• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

GAF Indie Game Development Thread 2: High Res Work for Low Res Pay

Status
Not open for further replies.
hi guys. I need advice on how can I modify these existing art we have so that Square won't come after us?
Originally we were trying to spoof the FF7 characters but square warned us they might not be comfortable with this idea. Instead of having these assets wasted I'm thinking we can still use it by perhaps having Barret have lighter skin tone and for Cloud to totally change his hair?

u6ck7LR.gif

tFqBhbX.gif

Bq79Pti.gif
 
hi guys. I need advice on how can I modify these existing art we have so that Square won't come after us?
Originally we were trying to spoof the FF7 characters but square warned us they might not be comfortable with this idea. Instead of having these assets wasted I'm thinking we can still use it by perhaps having Barret have lighter skin tone and for Cloud to totally change his hair?

I would avoid this. It could be considered whitewashing. I would change his body type to be thinner and/or change his clothes (perhaps give him an undershirt?). Or literally anything else BUT his skin tone.

Cool artwork by the way.
 

Popstar

Member
hi guys. I need advice on how can I modify these existing art we have so that Square won't come after us?
Originally we were trying to spoof the FF7 characters but square warned us they might not be comfortable with this idea. Instead of having these assets wasted I'm thinking we can still use it by perhaps having Barret have lighter skin tone and for Cloud to totally change his hair?
For Barret I think you could lighten his skin a touch and make a couple small changes to his head and turn him into someone from a Indian film.

I agree that just changing the hair on Cloud might be enough.
 

Blizzard

Banned
I really wanted A.I

I might participate if Hallucination wins.
I voted for Artificial Intelligence, though trying to do something interesting with AI in 48 hours would probably be really difficult.

On the bright side, the biggest theme winner thus far is A World In The Skies. I'd love that to win even though I imagine it would require a lot of time spent on art. :p
 
hi guys. I need advice on how can I modify these existing art we have so that Square won't come after us?
Originally we were trying to spoof the FF7 characters but square warned us they might not be comfortable with this idea. Instead of having these assets wasted I'm thinking we can still use it by perhaps having Barret have lighter skin tone and for Cloud to totally change his hair?

Give Barret a machine gun leg and remove the machinegun arm. Change his clothes a bit(give him one more layer).

And instead of giving Cloud a big ass sword give him a big ass razor, because he is such an emo. Just joking. Give him a different weapon and again give him an extra layer of clothes like a leather jacket. And let the hair stay. Its not like Square has a patent for generic anime hair.

Edit: Let the clothes like they are. They look totally different than FF7.
 
For Barret I think you could lighten his skin a touch and make a couple small changes to his head and turn him into someone from a Indian film.

I agree that just changing the hair on Cloud might be enough.
I would avoid this. It could be considered whitewashing. I would change his body type to be thinner and/or change his clothes (perhaps give him an undershirt?). Or literally anything else BUT his skin tone.

Cool artwork by the way.
Give Barret a machine gun leg and remove the machinegun arm. Change his clothes a bit(give him one more layer).

And instead of giving Cloud a big ass sword give him a big ass razor, because he is such an emo. Just joking. Give him a different weapon and again give him an extra layer of clothes like a leather jacket. And let the hair stay. Its not like Square has a patent for generic anime hair.

Edit: Let the clothes like they are. They look totally different than FF7.

Thanks for the feedback. I will pass these along to my teammates.
I lol-ed at the machine gun leg haha.
The original idea is that they were supposed to have more of a medieval feel but I guess the artist interpreted differently.
 

Pehesse

Member
I'm planning to dig into animating my characters soonish, any recommendations on 2D animation programs? Glanced at Spine and looks great, though Unity's built-in animation tools might be enough for me at the moment. Any thoughts?

NO don't listen to the others, you have to do it the traditional frame by frame way, ideally while whipping yourself!

I'll throw in another vote for Spriter (and by extension Spine from what I've seen/heard). Don't know how well they mesh with the different game engines, but it should save you quite a bit of time. It all really depends on the 2D aesthetic you're going for, I'd say.

Do these plants look too much like sperm?

CMvUDDNUYAAWjgU.png:large


(This was unintentional, the penis plant was intentional, in any case, my game seems to have some real subtext going on)

They... kinda do, I'll admit. But if it works in context, hey, why not? And now that you've mentioned mushrooms I can see those as well. Their perception will depend on movement, I guess? Will you have them static, or swaying left/right, or... I don't know, bouncing up/down? (that'd give off massive spermy vibes).

