Support NeoGAF

[wwm0nkey]

I won't lie, I thought this was hilarious, I really dug the recreation of the a e s t h e t i c (though it lacked cans of Arizona iced tea). The only thing I couldn't figure out is how to actually damage enemies. Holding down LMB made little squares hang out at the front of the "gun", but that's it - no projectiles, nothing to hurt enemies with.

Yeah we didn't get that weapon where we wanted it, it just spawns a huge ass cone that triggers enemy damage. Switch to the CD with "Q" that weapon is more finished

 

[shysaursoft]

Yeah we didn't get that weapon where we wanted it, it just spawns a huge ass cone that triggers enemy damage. Switch to the CD with "Q" that weapon is more finished

Ah, that works better - actually managed to kill a few. I really like the screen desync / scanline effect, it's pretty spot-on.

 

[Makai]

There was a magical moment where I saw the strategy of the game develop. It was dominated by a game-breaking exploit until somebody found a bug that countered it - just like in real fighting games!

 

[wilboo]

Thanks for people's kind words about the HK Switch gif!

This GGJ was a bit of a downer for me. I worked as graphic artist on a puzzler, and then halfway through the one programer that constantly shot down all ideas and changed everything all the time, just switched teams and left the other programmer hanging. Now I wasted 15 hours coming up with designs and the game remained unfinished.

At least my second project got done.

Ahh, that sucks, but if you like the idea you can always put aside like a weekend or so to finish it off if you wanna see the idea through.

 

[dinopoke]

This GGJ was a bit of a downer for me. I worked as graphic artist on a puzzler, and then halfway through the one programer that constantly shot down all ideas and changed everything all the time, just switched teams and left the other programmer hanging. Now I wasted 15 hours coming up with designs and the game remained unfinished.

At least my second project got done.

Yeah, this was my first GGJ and I had a terrible time. Our group was comprised of strangers and everything that could go wrong went wrong (completely changed game ideas halfway through, team members left, morale was low etc). Early Sunday morning, I went back to the original idea and attempted to salvage it and we ended up with this:
http://globalgamejam.org/2017/games/newton-defence

 

[missile]

I improved my bump-mapping code a bit. There were a few bugs in there. One was
the following; for making some cool animation later on, I converted the
tangent frame to a quaternion. The normal in tangent space is (in my case)
(0,1,0), which I use for bump-mapping. Having computed the bump-mapped normal
in tangent space and transforming back using said quaternion produced a few
odd normal here and there. I found out that the quaternion was a bit off in
mapping the (defined) tangent space normal back to world space normal by a few
digits. As it turned out, my fast conversion from matrix to quaternion hasn't
enough precision for this to work properly. xD I fixed that and everything was
just fine. I also tried another way in transforming the world space normal into
tangent space and using this one as the normal instead of (0,1,0), but that
gets a bit messy further down. I also implemented bilinear filtering and fixed
some minor texture bugs.

old

new

good old world map

Also tried fading the bumps with distance. Don't know if that makes sense
further down, but does lessen the aliasing a bit (see second image above as
well) and may also help to reduce specular flickering later on. Will see.

normal map

 

[asa]

Missile, what exactly are you doing? a engine for limited hardware? a lowres game? Your gifs are cool, so I wanna know more

 

[Blizzard]

If I ever manage to carve out some time, I want to dive seriously into UE4 and try to make something.

I've used UDK off and on over the years, and messed briefly with UE4. Are there any references, guides, intros, summaries, tutorials, etc. that are highly recommended? Or should I just dive in and look up stuff when I need it like usual?

 

[Timeaisis]

Hey, all! I have really been missing out on this thread. Been super busy at work and stuff. Everyone's stuff looks great as usual.

Anyway, spent the last weekend readying a greenlight campaign for a game I've been working on off and on for six months. And we're it's on greenlight today! It's called A Quick Death and it's a 2-player competitive western shootout.

Here's the trailer I recently cut:
https://www.youtube.com/watch?v=5KsJ9fwJvfQ

And you can vote for it here: http://steamcommunity.com/sharedfiles/filedetails/?id=846583048

I'd really appreciate it. And it ain't your bag, no sweat. Thanks, everyone!

On my end, I've been super quiet/antisocial as we crunch non-stop towards Hollow Knight's release. I do have this cool GIF that Ari made to announce our Switch release though:

I'm putting achievements in today too which'll be fun to try out.

Man, this looks awesome. Love the art.

 

[FlashbladeGAF]

I had written stuff about the animation and coloring process for Honey - those posts are a bit old now, but I still work mostly the same for Pacha, except with an added sketch layer, and I now use the animation timeline in Photoshop to check the motion as I make it, rather than export the frames to Flash for the same purpose.

Other than that, the process is fairly straightforward: all animation is done in photoshop, when I'm done I export each frame individually (no spritesheet), import them in C2 and let it build the spritesheet automatically. The process can indeed be done in seconds, if that's what you're worried about.

When importing an animation in C2, it goes like this: open object/sprite->new animation->import all frames->set anchor point on frame 1->set anchor point for all animation based on frame 1 (one click option) (this step can take more time if you need to displace the anchor point during the anim)->(add special anchor points if needed, + work the collision boxes if needed - faster to do it before cropping, as you can "apply to all animation" when all frames are still the same size with more predictable results)->crop all frames (one click option)->set animation play rate->set individual frames hang time if needed. All in all, this takes between 10-15 seconds for something with a single defined anchor point and nothing else to tweak to a couple minutes if you need to tweak the collision box on every individual frame of a 40+ frame animation.

