next »

[Paul Jeffries]

I have issues with your statement that a traditional roguelike is beatable on the first go by a skilled player. Prior game knowledge and spoilers in classic roguelikes are used to substitute a progression system (ADoM, NetHAck and Crawl are what I played most, and they all present this problem in different degrees -NetHAck being the most obvious offender-). Sure, there were no mechanics forcing the grinding, and I agree that's ideal.

Well, knowledge of the game and it's systems is part of what I'm counting under 'skill' but yes I take your point.  What I'm trying to say is that there is no artificial barrier to progress that the game itself is putting in your way, whereas in RL you have to die a couple times to even unlock basic stuff like a double-jump.  I would also say personally that being too spoiler-reliant is a flaw in those older games, though to a certain extent I think it's unavoidable.

Actually, Binding of Isaac and Shiren the Wanderer are my main references for a metagame.

Yes!  Shiren has an awesome metagame.  I think Spelunky also has quite an interesting system with its shortcuts to later levels - they theoretically help a less-skilled player out, but using them comes with the disadvantage that you can't use the relatively easy early levels to gather equipment, so there's a trade-off in deciding to use them and the separate challenge of completing the game going through all the zones still remains unblemished.

On the graphics front: I think my issue with the palette is not so much the colours involved (I quite like that the colour selection is limited) as how dark everything is - it makes the screen look nicotine stained!  I think that just by brightening up those cream colours a bit you could make the whole thing seem a bit less oppressive.

[Juan Raigada]

Daily Devlog (1 of ?):

As promised, I will post an update today.

The BombHead should be pretty self explanatory, but just in case: it makes the NimbleBot explode (and damage you and the enemies) on dead.

I probably will tweak the AI at some point, but right now, it feels *good enough*.

Next update will be coming later today, hopefully. After that, we will have to wait until the 29th -this week I travel away, so no work to be done in the game-.

Current schedule is to have the NimbleBot and the new weapon (Blade-on-a-stick) finished by August 13th -plus whatever else I can fit-. After that it's a race to finish all the not-so-hard to do content and release a playable.

[laughing Frogs]

i really want to play this.
Have you seen archeblade on steam? It's a free to play (pvp, only i think) brawler, i kind of like their system, but imo it's too complicated for mouse&keyboard, it basically longs to be played with a controller. I think my question is will your game be better for one of the input schemes, (disregarding personal preferences ) ?
Did you think about how "granular" your upgrade system will be? What i mean is, a damage upgrade in something like isaac feels a lot more powerful than a damage upgrade in rogue legacy (for obvious reasons). I'm just asking because i find it fascinating and it can change the "game feel" quite a bit, i think.

[Juan Raigada]

Daily Progress Update (2 of 2):

So did some random bugfixing and started to model and bring into the game the new piece for the NimbleBot.

The Radial:

I think I went too far with the yellow stripes, so I will bring the non-saw part of the arm more in line with the standard look.

No functionality yet, but it doesn't break any of the existing attacks, so that's good. I plan on implementing two different attacks for this part. One will be a "pirouette kind of attack with little"horizontal translation but a wide hitbox. The other one will be a rush forward, similar to the current and standard jump attack, but ground based.

I'm also thinking about making those saw automatically damage the player on touch, but I dunno if it will be too frustrating to navigate around them.

These attacks should not be too hard to implement.

1 part and a half until the NimbleBot is done... Next content update will be the 29th.

[Juan Raigada]

i really want to play this.

Soon, I hope...

I think my question is will your game be better for one of the input schemes, (disregarding personal preferences )?

I'm designing with a controller.

I haven't yet thought too much about keyboard inputs, although I will include them. There are two attack buttons that can be mapped to the mouse buttons, and a jump that can be mapped to the spacebar. Evade could be control or a third mouse button. Apart from that we need inputs for running (shift, I think) and to change weapons (trigger hold on the pad, but could be mapped to toggles mapped to Q and E.

I think it can be playable, although probably will have issues on the first playable. It will require some polishing to get those controls right! I will try my hardest, though, but this kind of game does ask for a controller.

Any suggestions for a good M&KB brawler inputs?

