Support NeoGAF

[amanset]

Do you happen to remember what the lessons learned from that experience were -- what the most important things that should have been optimized were for example?

As a side note, one very simple optimization I will probably do is only calculating costs of squares that I know are possible to move to, instead of parsing all map squares. The worst case cost would remain the same if a unit could theoretically move everywhere, of course.

The two main ones were C++ related. I wasn't really used to namespaces and the new (to me) data structures. As such I kind of panicked with doing a find in a set (IIRC) and implemented my own find routine as the iterator concept was so new to me. Turns out it is really easy and not using it looked amateurish as well as failed to use any inbuilt optimisations.

For the other major one, referring to the wiki page (I basically implemented the pseudocode on there):
http://en.wikipedia.org/wiki/A*_search_algorithm

For this:

current := the node in openset having the lowest f_score[] value

I originally had everything going into (IIRC) a set ordered by f score, which would make finding the lowest very quick. However, I ran into problems (obviously) when different nodes had the same f score, so I panicked (again) and just ordered them by an ID number. The correct thing would have been to use (IIRC) multiset which allows entries with the same value. So instead of iterating through the entire set to find the lowest f score I could have taken it from the ordered multiset.

To be honest, if you follow the pseudocode it only does the nodes you need to do, rather than the whole map, which I understand is your major issue.

I also found this page really useful:
http://www.policyalmanac.org/games/aStarTutorial.htm

 

[superNESjoe]

My team's current project is going to be revealed at the end of this month. I have a set of contacts from both my personal life and my professional time spent as a Game Journalist, so I feel like I have a good starting point as far as getting word out.

Outside of that though, do any of you that have shipped games have advice for spreading word or convincing websites to cover your material?

 

[Blizzard]

I also found this page really useful:
http://www.policyalmanac.org/games/aStarTutorial.htm

Thanks for the response and the very helpful link! I notice that they promptly point out the Manhattan heuristic means they don't always get the shortest path.

I personally detest using templates in C++ because I feel they can very easily hurt readability. I've also probably run into either requirements or restrictions where using std namespace C++ API functionality either made code really bloated, or slower, or was outright disallowed (because of embedded requirements, rules, older compilers, or similar). However, this may indeed mean I will never get a job again because the API is probably the way to go with newer compilers.

I feel like I've also been burned by stuff like code that is complex and carefully designed through layers of function or method calls...which did not get inlined, resulting in huge call overhead.

*edit* Granted, it's laughable for me to talk about efficiency when I mentioned that my own algorithm was super inefficient, but I just wanted to comment about the C++ stuff. I hope you found a better job after that!

 

[amanset]

Thanks for the response and the very helpful link! I notice that they promptly point out the Manhattan heuristic means they don't always get the shortest path.

I personally detest using templates in C++ because I feel they can very easily hurt readability. I've also probably run into either requirements or restrictions where using std namespace C++ API functionality either made code really bloated, or slower, or was outright disallowed (because of embedded requirements, rules, older compilers, or similar). However, this may indeed mean I will never get a job again because the API is probably the way to go with newer compilers.

I feel like I've also been burned by stuff like code that is complex and carefully designed through layers of function or method calls...which did not get inlined, resulting in huge call overhead.

*edit* Granted, it's laughable for me to talk about efficiency when I mentioned that my own algorithm was super inefficient, but I just wanted to comment about the C++ stuff. I hope you found a better job after that!

Yeah, I think it all depends on the heuristic used. I've seen other implementations that do other stuff, for example the square of the hypotenuse.

I have a pretty good job anyway (in a really, REALLY, niche part of the games industry), but this one sounded interesting too. Thanks though

Also, this job is what gets in the way of me doing my own stuff. I'm finding it hard to find the time. Damn full time employment. Grr.

 

[JulianImp]

So... I had been thinking about how I'd be implementing bezier curves for Quark Storm's levels in order to both make level design faster and the resulting environments being more organic-looking in a single stroke. Yesterday, a local indie dev I'm following on Facebook posted a link to this video, which basically explains the math behind bezier curves: https://www.youtube.com/watch?v=S2fz4BS2J3Y

Before implementing that, I've been researching how to create additional handlers inside Unity's scene editor for managing the curves' tangents. It took me a while to wrap my head around when and how I'd have to use handlers, but I'm done with that for now.