I've been playing my game as it stands in its current form for over a year. I have beyond lost all objectivity. In fact, I'm starting to hate it. :) I was talking about this before, but when you're this far in the shit, you can't see what's up or down, you just have to "fly by instruments" like a pilot in a storm. What are the design principles you're following, what are the design principles you believe in, etc...

Agreed. I wonder how much the enjoyment you get out of playtesting your own stuff is a sign that you're making something "good", or does everyone end up hating their own stuff? Read stories about devs only ever playing their own WIP games during their break, etc. I certainly couldn't do that (full disclosure: I'm getting to the hate part myself as well).

It's really sad how the one person that can't enjoy your game is you! Hopefully, after a few months/years, once you've got fresher eyes, it can become enjoyable again... though you can never really turn off the creator part of your brain that made it all happen.

hi guys. I need advice on how can I modify these existing art we have so that Square won't come after us?
Originally we were trying to spoof the FF7 characters but square warned us they might not be comfortable with this idea. Instead of having these assets wasted I'm thinking we can still use it by perhaps having Barret have lighter skin tone and for Cloud to totally change his hair?

Did you contact Square, or did they contact you?

If they stumbled upon your work and contacted you then the situation could be serious, but if it's the other way around, I'd say you have a bit more leeway. I already think Cloud/Red are different enough, anime hair being what it is. Barret might just need a hair change, or anything else that'd be distinctive added on top of the current design.. A new design element to focus attention on, instead of the gun-arm? Though as said above, don't change his skin color!

(Like the style, too!)
 
Did you contact Square, or did they contact you?

If they stumbled upon your work and contacted you then the situation could be serious, but if it's the other way around, I'd say you have a bit more leeway. I already think Cloud/Red are different enough, anime hair being what it is. Barret might just need a hair change, or anything else that'd be distinctive added on top of the current design.. A new design element to focus attention on, instead of the gun-arm? Though as said above, don't change his skin color!

(Like the style, too!)

We are part of Square Enix collective. Here is our pitch for the game in their website so we are in constant communication with them. We were supposed to post an update regarding the different Hero Parties you may encounter and they gave us a heads up that their Japanese counterparts may not be too happy with the ongoing spoof. Thanks for the suggestions again. :)
 

Jobbs

Banned
BTW, I de-spermed it a bit. Also added some more FG elements. This area has a lot more parallax layers than my average room. I don't usually set up things as complex as this because stencyl makes it pretty tedious to do so.

 

Vanguard

Member
Anyone else having a problem in vs2015 where it just ignores formatting settings? Have it set so it should leave stuff on one line and it's just not... at all. Keeps formatting it regardless, don't have resharper or anything. Just wondering if anyone else has the issue :\
 

Blizzard

Banned
What is a decision matrix (AKA "trade matrix" AKA "do a trade")?
A decision matrix is a numerical system that allows you to compare solutions, choose one, and record the factors behind that decision.

Any particular thing you've used this on recently?

For folks interested in that kind of thing I recommend "Thinker's Toolkit: 14 powerful techniques for problem solving" by Morgan Jones who apparently has worked as an analyst for the CIA. I randomly bought it a long time ago for casual reading. It's easy to digest and covers decision matrices among other things, practical examples and applications. Doesn't require engineering or math knowledge.

Here are some rough numbers I came up with:
server_scoreswhlkz.png


I adjusted the numbers after filling it out the first time, since at first the single-threaded system won. I did not expect that since I was thinking that scalability was one of my biggest concerns. Of course I could have made its weight higher and probably had it win that way as well.

If anyone has comments on these server options for a turn-based game, or factors I might have missed etc., I appreciate input.
 

Pehesse

Member
Said I'd post something about the color process, so here it is! It's a lot simpler than the previous posts, but hopefully there are still some info/tricks in there that might be of some use to some of you.

I start by coloring a full single frame to determine the palette, that ends up being something like this:


For the cel-shading aesthetic, I use two-tone shades for each different color type: one for base (=light), one for shadows, and possibly one extra for highlights on specific spots (remember, this isn't a coloring process for a single illustration, details WILL kill you. You're going to be doing this over and over and over so the less colors you have, the faster it'll go/the better you'll feel overall).