Did you think about how "granular" your upgrade system will be? What i mean is, a damage upgrade in something like isaac feels a lot more powerful than a damage upgrade in rogue legacy (for obvious reasons). I'm just asking because i find it fascinating and it can change the "game feel" quite a bit, i think.

I agree it's extremely important. I have thought about it a lot and most likely it will have very little granularity. At the moment, weapons do double damage at max level. You can probably combo two or three equipment items to get 8x damage in certain circumstances, but that SHOULD break the game, or at least allow for crazy damage output (at the expense of versatility). Balancing is quite far away still, though.

[Juan Raigada]

Random Development Update:

So I'm back teaching at DigiPen Bilbao, and I was preparing a lecture on game *physics* and reading Game Feel and I had an idea that I wanted to show my class in a demo.

So I quickly coded

(video). Notice the arc the ball performs as opposed to a more traditional gravity based approach. It's subtle, but it should be noticeable (it "hovers" a little and then falls faster).

Basically, I´m mapping the speed at any given point in the jump to an Unity AnimationCurve that allows for much more precise control of the behavior *and* gives the movement much more personality.

This is the curve I´m using. It represents speed over time, with the speed inverted when I go over 0.5 in time.

Of course, I will adapt this system to Heart & Slash. There was a lack of polish in the movement that was bothering me. This will solve it (and allow for more precise tweaking of the feel) although it might be some time until I can get to it. (It won't be very hard to do, just slightly time consuming).

[Juan Raigada]

Daily Progress Update (1 of 2):

So the first of the two bumps in productivity is over. I will be working on this Mon, Tues & Sun for the next couple weeks.

I am rusty, though, so it's going to take a while to get back up to speed.

Today I finished the Saw attack for the NimbleBoy. It worked out better than I thought.

Now I will implement the last part/variation on this enemy (tron-like trails) and polish things a little (the explosion of the bombhead and the gigantism modifier could use some extra time spent on them -they lack feedback-).

I will be done with this tonight, I hope.

[Calum Bowen]

those graphics are giving me pangs of dreamcast nostalgia. Reminds me a lot of Jet Set Radio! Really cool!

[Juan Raigada]

Daily Progress Update (2 of 2):

Day was slow, due to unforeseen errands and reforms going on in the apartment next to mine (hard to program with a drill going on at full strength next to your ear).

Anyway, I got the tron-like trail working, but it's not yet polished, so I'll show it tomorrow.

Even with these complications I'm still going to finish the NimbleBoy one day ahead of schedule. This is good. Only hard thing to do for First playable is the new weapon (other kind of bodyparts tend to be 1 or 1.5 days affairs). By the end of August all content will be in and then I'll have a couple of weeks of polish/bug fixing before release.

Right now I'm pondering whether tomorrow I should start implementing the new weapon or just do some of the overdue polish/movement code rewrite.

[Juan Raigada]

Daily Progress Update (1 of 2):

Finished everything related with the NimbleBot. I have included a video of four of them destroying me:

The radial attacks are already pretty hard to evade, and the electricity trails make everything work. They are definitely balanced as HARD. Which is good because it's always easier to cut down than to make them more difficult (plus, I need some range in difficulty for the random spawning to work).

Expect a hard first playable.

I'm going to rest a little and then I will start tackling the movement rewrite for the main character.

[Juan Raigada]

Daily Progress Update (2 of 2):

I started to implement the curve based movement (already feeling way better) but it's not finished. So no video. I'm optimistic. I think it will be time really fell spent (since it deals with the most direct "feel"of the game).

Also... How to say this???

The most important thing I realized today is that the game IS HARD AS HELL. Basically I brought the two new enemies into the game and did some preliminary balancing with their experience levels. 

And I got killed many, many times. The new enemies are way more agile, so they are way harder to crowd control than the old BigGrunts. I will need to do a lot of balancing on their AI to bring them back to manageable levels.

I am conflicted. On one hand I love the game can get rough enough, but on the other hand, it's waaaay too much right now. We'll see how much I can tweak before the playable.

Another Thing I realized it that I'm lacking a lot of variation in room design. This was expected, but right now I feel the right amount of environmental content is going to be higher than expected. Also, the regular spawing mechanism I have implemented is decent, but not enough. I need to have enemies already spawned into the rooms in certain circumstances, to create more of a feeling of "place", so to speak.