All that's left now is using the equation shown in the video and a sample size N which should result in N subdivisions between two bezier vertexes, so N=1 would mean a single vertex halfway, 2 would be two vertexes at 0.33 and 0.66, and so on.

 

[Jobbs]

I showed a friend what I was doing with maps, and he said "why doesn't the dot move inside the room to correspond to your position?". As I prepared to say "too hard, not worth it" I started thinking about it and realized it's not that hard.

So I did it: http://gfycat.com/DearestSecretBarasingha

Now the question is do I even want this. I'm honestly not sure. Not sure if I want this precision. It's not as charming.

 

[show me your skeleton]

Now the question is do I even want this. I'm honestly not sure. Not sure if I want this precision. It's not as charming.

for what it's worth, i definitely agree. too precise.

 

[Five]

I agree, you don't need that level of precision. That said, I'd probably also change the level of abstraction one bit further, highlighting or outlining the individual room instead of using a dot marker.

 

[Roubjon]

I showed a friend what I was doing with maps, and he said "why doesn't the dot move inside the room to correspond to your position?". As I prepared to say "too hard, not worth it" I started thinking about it and realized it's not that hard.

So I did it: http://gfycat.com/DearestSecretBarasingha

Now the question is do I even want this. I'm honestly not sure. Not sure if I want this precision. It's not as charming.

Just to throw my 2 cents in. I think the only reason you friend thought to ask that question is because there is a little blue icon in the center of the room that is representing you being there. So when there is already a little avatar at the center of the map and it's not moving, you wonder why doesn't it move along with you.

If you just highlighted the room you are in a different color, or made it glow softly or something, then the user would never even think about "why isn't there an icon that shows where I am", because the idea would never be planted in the persons head in the first place.

But yeah, do whatever you feel is right. I'd want to play your game, not someone else's. It's looking really nice.

 

[Pehesse]

I showed a friend what I was doing with maps, and he said "why doesn't the dot move inside the room to correspond to your position?". As I prepared to say "too hard, not worth it" I started thinking about it and realized it's not that hard.

So I did it: http://gfycat.com/DearestSecretBarasingha

Now the question is do I even want this. I'm honestly not sure. Not sure if I want this precision. It's not as charming.

To be honest I wondered the same thing as your friend, but after reading your thoughts on the map and how you use it, I believe having it not moving fits more with what you have in mind (or how I understand it). I like to see the dot moving around, but it is indeed an unneeded level of detail, especially if all areas of the map are not visible.

Just to throw my 2 cents in. I think the only reason you friend thought to ask that question is because there is a little blue icon in the center of the room that is representing you being there. So when there is already a little avatar at the center of the map and it's not moving, you wonder why doesn't it move along with you.

There's probably some of that going on, too! Nice observation!

 

[Jobbs]

yeah, I'm kinda leaning away from the specificity, you guys seem to be on the money. it's kinda "neat" but it just doesn't really fit the game, which doesn't want you to use this as a GPS but as a rougher guide.

thanks

 

[Ashodin]

I showed a friend what I was doing with maps, and he said "why doesn't the dot move inside the room to correspond to your position?". As I prepared to say "too hard, not worth it" I started thinking about it and realized it's not that hard.

So I did it: http://gfycat.com/DearestSecretBarasingha

Now the question is do I even want this. I'm honestly not sure. Not sure if I want this precision. It's not as charming.

I think the dot is too bright and attention grabbing. More something like you are here would be good.

 

[Whitebison]

I showed a friend what I was doing with maps, and he said "why doesn't the dot move inside the room to correspond to your position?". As I prepared to say "too hard, not worth it" I started thinking about it and realized it's not that hard.

So I did it: http://gfycat.com/DearestSecretBarasingha

Now the question is do I even want this. I'm honestly not sure. Not sure if I want this precision. It's not as charming.

Yep, I think highlighting the room is better.

 

[UncleLeo!?]