To get shadow colors, I use the HSB selection:



For the light tones, I generally try to stay around the 30-45 saturation range. Figuring out "your" range is how you get an homogenous palette overall, and having that is a big part of having a cohesive aesthetic. So this part is pretty important!

For the shadow tint, I overall raise the saturation by a couple %, and decrease the brightness by a couple more (generally 2x the raise for saturation).

HOWEVER, and this is where it's important, it's also quite good to change the hue a bit: to the left if you want something "darker", or to the right if you want something lighter (or at least, that's how I interpret it). The slight difference in hue between the light and shadow will help add volume. Since you only have two tones, you have to pick them carefully!

Once you have your palette, I find it best to make it into a set of swatches, like so:


(Here, the Big Blue palette are the bottom two lines - the others above are other palettes for other chars).

Now that you have your palette handy and swatched, it's time to start painting all of the remaining frames!

The process is fairly simple: first paint in the base color, then your shade tint using the mask layers I talked about in a previous post.

In detail: to paint the base layer, I tried several processes, but I find the one that works best is using the lasso tool while zoomed in to have leeway in your precision when tracing the line, then filling it in with the color pot. For smaller details/patches of color, though, directly painting it can be just as fast. Your mileage may vary!


The main idea here for speed's sake is to keep the same color selected for as long as you can, to avoid going back and forth to your swatches and swapping tools. This is why I do a layer of the color for each frame of the animation: here, I'll do the skin light tone for each frame, before moving on to shadowing, etc. Like so:


I eventually fill in the other colors, following a non-adjacent color principle: I'll put on the same layer most of the colors that are generally non adjacent. There will be some overlap during specific moves, of course, but we're trying to design a generic process here!


The other thing to note is that as I add more layers, I try to eventually merge them down when I complete a cycle (reach the end of the animation and get back to the beginning for a new layer). I keep the skin layer separate, but for everything regarding clothing, I eventually merge it down into a single layer to save on memory use.


As you're trucking along in might be a good idea to toggle the background between two different contrasted colors to see what's out of the lines, if you care about having eveything fit in.


The last step is to add some simple gradients to add some volume with a low time cost. Again, mask layers are key!

And this is the end result:

GrizzledIndelibleGar.gif


That's about it, really. It's a lot of cycling over and over the whole animation, slowly filling in the colors, and trying to be as efficient with your hand movement as possible. The more mechanical and mindless the process becomes, the faster it goes! Hopefully, with all that, you too can become an unfeeling, uncaring coloring machine!
 

Bloodember

Member
When I'm using Visual Studio Community 2015 it hangs quite a lot saying Not Responding for a few seconds. It's getting on my nerves. I have it being used with Unity 5. I have it updated, the only add-on I've installed from the Internet is "Error Catcher". Anyone know why this is doing this? I love VS but if this is going to persist I'll just go back to Mono Development.
 

2+2=5

The Amiga Brotherhood
Now we're arguing in circles as we're clearly not discussing the same point.

Correct me if I'm wrong, but you're talking about how it's easy to be fast when you don't have to take many features into account, and that you would rather take more features over speed. This is all well and good, but is also orthogonal to the point I raised which limited itself to explain why the brush tool is slow to begin with; That is, poor time complexity.

No, i'm just saying that because the brush is complex it takes more time than a simple one, i didn't say one is better than the other, a simple and fast brush is good for pixel art while a complex one is better for paint, even if slower.

Complex algorithms require more time than simple algorithms, have you ever thought that maybe that time complexity is the best that experienced and well paid engineers working for years on the absolute market leader product(i'm talking about photoshop, what started the discussion) can achieve without making the brush simpler?
 

anteevy

Member
I find difficulty to be a very hard thing to access when I'm playing since I've played this levels over and over again and I pretty much know all the paths and stuff. But that's why there's play testings!
Yeah, that's quite a problem for me with my game. My testers always tell me how frustratingly hard some levels are and how they needed several hundred respawns to beat one of these while I have no problems to finish it without respawning even once. But I fear that by making these levels too easy, they will feel less rewarding to some more skilled players. In the end it's just testing, testing, testing until it works for most of the testers. I feel like I'm blind and need others to tell me what I see.
 
And this is the end result:

GrizzledIndelibleGar.gif


That's about it, really. It's a lot of cycling over and over the whole animation, slowly filling in the colors, and trying to be as efficient with your hand movement as possible. The more mechanical and mindless the process becomes, the faster it goes! Hopefully, with all that, you too can become an unfeeling, uncaring coloring machine!