Nothing I can do anything about right now until I bring all the planned content into the playable, but I will have to think hard about this problem further down the line. Coding wise all of this is trivial, but it will require a certain amount of content creation...

[nickgravelyn]

Random Development Update:

So I'm back teaching at DigiPen Bilbao, and I was preparing a lecture on game *physics* and reading Game Feel and I had an idea that I wanted to show my class in a demo.

So I quickly coded

(video). Notice the arc the ball performs as opposed to a more traditional gravity based approach. It's subtle, but it should be noticeable (it "hovers" a little and then falls faster).

Basically, I´m mapping the speed at any given point in the jump to an Unity AnimationCurve that allows for much more precise control of the behavior *and* gives the movement much more personality.

This is the curve I´m using. It represents speed over time, with the speed inverted when I go over 0.5 in time.

Of course, I will adapt this system to Heart & Slash. There was a lack of polish in the movement that was bothering me. This will solve it (and allow for more precise tweaking of the feel) although it might be some time until I can get to it. (It won't be very hard to do, just slightly time consuming).

This looks really cool. I've used the Unity curves for a few similar effects (controlling scaling or translating for screen shake type effects). Care to share more details about how you mapped this curve to a character's jump height? And also does it cause issues if you're jumping up over something (so the ground height underneath is changing)?

The game's looking mighty cool, btw. :D

[Juan Raigada]

This looks really cool. I've used the Unity curves for a few similar effects (controlling scaling or translating for screen shake type effects). Care to share more details about how you mapped this curve to a character's jump height? And also does it cause issues if you're jumping up over something (so the ground height underneath is changing)?

The game's looking mighty cool, btw. :D

It's really simple. The curve just maps directly to the vertical speed (speed = maxSpeed*(1-curve.Evaluate(t) ) with a flip to negative once we are over the apex (0.5 t in this case). You put that into your FixedUddate and you get something very similar to an aceleration based jump in behavior (you just change speeds differently) so it has no problems with ground changing or things like that. Like an acceleration based jump, behavior can become slightly inconsistent on very high dT. But if you are getting that low a framerate you are gonna have many more issues...

It just gives you more control. Like how the second part of the curve simulates a bigger gravity than the first part along with a drag applied to it.

As to height, you can't use the standard 1/2*acceleration*(time^2) so you reverse engineer changing the variables (curve shape, max speed and time the curve takes) until the height is appropiate for the game...

[nickgravelyn]

Oh that makes sense. I'll have to try it out later and see how it feels to me. Thanks for the explanation. 

[Juan Raigada]

Daily Progress Update (1 of 2):

I spent some time yesterday setting up a Github repository for the project. With that I can:

1- Go to sleep without fearing tons of work could be lost anytime.
2- Work at the game while I'm on Bilbao teaching (if I have some free time).

Both of these should speed up development.

On other news, I finished implementing Heart's curve based jumping. I also brought that approach over into the Katana air attacks.

As a result the game now feels better to control. It's still not 100% done (There are tweaks to do with horizontal translation for a better feeling) but combat is snappier, which was a necessity due to how hard the enemies can be right now.

And let me speak about these enemies:

For a while I have focused on implementing content, which is good and has given me a good sense of progress. However, as I work in the game, what I implement is better/faster/more complex. When I brought back all the new enemies into the game, what was supposed to be smaller, simpler enemies, have become fast predators that in some configurations are REALLY HARD to deal with. Thus, I've decided to hold for a couple of days on implementing too much new stuff and tweak the system and the sense of pacing a little bit, so as the game can be enjoyed.

Some basic tweaks on enemy "danger cost" has already proven useful in mitigating the frustration. I know I will have to tweak their AI at some point, but that AI is right now (I think, I don't have a profiler) one of the big performance bottlenecks on this (I check too often on whether to choose a new state or not). So I will wait until I optimize that bit before tweaking further.

This afternoon I want to take the basic square room and add a lot of variants to it, as to make the game feel more varied and interesting (It's still too repetitive). This work will probably continue tomorrow until I have a room I can call first-playable complete. That is, a room with enough variety and polish as to serve as a frame of reference for the other rooms in terms of content.