Sorry, coming into this one late but another alternative a friend of mine has been using if you ever want to use Unity is 'Playmaker'. He's not big on programming but he's found the visual interface and state machine playmaker provides has let him create all kinds of stuff.
Admittedly though, I'm not exactly sure how good Unity is yet for 2D stuff compared to dedicated 2D engines like Gamemaker and co :3

Thanks for that as well.

 

[Noogy]

Huh, another Unity animation roadblock.

I'm still using mecanim (the default Unity animation tools) for 2D, mainly because I want to adhere to the workflow once I start adding polygonal elements.

It's fine for the most part... say you have a character comprised of a number of sprite elements, and every animation uses the same elements.

The problem is when you want to switch elements out. For example, a running animation may use 10 run elements (bodies, legs, etc)... but a hurt animation could use a totally different set.

This is an issue because you can't just add or remove elements per animation, they have to persist throughout all the animations. So you have to add a ton of active flags on anything you don't want to do use for each animation. I'm trying to mess with having children objects for the character so that I can keep each 'animation set' separate, but then I have to have multiple animation controllers per character.

Anyway, it's messy, and clearly meant for 3D animation. I'm wondering if Spriter, Spine, or the 2D toolkit would be a better option here, although I haven't seen enough examples to convince me that they have made the process any easier. I had to build my own animation engine for my last game, and I think it's spoiled me.

Edit: I guess I should pose a question for those in the know... is there a simple and efficient way to manage a library of animations each with different elements, sprites, and skeletons for a single character?

 

[Jacksinthe]

So I managed to fix the cursor issue (at least in FF, so ymmv). In the end what I did was instead of setting the cursor through the player settings, I attached the cursor texture to my player, and had a separate cursor script assign that texture to the cursor when needed. All very simple, but slightly unintuitive to have to manually code something that there's already a built in option for!

But with that, my first game SEAL, BE MY FRIEND! (built from scratch in Unity) is complete. I say game, really it's more of a joke for a friend who recently told me about her futile attempts to befriend seals while sailing. There are MULTIPLE ENDINGS though, and that's a thing that games have, right?

I only started fiddling with Unity three weeks ago, although I already had some not-so-recent coding experience in C++ and Java so I've not been starting entirely from scratch on the knowledge front.

There's a couple of known bugs:

Spoiler

you're not supposed to be able to 'force' yourself to travel backwards beyond the limits of the camera, but in the end I had to give up trying to fix the camera code to get this finished on time (and preserve my hobbyist sanity)

and

Spoiler

you can shoot seals slightly more frequently than the gun actually shoots

. If you find any others though, I'd love to know what I've missed.

Play, don't play, I won't be offended



CLICK HERE TO PLAY

I'm cracking up over here haha! A bug I ran into tho is that whenever I click with the mouse - the ability to control the boat stops. I am unable to resume play

 

[electroflame]

I guess I should pose a question for those in the know... is there a simple and efficient way to manage a library of animations each with different elements, sprites, and skeletons for a single character?

I'm sure that there is, but there isn't going to be a one-size-fits-all solution, at least in native Unity. It seems that Unity Technologies added in 2D sprite support but kind of forgot about animation.

There are some 3rd-party solutions, though.

SmoothMoves seems to be the most popular one, but at $75 it's a little pricey.

Puppet2D seems to be pretty popular as well, some even saying that it's the best (and better than SmoothMoves). It comes in at a less-expensive $30.

Then there's also Sprites and Bones. It's not very popular, but I believe that that's largely due to it being Open Source and Unity not allowing it on the Asset Store. It seems pretty good at first glance, so maybe try it out and move on from there?

Personally I'd go with Sprites and Bones, see if it'll work, and then move on to Puppet2D if necessary as it looks to be better than SmoothMoves (to me, at least).

 

[Noogy]

I'm sure that there is, but there isn't going to be a one-size-fits-all solution, at least in native Unity. It seems that Unity Technologies added in 2D sprite support but kind of forgot about animation.

There are some 3rd-party solutions, though.

SmoothMoves seems to be the most popular one, but at $75 it's a little pricey.

Puppet2D seems to be pretty popular as well, some even saying that it's the best (and better than SmoothMoves). It comes in at a less-expensive $30.