First and foremost, let me say, thanks for the fantastic tutorial again, creating a color pallete can become really handy, even if im not used to it.
Time to give something back to this community that has given me so many greast tips.
Don't know if it would be faster this way for you, but when working in scribblenauts we learned a really fast process of painting basic colors (sorry for having spanish photoshop to make the tutorial):

-Get the wand tool (its usually shit with this things, but keep on with me), and select the OUTSIDE of the outline of your figure.

-Then choose Selection-Modify-Expand... (I have a hotkey for this, but you will need to add it as photoshop doesnt have it in a fresh install, I chose control+alt+1 for it), and choose a number, as you are working at hight resolution 3px would be a good number.

-Right mouse click and select reverse. You will see the selection outline is inside your black lines.

-Create a new layer below the one with the black outline, get the fill tool and use the color that your figure has most (or the base color). You can erase the selection line now as you wont need it anymore (control+d). You can also erase some of the problematic filler the color has done (you can see it here in the part between the neck and hair), BTW at higher resolutions, less problems with this last part becuase the wand can get easily to those tiny places.

-In the layers window choose the icon "block transparent pixels" (the one that looks like a chessboard).

-Now you can add the rest of the colors easily using the example used in the gif. Create different layers for each color or zone. Make this layers masks by Create clipping mask. After that you can color fast winning time by not having to be precise. For another color or zone just create a new layer and repeat the same create clipping mask process.

-As I see in your tutorial, you use light and dark colors in your "zones". Another tip for doing that fast with my example is that the layer you just painted with the create clipping mask, you choose block transparent pixels for it. And then add your dark and bright colors that will not get out of your original layer, even coloring fast.

I use a more simple method that what pehesse is using for coloring darks and lights. A more flat cartoon shadind, but I think the method will work in both basic flat coloring and a more complex shadowing.
In my coloring, after finishing all the basic 1 color layers, I add one last mask layer, and choosing full black color, I make that layer 30% transparency. With the black color brush I then add all the simple shadowing. This is the ending result:
capitanapis2w.png


Of course this tutorial is not only for Pehesse, I recommend anyone to try it and see if they are faster by doing so. I can say that it became really handy for me, but other people may have other fast symtem to cope with coloring.
The key thing in this tutorial though is the first part, the one using the wand, after creating your base color super fast, you can use other methods you are more comfortable using.
 

Pehesse

Member

Wow, very nice tips! I'll try some of those, especially the wand thing, see if it can't help move things along even faster :-D Thanks a lot!

EDIT: ...and yeah, after trying it just now I remember why I didn't go that way - it doesn't work with the way I line things, since I don't always properly close the lines (some sort of extreme thick/thinning lines :-D), so the wand selection ends up all over the place. But still! Very cool info and process, and I'll try to see if it works better with cleaner defined shapes at some point :-D
 
Wow, very nice tips! I'll try some of those, especially the wand thing, see if it can't help move things along even faster :-D Thanks a lot!

EDIT: ...and yeah, after trying it just now I remember why I didn't go that way - it doesn't work with the way I line things, since I don't always properly close the lines (some sort of extreme thick/thinning lines :-D), so the wand selection ends up all over the place. But still! Very cool info and process, and I'll try to see if it works better with cleaner defined shapes at some point :-D

Oh yeah! didnt remember about that in the other tutorial. Yep, the only way this is faster is if you colse all the lines, so more free lining will not work.
Well, at least I hope it can help other people or even you can use it in some other way :)
 
hi guys. I need advice on how can I modify these existing art we have so that Square won't come after us?
Originally we were trying to spoof the FF7 characters but square warned us they might not be comfortable with this idea. Instead of having these assets wasted I'm thinking we can still use it by perhaps having Barret have lighter skin tone and for Cloud to totally change his hair?

Honestly? I don't think you need to change anything. As long as it's not 1:1 their characters, they can't actually do anything to you. They can huff and puff a little, but let's be honest, here: unless your game is a success on the same scale as Minecraft, they probably won't notice or care.

It's going to cost them more money to get together any kind of legal action against you than they'd probably win in a court case. So they just won't bother. Plus, fair use does protect parody to an extent. There was a big deal with Coolio getting really mad about Weird Al's "Amish Paradise" back in the day, but since it was a parody of Gangsta Paradise and not a direct rip, Weird Al was in the clear.

https://en.wikipedia.org/wiki/Campbell_v._Acuff-Rose_Music,_Inc.