In some ways I'm behind in my planning and in other ways I'm ahead. A mid September playable is a given at this point. Since the only piece of content I really need to make is a weapon I'm positive I can deliver in time. Some of the other planned content might be behind schedule and thrown into the playable later, but it's not a big deal, since, but for weapons, I feel there's enough collectible variety for a first playable...

Tonight, on my second update, I'll show what I am doing with the square room right now...

[Juan Raigada]

Daily Progress Update (2 of 2):

Ok, so this was a productive afternoon!

I implemented basic variants of the same old square rrom you guys have seen in previous videos and screenshots.

Apart from the regular, old, empty room, we now have:

-A room with a special floor that slow down enemies (not-flying ones) that touch it.

-The same thing, but this time with a floor that slows YOU. Also notice the fans making the rrom much tighter. Both versions of the floor have the open room shown above and this variant.

-Cool electricity spikes that damage you. This time no tighter version. I will probably implement that variation, but I need to put a platform in there or otherwise there would be no way to avoid damage!

-A moody warehouse variant of that room:

-And the warehouse with badly maintained radioactive vials. this gives a room wide effect (the skulls) that multiply damage given and received by 2!

Everything works as intended. More polish is obviously necessary, but I am happy it took only one afternoon to bring this much variety into the game. Along with the tweaks I did this morning on enemy experience and placement, the game already feels much richer (if still quite sparse).

I'm again worrying about optimization. This time I'm sure it's not the graphics, but the scripting. I get down to 200 frames per second. Once I get a hold on a Unity Pro license (might be a while) I will have to really dig into the profiler to catch the offending scripts!

I don't know yet what I will be doing tomorrow. Maybe more rooms (polishing this one and bringing variety to the other two rooms I have the basic structure of)...

[Juan Raigada]

Daily Progress Update (1 of 1):

Well, yesterday was extra productive, so today had to be slow .

I have been doing some system restructuring to better cope with spawning enemies in very different positions depending on the specifics of the room that has spawned. I'm halfway through it. What I'm missing is giving each enemy a particular type of "insertor" so that some of them appear on the floor, some on the ceiling, etc (right now, since they share an insertor, they need to appear from the same kind of place).

That will take another half day of work, but once finished, I will be able to finish polishing the three rooms that are already into the game.

I realized it was time to tackle this issue as I was doing today's work: polishing the chasm room to add enemies and variety.

Some screenshots of that:

This weeks work has proven useful in making the game way more fun and engaging, but that's something you guys won't enjoy until the playable.

[Juan Raigada]

Daily Progress Update (1 of 1):

Just one update today, since most of the day was spent squashing bugs that came up in the latest content updates. Also, I made some progress in the build-up polish pass towards releasing a playable.

The most "showable" stuff I did today was implementing an almost final version of the "enemy insertors" in the first level.

Short video:



Next up is continuing the polish so I bring my non-content to-do list back to a very reduced size (it's already quite small). I should be finished with this pass tomorrow, so after I come back from holidays I can focus on implementing the last pieces of content needed for the first playable.

Which is still happening mid-September.

[Juan Raigada]

Hi Juan,


Sent you an email about a week ago and I was wondering whether you've received it.


Cheers!

Wrote you back!

[Juan Raigada]

Daily Progress Update (1 of 1):

I'm going to NYC for a week, so I wanted to get a lot of stuff done before that. Basically that means a lot of bugs were fixed and many, many were discovered (and fixed on the spot).

I added some stuff, but this time it's a secret, so I'm not showing it (come on, the playable is coming soon, you don't want to have seen everything that is in there)  .

Right now, the foundation for the project is the most solid it's ever been. I'm very happy by how moderately bug free it is right now, and how solid most of the core systems are. The only system (aside from menus) that I think I need to touch at all at this point -other than coding behavior- is the level generator, which needs, secret rooms and boss room implemented. Also it needs to know not to stop putting room in until the boss is generated.

There are some systems that might come in place later (NPC stories -basically the progression system-, unlocks, etc...) but those are very concrete and not overreaching, so I'm not worried about them.

Plus, I sent a couple of build to collaborators and it seems performance is good on a three year old computer. This bodes well for many of you to be able to play the first playable pre-optimization.