Then there's also Sprites and Bones. It's not very popular, but I believe that that's largely due to it being Open Source and Unity not allowing it on the Asset Store. It seems pretty good at first glance, so maybe try it out and move on from there?

Personally I'd go with Sprites and Bones, see if it'll work, and then move on to Puppet2D if necessary as it looks to be better than SmoothMoves (to me, at least).

Many thanks, I'll go through those. I own a license to Spriter so I'll play around with that, and Smooth Moves seems to implement what I need.

 

[Roubjon]

Had some fun programming these respawning spirit dudes. They exist only in this level so that the chick can get back to the top if she falls down from fighting the BIG BADDIE up top.

EDIT: Added some slowdown to the hits.

 

[MDave]

Just want to quickly mention I've just released my first indie game on the Vita! Oh the buzz of seeing it on the store, is quite something special It's on the Playstation Mobile section:

It's BallFlip, just a simple arcade game of dodge the incoming blocks. Has a couple of powerups to mix it up a bit. It's no fuss, straight into it kind of game. It goes on endlessly, but there is some little things that change over time.

The initial loading time might be a while, but after that each play is fast

Let me know what you think of it, and what you would like to see

 

[Jobbs]

Had some fun programming these respawning spirit dudes. They exist only in this level so that the chick can get back to the top if she falls down from fighting the BIG BADDIE up top.

beautiful!

 

[Geminosity]

Had some fun programming these respawning spirit dudes. They exist only in this level so that the chick can get back to the top if she falls down from fighting the BIG BADDIE up top.

wow, looking very stylish

 

[Anustart]

I think tomorrow I'm going to pick up 2D ToolKit. Are there any other tools that I might find handy in development of a 2d game?

2D TK seems like it will increase my output dramatically, that will be a godsend.

 

[Blizzard]

I showed a friend what I was doing with maps, and he said "why doesn't the dot move inside the room to correspond to your position?". As I prepared to say "too hard, not worth it" I started thinking about it and realized it's not that hard.

So I did it: http://gfycat.com/DearestSecretBarasingha

Now the question is do I even want this. I'm honestly not sure. Not sure if I want this precision. It's not as charming.

What do you think of only highlighting the room, but not showing a dot at all? That way you see what area you are in providing the general rough map you want, but people don't get confused by the dot.

Also, I will second the request to please disable (or maybe a checkbox option to disable) the flickering for people who are seizure/photosensitive.

 

[Jobbs]

Many thanks, I'll go through those. I own a license to Spriter so I'll play around with that, and Smooth Moves seems to implement what I need.

I've kind of come around to the idea that Spine 2D is the future, so I decided I'm going to really learn and master it once I get the chance.. That said, I haven't yet. But as far as swapping out individual graphics, yeah, you can do that. it's only the beginning. What interests me is the ability to do that in concert with the ability to warp the graphics along meshes or set the weight of bones within the meshes so they deform naturally along joints. Using all of this stuff properly, I believe, means amazing animations could be made that don't really have that "2d skeleton" look, but instead look totally authentic.

What do you think of only highlighting the room, but not showing a dot at all? That way you see what area you are in providing the general rough map you want, but people don't get confused by the dot.

Also, I will second the request to please disable (or maybe a checkbox option to disable) the flickering for people who are seizure/photosensitive.

This is what I want to do, that said, oddly it's going to be more work to set up than the last thing was, because each individual room isn't an object. The way I set it up is that each cluster of rooms representing a fragment of the map is its own object, so highlighting individual rooms would mean manually getting sizes and coordinates of every room and setting it all up by hand. Possible, yes, but a bit tedious. That said, I'm sure I'll get it done one way or another, because I think it's for the best.

 

[popyea]

Huh, another Unity animation roadblock.

I'm still using mecanim (the default Unity animation tools) for 2D, mainly because I want to adhere to the workflow once I start adding polygonal elements.

It's fine for the most part... say you have a character comprised of a number of sprite elements, and every animation uses the same elements.

The problem is when you want to switch elements out. For example, a running animation may use 10 run elements (bodies, legs, etc)... but a hurt animation could use a totally different set.