All three of your characters here are far and away different enough for you not to have to worry about anything. I'd argue that changing them any more than they are might actually make it hard to tell it's supposed to be a spoof, actually.
 

SystemBug

Member
Not really art for a specific game, but i have been working on this 3d sculpt of a head. actually finished this months ago but I have been so busy with my job (which sucks, the job and the lack of time) that I didnt get a chance to make a new render

20700698552_61567f09bb_b.jpg
 

mStudios

Member
I'll admit. When I see others doing cel-shading it looks amazing.
When I do it myself I never like the result.

It might be that I'm not used to cel-shading or something, but I never like my own result.
 

JNT

Member
Complex algorithms require more time than simple algorithms, have you ever thought that maybe that time complexity is the best that experienced and well paid engineers working for years on the absolute market leader product(i'm talking about photoshop, what started the discussion) can achieve without making the brush simpler?

Yes, I have. And, no, it's not relevant to the discussion of why the brush is slow.
 

EDarkness

Member
I'll admit. When I see others doing cel-shading it looks amazing.
When I do it myself I never like the result.

It might be that I'm not used to cel-shading or something, but I never like my own result.

I'm the same way. I want my next project to use some sort of cell shading, but I can never make it look good. Just need to play with it more, I guess.
 

Cptkrush

Member
I sat down this morning and started a spreadsheet for my ideas for my game so far. I ended up getting some serious writers block, but what I have written so far I am very happy with. It definitely reads like a game that I would love to play, but execution is everything of course. If anyone is interested in checking it out, it's on my public Drive here: http://bit.ly/1PzMNGC

I'd of course love to hear any feedback, suggestions, or whatever. Now I need to think of missions, mission structure, and a timeline so I can keep myself organized. I've been jokingly calling the game Grand Theft Spaceship, and Sci Fi Tropes the game, but I think those are both accurate in this situation.
 

Blizzard

Banned
Yes, I have. And, no, it's not relevant to the discussion of why the brush is slow.
This discussion has confused me since what 2+2=5 has been saying makes sense to me.

The question was, why is Photoshop slow?
The answer suggested was (I think?), here's a comparison with a very simple brush using a GPU algorithm compared to another image editor. Maybe Photoshop has similar problems to the GIMP.

To me that does not answer the question -- what is being suggested?
  • Are you suggesting Photoshop does not GPU-accelerate its rendering?
  • Are you suggesting Photoshop uses a poorly chosen brush algorithm?
  • Are you suggesting Photoshop uses a valid, but complex algorithm?
I'm not sure any productive discussion can happen unless this is clarified as a starting point.



On another note, someone just pointed out to me that Blender has pie context menus if you enable an official add-on:
http://wiki.blender.org/index.php/Dev:Ref/Release_Notes/2.72/UI/Pie_Menus

This looks pretty handy for tab and Z especially, and maybe Q as well.
*edit* Screenshot:
600px-pie_menu_manualy0y3u.png
 
Honestly? I don't think you need to change anything. As long as it's not 1:1 their characters, they can't actually do anything to you. They can huff and puff a little, but let's be honest, here: unless your game is a success on the same scale as Minecraft, they probably won't notice or care.

It's going to cost them more money to get together any kind of legal action against you than they'd probably win in a court case. So they just won't bother. Plus, fair use does protect parody to an extent. There was a big deal with Coolio getting really mad about Weird Al's "Amish Paradise" back in the day, but since it was a parody of Gangsta Paradise and not a direct rip, Weird Al was in the clear.

https://en.wikipedia.org/wiki/Campbell_v._Acuff-Rose_Music,_Inc.

All three of your characters here are far and away different enough for you not to have to worry about anything. I'd argue that changing them any more than they are might actually make it hard to tell it's supposed to be a spoof, actually.

Im with you.

Don't change anything rhophiehalul78. The can't and will not do anything.
I wouldnt have even asked them permission though. If its not a small company that usually likes to do this stuff they will say no everytime they are asked.
 

Popstar

Member
Im with you.

Don't change anything rhophiehalul78. The can't and will not do anything.
I wouldnt have even asked them permission though. If its not a small company that usually likes to do this stuff they will say no everytime they are asked.
I think the two of you missed the post where he mentions being part of the Square Enix Collective program.

He's just trying to keep a good relationship with a partner.
 

