapon that does area damage (grenade) or that has to be retrieved (boomerang) could be an option perhaps. Maybe a weapon where bullets ricochet off the wall leaving a trail as it leaves the player gun could be interesting. Feeling of satisfaction succeeding at a difficult trick shot etc.
5) Adding a relic slot where you can equip one item and obtain a passive perk (speed, double jump etc) like the shield thing from Gaia which I thought was pretty cool.
Anyways, just some random thoughts having completed earth dungeon 1.
Cheers,
K
With the reward system a maximum of 5 Gun parts can be found in any given level, this allows players to have a certain degree of choice in terms of which upgrades they decide to take without forcing the player down to specific a play-style during any particular run.
I still liked the idea of having power ups with trade-offs but realise how stat trade offs aren't a fun way to implement this, so instead most of the power ups having gameplay trade off and some change how your weapon behaves. I was also thinking about having some rewards which are unambiguously positive for the player (like a damage up or fire rate up in BOI) these rewards would be for players who don't want to alter their play-style but still want power ups and need upgrades, these rewards of course would be less frequent.
I've also implemented a simpler power up combination system which allows all the upgrades in game to stack with each other, unlike the Pixel Boy system the amount of combinations aren't limited in any way and all the power ups can just be stacked onto the gun you start with.
Upgrades in Game:
1) Hand of Wrath -> Increases the amount of bullets you shoot per shot and the amount of damage you take. (more "Hand of Wrath" upgrades means more bullets shot and more damage taken)
2) Hand of Chaos -> Right click to double damage for your next shot. This ability can stack (stacks are limited to the number of "Hand of Chaos" upgrades you've found).
3) Heart of Dance -> Increases how fast you move but standing still causes you to take damage.
4) Vampiric Aura -> Each shot you fire drains 1 HP of your life, each shot which hits an enemy heals 2 (or more based on how many vampiric aura parts you have) hp.
5) Shield -> Right-click for shield, lasts 1 second, regenerates 1 second, you can't shoot while shielding. Blocks all damage.
6) Bullet Core -> Simple damage up power up.
7) Metallic Core -> Simple Atk Speed up power up.
8) Lightning Gun -> Makes Bullet Speed instant but decreases firing rate.
Upgrades planned for next build:
9) Proximity Mine Bullets -> Changes your bullets into prox mines which only detonate after 1 second.
10) Shrapnel Blast -> After 1 second bullets explode releasing random shrapnel in all directions.
Changes to Enemies:
I've removed the bobble head buggies and replaced them with weaker chase enemies which will chase you and shoot at you. I've also made the floating enemies more aggressive, if you shoot them while deactivated they will attack you and charge you and if you shoot them while they are floating around they will drop to the floor and agro you in close combat.
Link to Latest Build (windows): https://drive.google.com/file/d/0B4Cx5fJAQK5ndjRTbm9keFZyM1k/view?usp=sharing
If a Mac build is requested I will supply one.
Risk of Rain offered players some collectible modifiers that would do things when the player leveled up. These made players consider when they wanted to hit the next level and how best to use the effect it offered.
Obviously leveling up at the right time in Desktop Dungeons was a huge part of the strategy and benefited that game immensely.
In general though exploring the kinds of levels you want to generate to find random upgrades is better than getting upgrades based on leveling up (for the reasons you mention).
I'm going to give this build a try later tonight I hope.
The other thing about leveling up based on experience is it adds a bit of compulsion to killing enemies (Diablo obviously is all about getting experience and random loot from killing enemies and that is incredibly compelling).
Risk of Rain has a kind of hybrid system, with experience from killing enemies, gold from killing enemies, and random upgrades strewn around the map that require gold to access... (Sorry I keep mentioning it, but it really is very clever).
This is yet another reason I need to play risk of rain, the hybrid system sounds very interesting, maybe I could tie the Deities back into the game with this somehow. (Just brainstorming here) I could have it that when you kill enemies you gain souls and you have to sacrifice souls at the Alters to receive the upgrades.
Keys opened chests and doors that gave you things. You could use bombs to get places that might give you keys and coins. Coins could be used to buy bombs and keys. etc
And so making choices to balance these resources, and predict what you'd need in the future, was always in players' minds. FTL also had a few resources that could be traded for each other and were necessary to balance.
I think the fact that coins in Binding can be used in multiple ways to achieve multiple goals (as can the other resources) makes coins really interesting. Risk of Rain has some choices about where to spend money, but the choices mostly go one way and are only supply-blocked from one side.
There's of course LOADS of ways to reward players with things that they can use to acquire other things they want (like souls for blessings). I'd suggest though that one of the hallmarks of a really interesting system is that there are multiple uses for each resource.