This is an issue because you can't just add or remove elements per animation, they have to persist throughout all the animations. So you have to add a ton of active flags on anything you don't want to do use for each animation. I'm trying to mess with having children objects for the character so that I can keep each 'animation set' separate, but then I have to have multiple animation controllers per character.

Can you elaborate on this? I'm a Unity newbie, but all my experience with it has been creating a pipeline that suits a traditional 2D animation style. I used a plugin called Uni2D, which is fairly broken and difficult for the most part. But I was able to figure it out enough that I could ignore most of it's functionality and just use it to create a character made out of separated 2d parts, which can be deformed quite nicely with bones. The way I had it set up was each section of the character had it's own folder in the hierarchy (eg. an arm folder, which contains all the various arm poses with their own bones), and that I would just disable and enable pieces using the keyframe editor in order to switch parts in and out.

I'm curious if this method is going to cause problems down the line.

 

[JulianImp]

I've just finished writing a news update for Quark Storm on IndieDB after making a much-needed breakthrough on the tools I've made for assisting in level design. Now, all that's left for me to do is wait until the site staff approves it! (so tired)

Had some fun programming these respawning spirit dudes. They exist only in this level so that the chick can get back to the top if she falls down from fighting the BIG BADDIE up top.

That looks really cool! Having to go up after falling off of a boss room seems like it could be a bit annoying, though, but if you're having these spirits show up specially in there, then I guess I could be missing some interesting tidbit behind that design decision.

 

[Roubjon]

That looks really cool! Having to go up after falling off of a boss room seems like it could be a bit annoying, though, but if you're having these spirits show up specially in there, then I guess I could be missing some interesting tidbit behind that design decision.

 

[Turfster]

Never ever try to update the collider for something that is in a hierarchy that contains a rigidbody upstream in Unity. It leads to a lot of cursing and hair-pulling.

 

[Jobbs]

What do you think of only highlighting the room, but not showing a dot at all? That way you see what area you are in providing the general rough map you want, but people don't get confused by the dot.

Hm. There's one thing I forgot about, which complicates the issue a bit.

The dot does provide some utility. The point is that when you go off-map, (which is something you'll be doing quite a lot) you still have a rough idea where you are in relation to the map.

example: http://gfycat.com/InexperiencedLightIvorybackedwoodswallow

 

[missile]

Had some fun programming these respawning spirit dudes. They exist only in this level so that the chick can get back to the top if she falls down from fighting the BIG BADDIE up top.

EDIT: Added some slowdown to the hits.

Am really a fan of the art style! :+

 

[Noogy]

Can you elaborate on this? I'm a Unity newbie, but all my experience with it has been creating a pipeline that suits a traditional 2D animation style. I used a plugin called Uni2D, which is fairly broken and difficult for the most part. But I was able to figure it out enough that I could ignore most of it's functionality and just use it to create a character made out of separated 2d parts, which can be deformed quite nicely with bones. The way I had it set up was each section of the character had it's own folder in the hierarchy (eg. an arm folder, which contains all the various arm poses with their own bones), and that I would just disable and enable pieces using the keyframe editor in order to switch parts in and out.

I'm curious if this method is going to cause problems down the line.

It sort of depends on what you are doing with your animations.

For example, lets say you have a character that you always see from the same angle. The number of body parts never changes, and the parent/child links are fixed. You are basically treating this like a 3D model. Then you are probably fine.

However, take this example. it's a bit exaggerated on purpose, but explains the issue I'm running into.

You have a simple humanoid character with 4 body parts. When the character jumps, he turns into a tarantula from a top-down perspective. When he runs, he is a single sprite with the animation baked into a spritesheet with zero interpolation.

Now say I want to adjust animation types per frame. That's the sort of functionality I'm trying to figure out, and it seems to be a hassle with native Unity. But I like enough of the native tools to try and figure it out before moving with a 3rd party app.

 

[popyea]

It sort of depends on what you are doing with your animations.

For example, lets say you have a character that you always see from the same angle. The number of body parts never changes, and the parent/child links are fixed. You are basically treating this like a 3D model. Then you are probably fine.