Jumplion

Member
NO don't listen to the others, you have to do it the traditional frame by frame way, ideally while whipping yourself!

I'll throw in another vote for Spriter (and by extension Spine from what I've seen/heard). Don't know how well they mesh with the different game engines, but it should save you quite a bit of time. It all really depends on the 2D aesthetic you're going for, I'd say.

I'm literally going for a stick-figure look. Any of those really old stick figure animations in flash, or Pivot (which brings back so many memories of messing around with it in middle school), is what I'm planning on doing since I can't draw to save my life. I'm thinking of experimenting a bit with it, maybe rotoscoping on top of some footage to get the exact feel I want for it.
 

rje

Member
Trying to work on mod tools today, but getting frustrated by a lack of good solutions for cross platform UI work in C# - does anyone have any recommendations? The best choices I've found so far are GTK# and XWT. If you've used either library I'd love some impressions on how you liked it.

In a perfect world I'd write native UI for each platform, but we're a 2 man dev team and I can't possibly justify the dev time on that. So looking for alternatives if anyone has them.
 
Thanks for the coloring tips, Pehesse and SpacePirate Ridley!

My main concern right now is figuring out how to develop a good limited color palette for pixel-style characters (even though mine are voxels). There is some info I've seen around the web but there appears to be a lot of "art" to it that I'm still trying to wrap my head around.
 

UsagiWare

Neo Member
Yeah, that's quite a problem for me with my game. My testers always tell me how frustratingly hard some levels are and how they needed several hundred respawns to beat one of these while I have no problems to finish it without respawning even once. But I fear that by making these levels too easy, they will feel less rewarding to some more skilled players. In the end it's just testing, testing, testing until it works for most of the testers. I feel like I'm blind and need others to tell me what I see.

Do you watch your playtesters in action?
Sometimes it can be due to an isolated difficulty spike in a level that makes it really hard, or they approach/solve a situation completely different than you would.


@Pehesse and SpacePirate Ridley:
Interesting read! thanks for sharing your workflows like that,
 

JNT

Member
This discussion has confused me since what 2+2=5 has been saying makes sense to me.

The question was, why is Photoshop slow?
The answer suggested was (I think?), here's a comparison with a very simple brush using a GPU algorithm compared to another image editor. Maybe Photoshop has similar problems to the GIMP.

To me that does not answer the question -- what is being suggested?
  • Are you suggesting Photoshop does not GPU-accelerate its rendering?
  • Are you suggesting Photoshop uses a poorly chosen brush algorithm?
  • Are you suggesting Photoshop uses a valid, but complex algorithm?
I'm not sure any productive discussion can happen unless this is clarified as a starting point.

Nothing 2+2=5 has said is wrong. It's just besides the point.

The point, which I sought to clarify, was why the Photoshop brush is slow. The comparison between the simple brush and the complex brush was never the point. Ever. I wanted the readers to fix their attention to this snippet from the article I linked:

In its simplest form, the brush tool requires a circle filling algorithm. Given a mouse click position and a brush of diameter n, the program has to perform a distance test on a square of size n x n around the mouse. If the distance from the given pixel is less than or equal to the radius, the pixel has to be filled. This algorithm becomes exponentially slower as n increases linearly. Things get even worse when you consider that the user drags the brush around. So instead of just filling in a circle, you have to fill in all the circles along a line between the user’s last drag position and the user’s current drag position.

It's a short and concise point which clarifies why something seemingly simple is slow. To me, this does answer the question "Why is the Photoshop brush slow?". The answer, in case you missed it, is "because the algorithm used to fill the brush is order n-squared complexity, where n is brush radius" (I've extrapolated this conclusion from the article, so it's not necessarily true, but to me it doesn't seem unthinkable). You can disregard the rest of the text in the article as it doesn't lead to another conclusion.

As for what I'm suggesting; None of the above. We already know Photoshop has GPU accelerated rendering, so what's the point of suggesting otherwise? However, we also know that O(n²) scales poorly regardless of what silicon it is being executed on. As for suggesting that the chosen algorithm is poorly chosen, valid, and/or complex; If I were suggesting any of those alternatives, would that somehow make it not an O(n²) algorithm? Would it modify the original point I was trying to clarify? The way I see it, the answer would remain the same regardless, which is why I refrain from such value judgements.
 