The obvious suggestion, based on the idea you threw out about souls being used for blessing from deities, is that these souls can also be used to buy consumable items at a kind shopkeeper (like spare health). Binding of Isaac obviously took it a lot further than this, so I'd say, if you're introducing a resource, then making the decisions about when/how to use that resource interesting (by having to decide between multiple uses) has got to be your goal.
Of course, if you think this avenue holds promise then you can probably quickly prototype a resource with a single use and find other uses for it as you go along. I'm not suggesting that you plan the whole thing out now, just don't introduce a resource into the game and leave it as only ever being used in one way.
Version 0.03 https://drive.google.com/file/d/0B4Cx5fJAQK5nWG9SM2dpcWRxWXc/view?usp=sharing
I got an upgrade that made my bullets faster the first time, which felt good. I died before the gate.
I got a dance upgrade the second time (which damaged me before I finished reading the description). I died again before reaching the gate.
I would like some more decisions. I spent quite a while shooting enemies with only picking up one upgrade each time I played. What I'd really like is to choose between two different upgrades now and then, or to choose between two small upgrades versus one large one... Risk of Rain (I know I'm mentioning this again) has a very rapid rate of collecting upgrades, which certainly in the early game is very fun. Both Risk of Rain and Binding of Isaac have slightly more upgrades than you can reliably collect with the resources you have, which means you often have to decide what to leave behind.
Even without the dance upgrade, the speed of the character allows jumping across gaps which feels a little bit like parkour. I enjoyed jumping across gaps and around corners when I discovered I could.
Regarding the enemy behaviours:
Shooting enemies that are beneath the player in trenches is quite unfun and feels a lot like luck (there's a random chance they might be shooting upwards at you when you peek over the edge). Because the enemies are constantly charging at you circling round and firing from another angle is frustrating (because they quickly maneuver to be directly beneath you again). I want to be able to exploit the differing heights to outsmart the enemies, but instead of being an advantage the tranches feel like a disadvantage.
The charging at you in general feels like it needs to be adjusted to give the player more opportunity to exploit their behavior. It's very hard to feel like you've outwitted an enemy when both of you are running in circles round each other spamming bullets. As a result I always tried to kill enemies by sniping them from very far away, I don't mind sniping enemies, but I don't want that to be the only effective strategy.
I feel like enemies could alert other enemies, sniping them one by one seems a bit silly. Alerting other enemies might not be the best solution, but I don't want to kill off the enemies one by one.
With regard to you not picking up many upgrades I feel this problem could be rectified by me putting a minimum on the amount of power ups which spawn in a level and also spacing where the power ups spawn around the level better. As it currently stands the max for power up spawns is 5 however there is a very small chance that 1 or no power ups spawn (which I think is bad) also because I hadn't put logic in to adjust where they must spawn sometimes 1-2 power ups would spawn both towards the end of the dungeon forcing the player to basically play the entire level before getting a power up. I think I'm going to change it so that 5 power ups are guaranteed per level and that they will spawn spaced apart from each other all across the level, also that the number of enemies per levels prevents the player from easily being able to purchase all 5 power ups (I'll need to balance pricing and spawn rates but ideally this should force the player to manage their souls as a resource more carefully).
I think I'm going to pause the game with a small text pop up explaining what a power up does the first time a player picks one up (to cover cases like Heart of Dance which punishes you for not moving).
I think the enemy behaviours need the most work, in all honesty the two which are currently in game are mostly acting as placeholders (however I do like the idea of the floating mine enemy). I'm planning on having it that the capsule enemies are able to traverse the terrain using Jump Pads just like the player, which will hopefully make using the trenches feel a bit more advantages (trick enemy into trench, shoot enemy as it jumps out and is vulnerable). I agree with you that I should slow down the charge to make it a bit more exploitable to the player, it does seem very fast.
Alerting other enemies of your presence is something I hadn't thought of but is very interesting to me and shouldn't be to hard to implement (check within radius for all objects tagged as enemy and agro them or something like that). Possibly combining the altering with the proper environmental traversal on the capsule dudes could make them quite a fun enemy type.
@MCA, Here you go good sir: https://drive.google.com/file/d/0B4Cx5fJAQK5nb3RWQmc1MW1aS2s/view?usp=sharing
The Soul rest room could also be a market place with NPC's where you buy the upgrades (with a bartering or mission giving element? - e.g. kill all the X's in the next arena and I'll ... or retrieve the sacred blah and I'll give you ...), rather than in the arena, though a healer of sorts in the arena would be useful.