However, take this example. it's a bit exaggerated on purpose, but explains the issue I'm running into.

You have a simple humanoid character with 4 body parts. When the character jumps, he turns into a tarantula from a top-down perspective. When he runs, he is a single sprite with the animation baked into a spritesheet with zero interpolation.

Now say I want to adjust animation types per frame. That's the sort of functionality I'm trying to figure out, and it seems to be a hassle with native Unity. But I like enough of the native tools to try and figure it out before moving with a 3rd party app.

Can you say what you mean by "animation types"? I follow up until that part.

 

[Noogy]

Can you say what you mean by "animation types"? I follow up until that part.

Sorry, I should have said 'animation methods'. For example, skeletal/IK vs single sprite frame.

Example: You have a character getting hit, and all the body parts animate into position. But you want the last frame to be a static image that you custom created that looks nicer than anything you can manipulate the rig into doing. All of this helps to break up the 'paper cutout' effect of rigged 2D animation.

 

[popyea]

Sorry, I should have said 'animation methods'. For example, skeletal/IK vs single sprite frame.

Example: You have a character getting hit, and all the body parts animate into position. But you want the last frame to be a static image that you custom created that looks nicer than anything you can manipulate the rig into doing. All of this helps to break up the 'paper cutout' effect of rigged 2D animation.

Okay cool. That's the sort of thing I was planning on doing. I guess the way I would have done it is by disabling all the other parts that made up the animation, and having that last pose only enabled at the end. Not sure how your set up works, but I'm enabling/disabling the gameobject using the inspector. This worked when I did a prototype (with just 1 or 2 frame animation placeholders for each action), and the pose switching was pain free. I guess I will probably run into the same problems as you if I try something more complex.

 

[Blizzard]

I started adding some obstacles. I am faced with a stylistic choice.

The case on the right uses the Advance Wars style of drawing units overlaid on the map even if there are tall obstacles one row down. My choices seem to be:

1. Draw units behind obstacles, at least when the units are not selected. This results in the units being partially obscured and potentially difficult to read at a glance.
2. Design all tall obstacles to be exactly two "squares" in height so overlap is not a concern. They might look weirdly tall or fat in this case.
3. Design all obstacles to be exactly one "square" in height so overlap is not a concern. They might look weirdly short in this case.
4. Overlay units on top of tall obstacles, even though this looks a bit weird.

Opinions are welcome!

 

[anonymous_abc]

1.
Overlaying them just looks wrong.
But show us what 1 high buildings would look like.
The first image looks fine to me though. The vehicles aren't that obscured.

 

[Adam Scott Aukerman]

lol, super cute and pretty well presented too, especially for a first try. I also like the job you've done with the audio as the soundscape feels pretty complete with all actions having a sound and good background ambience.
That all said, I distinctly suspect that the friendmaker isn't making me friends though

Thanks very much. I spent much decade away from coding studying and working in video and animation, so I'd like to think I'm making use of some of that knowledge as well. The audio largely came from freesound.org, and was a total godsend (as it usually is).

I'm cracking up over here haha! A bug I ran into tho is that whenever I click with the mouse - the ability to control the boat stops. I am unable to resume play

Cheers! What platform/browser are you using?

 

[V_Arnold]

I started adding some obstacles. I am faced with a stylistic choice.

The case on the right uses the Advance Wars style of drawing units overlaid on the map even if there are tall obstacles one row down. My choices seem to be:

1. Draw units behind obstacles, at least when the units are not selected. This results in the units being partially obscured and potentially difficult to read at a glance.
2. Design all tall obstacles to be exactly two "squares" in height so overlap is not a concern. They might look weirdly tall or fat in this case.
3. Design all obstacles to be exactly one "square" in height so overlap is not a concern. They might look weirdly short in this case.
4. Overlay units on top of tall obstacles, even though this looks a bit weird.

Opinions are welcome!

Introduce opacity based on what is the current unit! If the one being obscured is the current unity, change the alpha of the unit obscuring it to half, and now you get the full view of your current unit. And vica versa.
(edit: just realized you were talking about obstacles. In which case, obscuring your unit when it is not selected should be fine, and looks fine as well, imho! The only question is whether you can/want to implement transparency when that unit is your focus.)