OldRoutes

Member
Yeah, that's quite a problem for me with my game. My testers always tell me how frustratingly hard some levels are and how they needed several hundred respawns to beat one of these while I have no problems to finish it without respawning even once. But I fear that by making these levels too easy, they will feel less rewarding to some more skilled players. In the end it's just testing, testing, testing until it works for most of the testers. I feel like I'm blind and need others to tell me what I see.

Implement some analytic, maybe? Collect the information on where they die on the map, investigate it if there's a common spot...
 

SeanNoonan

Member
Having one of those "why am I still working on this game" moments.

It'll be interesting to see if this effort will pay off at any point. I just bought another iOS license, so hopefully I can cover that when I'm done. I have a feeling the Jack B. Nimble update won't be able to cover the costs of Adobe CC this year though.

I'm curious; how much do you people bill yourselves hourly when working on your own projects. It seems like sustainability/success is very much out of reach for me anyway. Was just curious about other people's projects and allocated development budgets.
 
Having one of those "why am I still working on this game" moments.

"Because it's fun" should be the main reason if you are pouring your own money. Otherwise bail out, unless you get lucky (i.e. win the app lottery) you are much better working on other areas of software dev.
 

KevinCow

Banned
Almost played through my test level without encountering a major bug.

Almost.

Then on the last battle, an old bug I thought I'd fixed reared its ugly head again.

Ugh.
 

SeanNoonan

Member
"Because it's fun" should be the main reason if you are pouring your own money. Otherwise bail out, unless you get lucky (i.e. win the app lottery) you are much better working on other areas of software dev.
Yeah, I get that; my goal is just to have my games cover their expenses - I'm not billing my time right now though.

I work full time in AAA this really is just a hobby, but it would be nice if I could cover financial costs. The last release did, so hopefully these latest efforts will pay off.

Edit: I just checked my finance reports, and it seems that the game is no longer paying for Adobe CC - the pressure is on for the update to bring in some cash!
 
Hey there, I've been reading through this thread for a while and, I have to say, you guys are an inspiration.
One of my life-long dreams is to make a game by myself, and I've been reading for a while and I choose Unity. I have my share of programming and OOP knowledge, being my job, so I can comfortably code in any given language with proper documentation.

So, I've reading the Unity documentation and reading about how 3D works in a game, but I was wondering if there is something like examples with real-world (albeit simple if needed) applications like what PHP does with their documentation:

http://php.net/manual/en/function.call-user-func-array.php

Example #1 call_user_func_array() example

<?php
function foobar($arg, $arg2) {
echo __FUNCTION__, " got $arg and $arg2\n";
}
class foo {
function bar($arg, $arg2) {
echo __METHOD__, " got $arg and $arg2\n";
}
}


// Call the foobar() function with 2 arguments
call_user_func_array("foobar", array("one", "two"));

// Call the $foo->bar() method with 2 arguments
$foo = new foo;
call_user_func_array(array($foo, "bar"), array("three", "four"));
?>
The above example will output something similar to:

foobar got one and two
foo::bar got three and four

I think I know how to make everything using OOP notions, but my problem right now it's how to "materialize" my ideas.
 
I'm sick of constantly breaking my Unity 5 project without the ability to roll back.
I used to use perforce and SVN but decided Git and Source Tree might be the trick for me (need something with a GUI) as a non-coder.

Any tops on how to organise it? I just want to have the repository on a different local drive than the one I'm working on, not really interested in having it online just yet.
 

Blizzard

Banned
The final round of Ludum Dare themes is up for voting. I plan to make a thread tomorrow if there's still interest in that, just in case there are people participating who aren't already reading this thread.
 

bkw

Member
I'm sick of constantly breaking my Unity 5 project without the ability to roll back.
I used to use perforce and SVN but decided Git and Source Tree might be the trick for me (need something with a GUI) as a non-coder.

Any tops on how to organise it? I just want to have the repository on a different local drive than the one I'm working on, not really interested in having it online just yet.
So you want to have one repository as your working one, and then another on a separate repository on a different local drive where you push and pull (sync) your commits to? So they're both identical except for the fact that they're on separate drives?

I'd create your git repo on your "different local drive", then clone it to where you want your working repo to be. Then you can make commits on your working repo, and push your changes to the original repo when needed. I can get into SourceTree steps if you need them.