 

[Jacksinthe]

Thanks very much. I spent much decade away from coding studying and working in video and animation, so I'd like to think I'm making use of some of that knowledge as well. The audio largely came from freesound.org, and was a total godsend (as it usually is).



Cheers! What platform/browser are you using?

PC/Chrome

Speaking of audio - freesound is great. As a former audio producer, I like using subtractive, additive and FM synthesis for my 8-bit style sounds. But if you don't have the software or equipment you can cheat a little bit: BFXR and SFXR. They are like mini swiss army knives of bleeps, bloops and plings of various waveforms. No 2 sounds are ever alike.

Then for the more advanced "legit" sound: Famitracker
It has a pretty steep learning curve and is basically a tracker for NES audio. Supports up to 8 channels (thanks to the late-model chip design in NES carts) over the standard 6.

If you have never used a tracker, this tool is more like a wooden club with nails when you need a surgical knife. The results are as close as you can get to the real thing without actually programming on the real thing. I've spent months trying to perfect the perfect NES synths using subtractive synthesis only to come maybe 98% accurate where this tracker is like 99.9999%

Good stuff all around! BFXR, SFXR, Famitracker.

 

[-Winnie-]

I started adding some obstacles. I am faced with a stylistic choice.

The case on the right uses the Advance Wars style of drawing units overlaid on the map even if there are tall obstacles one row down. My choices seem to be:

1. Draw units behind obstacles, at least when the units are not selected. This results in the units being partially obscured and potentially difficult to read at a glance.
2. Design all tall obstacles to be exactly two "squares" in height so overlap is not a concern. They might look weirdly tall or fat in this case.
3. Design all obstacles to be exactly one "square" in height so overlap is not a concern. They might look weirdly short in this case.
4. Overlay units on top of tall obstacles, even though this looks a bit weird.

Opinions are welcome!

Agreed that the overlapping doesn't look right. You don't really need to see the entire unit unless they're in focus, at which point some transparency on anything that's obscuring it would probably look okay.

 

[2+2=5]

I started adding some obstacles. I am faced with a stylistic choice.

The case on the right uses the Advance Wars style of drawing units overlaid on the map even if there are tall obstacles one row down. My choices seem to be:

1. Draw units behind obstacles, at least when the units are not selected. This results in the units being partially obscured and potentially difficult to read at a glance.
2. Design all tall obstacles to be exactly two "squares" in height so overlap is not a concern. They might look weirdly tall or fat in this case.
3. Design all obstacles to be exactly one "square" in height so overlap is not a concern. They might look weirdly short in this case.
4. Overlay units on top of tall obstacles, even though this looks a bit weird.

Opinions are welcome!

1 is the only solution imo.
Look at Advance Wars

Not all the objects are contained in a square, for example on the left you can see some mountains covering other mountains.

This may be only personal preference, but imo only big objects like buildings, mountains, forests etc should be contained in a single square because they are "symbolic".

If you go this route(being no more forced to contain things in a square) then i suggest you to consider putting vehicles at the center of the square, and not at the bottom to fit them in a square.

 

[Adam Scott Aukerman]

PC/Chrome

Speaking of audio - freesound is great. As a former audio producer, I like using subtractive, additive and FM synthesis for my 8-bit style sounds. But if you don't have the software or equipment you can cheat a little bit: BFXR and SFXR. They are like mini swiss army knives of bleeps, bloops and plings of various waveforms. No 2 sounds are ever alike.

Then for the more advanced "legit" sound: Famitracker
It has a pretty steep learning curve and is basically a tracker for NES audio. Supports up to 8 channels (thanks to the late-model chip design in NES carts) over the standard 6.

If you have never used a tracker, this tool is more like a wooden club with nails when you need a surgical knife. The results are as close as you can get to the real thing without actually programming on the real thing. I've spent months trying to perfect the perfect NES synths using subtractive synthesis only to come maybe 98% accurate where this tracker is like 99.9999%

Good stuff all around! BFXR, SFXR, Famitracker.

Interesting, when I next get near a PC I'll have a look into that.