I don't really see the purpose in doing this two repo setup though, unless there's some special back up/sync that happens on the second repo. Might as well setup a private repo on BitBucket and push to there, or just work with the one repo. Not sure how familiar you are with Git vs. Perforce/SVN. Unless I'm mistaken, Git is not like Perforce and SVN where you have a single central repository that you check things in and out of. You own your own repo, and work on it directly.
 
So you want to have one repository as your working one, and then another on a separate repository on a different local drive where you push and pull (sync) your commits to? So they're both identical except for the fact that they're on separate drives?

I'd create your git repo on your "different local drive", then clone it to where you want your working repo to be. Then you can make commits on your working repo, and push your changes to the original repo when needed. I can get into SourceTree steps if you need them.

I don't really see the purpose in doing this two repo setup though, unless there's some special back up/sync that happens on the second repo. Might as well setup a private repo on BitBucket and push to there, or just work with the one repo. Not sure how familiar you are with Git vs. Perforce/SVN. Unless I'm mistaken, Git is not like Perforce and SVN where you have a single central repository that you check things in and out of. You own your own repo, and work on it directly.

I basically just want a simple workflow that allows me undo/revert when something breaks like an asset update. preferably ona separate drive to my main unity drive but local (my monthly internet quota would break if I used online backups).

if you feel like going into details, by all means. i would appreciate that.
 

Retroid

Member
The final round of Ludum Dare themes is up for voting. I plan to make a thread tomorrow if there's still interest in that, just in case there are people participating who aren't already reading this thread.

That would be cool! I'm hoping to participate this time, doubt I'll finish to my liking but hopefully I'll get something playable.
 

Jobbs

Banned
Spent all day revamping my whole character progression system. I've had a bit of anxiety and uncertainy around the pet aspect for some time, where it felt like it never quite had as much purpose as I wanted.

I'm working on a fundamental shift in how stats and levels work. Previously, you leveled up your character by killing enemies, and evolved dizzy by feeding him edibles (which could alternatively be consumed for stats yourself).

It occured to me a better way to go would be to take this "choose to consume or feed" concept to its logical conclusion.

Under the new system, killing enemies no longer grants any XP. The character gains XP by consuming edibles, and the pet gains XP by being fed those edibles. So it's a constant choice -- Alternate progression tracks.

Generally speaking, consuming edibles and leveling up grants you more reliable, predictable, and generally "defense oriented" stats, whereas feeding DIzzy contributes to his eventual evolution. It takes more food to make him evolve than it does to make teh character level up, but his evolutions can potentially have much greater bonuses and in some cases grant special powers. The "feed pet" upgrade path is therefore less predictable and more offense oriented.

I've spent the day working on the specifics of the system. Because there's now a finite source of XP (no edibles respawn), I needed to more carefully plot what their values are and what the required XP curve was.

Dizzy evolves when his overall exp is high enough, and the form he evolves to will be one from the next "tier", with there being 5 total tiers (so he can evolve a maximum of 5 times). Edibles, depending on the type, grant Dizzy varying amounts of three internal values, which I call "type 1", "type 2", and "special". Special is obtained from rare edibles. Which form from the next tier he evolves into will be based on these internal values. In some cases, extra conditions are required to achieve rare forms.

While I may write in vague hints from NPCs or something, generally speaking, none of how this works (when it comes to how evolving actually works) will be explained to the player. It'll just be a big mystery, though not entirely unintuitive. For example, type 1 are all edibles with certain things in common with eachother, as are type 2, as are special.

It's fun but really overwhelming. The good news is I think I have it mostly sorted out.
 
Can I ask you why you decided on Git? Perforce and SVN are both way more straightforward.

I asked some former colleagues for suggestions and Git was what everyone seems to love these days (since we all have horror stories of SVN and Perforce).

But hey, if going to SVN or P4 will do what I need better than Git, let me know.

As a non-coder, I personally love using Unity for the most part. Some programmers I used to work with won't shut up about how much they hate Unity. It won't do things the way they want it to, or it crashes or whatever. Which is ironic because I remember most of the non-coders saying the exact same thing about the tools and engines they used to provide for us to use.
 

Popstar

Member
It's more Github that everyone loves than Git itself I think.

Subversion or Perforce will do you what you need better than Git. In particular Git only works well with text files and works poorly with binary files. It's much more programmer-centric than than something like Subversion of Perforce.

I use Subversion myself. As a GUI for it I use Cornerstone on OS X and TortoiseSVN on Windows.
 
Status
Not open for further replies.
Top Bottom