I'm already away of SFXR and BFXR. In fact, that's what I used to create the boat engine and noises! I've also messed around with Bosca Ceoil, but my musical talent is so severely limited that I'd have just as much luck creating good music with a couple of bricks and my face.

Famitracker sounds... daunting.

 

[Blizzard]

Thanks for the comments thus far.

Introduce opacity based on what is the current unit! If the one being obscured is the current unity, change the alpha of the unit obscuring it to half, and now you get the full view of your current unit. And vica versa.
(edit: just realized you were talking about obstacles. In which case, obscuring your unit when it is not selected should be fine, and looks fine as well, imho! The only question is whether you can/want to implement transparency when that unit is your focus.)

About your edit, I actually implemented that as a test after posting.

If a unit is selected (e.g. preparing to move a unit), OR if the current cursor is on a unit (e.g. picking an attack target), the unit gets drawn on top, looking like the right picture. In all other cases, the units look like the left picture.

If people like this idea, that is certainly easier than the alternative of making special case partial transparency for the row below. However, partial transparency might look better, so I should still probably prototype it. >_>

 

[JulianImp]

Interesting, when I next get near a PC I'll have a look into that.

I'm already away of SFXR and BFXR. In fact, that's what I used to create the boat engine and noises! I've also messed around with Bosca Ceoil, but my musical talent is so severely limited that I'd have just as much luck creating good music with a couple of bricks and my face.

Famitracker sounds... daunting.

I've used DreamStation 1 in the past, which is currently free since the company behind it has released DreamStation 2 several years ago. to me, it was simpler to use than Famitracker if you want to stick with basic wave functions (just choose the first wave function and turn all optional consoles off), and it was quite serviceable.

Also, IndieDB has authorized my news update a short while ago, where I talked about adding bézier curve support to my level editing tools. If anyone wants to take a look, it can be found here: http://www.indiedb.com/games/quarkstorm/news/quark-storm-update-turbo-smooth-edition. As always, comments, critiques and retweets will be greatly appreciated.

 

[More_Badass]

Would you participate in a Screenshot Saturday thread if I started one here? Other gaming forums have one - Reddit, TIGForum, etc. - and considering just how many developers are active on GAF, I thought it might be interesting/worth starting one.

I'd most likely also be adding to the thread with the usual Screenshot Saturday posts I do for the Indie Thread

 

[JulianImp]

Would you participate in a Screenshot Saturday thread if I started one here? Other gaming forums have one - Reddit, TIGForum, etc. - and considering just how many developers are active on GAF, I thought it might be interesting/worth starting one.

I'd most likely also be adding to the thread with the usual Screenshot Saturday posts I do for the Indie Thread

I guess it could potentially work, but I'm not sure if dilluting our posts among one more thread would be worth it, given that this one doesn't grow at ridiculous speeds anyway. Perhaps you could, say, make a post here every Saturday night with a compilation of screenshots uploaded thorough the week?

 

[Jacksinthe]

Would you participate in a Screenshot Saturday thread if I started one here? Other gaming forums have one - Reddit, TIGForum, etc. - and considering just how many developers are active on GAF, I thought it might be interesting/worth starting one.

I'd most likely also be adding to the thread with the usual Screenshot Saturday posts I do for the Indie Thread

We'd be down.

 

[GulAtiCa]

Been spending a lot of time adding in stuff/fixing stuff up based upon user feedback. Miiverse is a great tool for developers to see what issues users might be having (not just bugs, but regular stuff people may be confused with or having issue with).

Edit: I'd be down with a Screenshot Saturday type thread as well.

 

[missile]

I like flat user-interface styles a lot. Who's with me?

Have spent some more time working on the UI internals esp. on things like
focus-follows-mouse, selections, button/mouse states/messages/events etc.
Isn't easy to manage them all if you want to have everything going at the same
time. Still not perfect, but things do come together. Edit boxes can now have
a delete button as well. List boxes up next, I guess. Once implemented I will
try some theme switching. Next to some high-res themes, I'm especially
interested in building some very low-res themes such that the UI can also to
be used on very low-resolution displays/framebuffers like for example 320x240
down to 64x64.