The spritesheets themselves are built during export, and I let C2 automate that on its own. It's not the most optimized result possible, but it's doing a good enough job for my purposes, the memory savings if I were to do it myself would be minimal (and the time loss would be monstrous). So all in all, it works well enough that I don't see myself using anything else for the foreseeable future, unless I'm forced to somehow! :-D

What resolution are you working in and how tall/big (pixel wise) will your final character sprites be when it's all said and done? After seeing traditional animated HD games like Cuphead and Indivisible I was curious to know what was a great size for image quality, yet decent enough not to take up alot of memory/space/etc.. (I realize there is no perfect size, just curious what works for you or anyone else doing HD traditional animated sprites)

 

[Tengrave]

Hey DevGAF,

Long time lurker. For the past year a friend and I have been working with Unity to make a Pokemon/Digimon Style game but with adult sensibilities and survival elements.
We want The Last Monsters to play like the best parts of Pokemon and Grandia 2 with a pinch of Oregon Trail. Combat is a blend of realtime and turn-based to allow for more strategic decisions.
[

https://www.youtube.com/watch?v=mFCXlcCcY5Q
https://www.kickstarter.com/projects/2041403205/the-last-monsters
We made it onto Kickstarter and received their "Project's We Love" but are looking for additional feedback about the project and where to go from here. Thanks.

 

[MrHoot]

Hey there

it me again

So i've started my devblog for helvetii, with my very first entry here
https://kevpec.com/2017/01/24/helvetii-dev-blog-1/
It's very exciting for me but also terrifying as an introvert, but I decided to share my road during the dev of this project which is super important to me. And since ya'll are super nice and were very supportive, I thought i'd share with you, maybe you'll find it interesting

 

[Jacksinthe]

So... Who is next?

I've been MIA from the site and from this thread but we are less than 2k posts from thread limit... Which one of you fine developers shall I be passing the reigns to?

 

[Makai]

So... Who is next?

I've been MIA from the site and from this thread but we are less than 2k posts from thread limit... Which one of you fine developers shall I be passing the reigns to?

me :u

 

[LordRaptor]

I think missiles 3d gaf logo and dev globe is part of his opening salvo for election

 

[Jobbs]

I finally added the flame weapon. It's still WIP, but it's mostly done. In particular I think the impact point needs cleaned up a bit and made more interesting.

https://gfycat.com/GrouchyYoungAsiansmallclawedotter

The way it works -- Molten shots make impacts that set fire to anything near. Additionally, enemies already tagged by a shot will allow future shots to pass through, though the ones that pass through will be diminished. Currently, diminished bullets don't start new fires, but I'm starting to think they should.

Enemies can also spread fires to eachother:

https://gfycat.com/ThornyWigglyArabianwildcat

 

[missile]

Missile, what exactly are you doing? a engine for limited hardware? a lowres game? Your gifs are cool, so I wanna know more

Since now that your game is out in the wild you're back at devdesk?
Are you going to write about how things went and stuff?

About the graphics, I wrote a couple of lines about it some time ago;

...
Since a couple of people asked me the same question, I want to explain a bit
about my doings behind all of it. Well, the engine is 100% software rendering
for a reason. The only dependency is the OS layer to get the pixels on the
screen. Since the engine is sort of a testbed for retro graphics, and for
testing some rendering algorithms etc., I currently don't need that much
performance or have any need to work around the limitations/bugs of any
hardware, library, or driver. Software rendering helps me to focus myself on
the core thing and not getting distracted from outside stuff I have no control
over. All faults are my own, and I can fix them! There is another advantage
when developing in software, for, the CPU isn't as fast as the GPU. These
performance constrains make you think a little harder about how to optimize
stuff as you go. But my engine isn't optimize the slightest. I don't do any
optimization, yet, but I do optimize in my head. For example, while
implementing a certain technique/whatever I'm looking for how it turns out and
estimate the restriction it will have on the game/graphics, i.e. when using
more complex resp. efficient data structures making it run much faster. For,
these optimizations will almost always lead to a lot of restriction the engine
needs to adhere to later on. And I need to see if I can still express my
vision with these restriction in place. This approach is a little bit
different from standard where you have a certain vision breaking it down on
existing parts. It's the opposite for me, currently. I'm creating the parts
with a certain vision in my head and see if they align or even enhance that
vision. That's the creative part of game development for me, finding new and
different stuff trying it out towards a certain vision. That's for example
how the pixelized defocusing blur came about. It wouldn't ever happen the
other way around. You would sit down using what everybody does like
accumulation and blending getting your defocus done, because that's the part
the GPU does best. For that reason, of being able to be quite creative doing
lots of experiments etc., it's good to have sort of a base (unoptimized)
software renderer which allows you to shoot in many different directions. But
once all the creativity is set and done, with respect to that vision, I'm
going to map it all onto graphics hardware as good as I can.

To give an idea for a part of my vision I have, considering the look of the
game, I'm after a certain stylized lighting and material reflection (which
is hart to describe) and which isn't possible to compute in an RGB color
space. I want to build special angle dependent reflective and absorption maps
over the spectrum of light combined with a complex index of refraction to
compute the amplitude (Fresnel equation) of said reflection. These reflective
and absorption maps should get modulated (artistically controlled) to produce
some pretty unique/stylized looks also changing with the viewing angle. The
light needs to be spectrally based as well to compute the correct lighting
(color rendering) for the given maps at the given frequency. The spectral
energy will than be converted as seen by an CIE standard observer (XYZ color
space) and from there to any RGB output device. With this technique I can
build a vast set of pretty synthetic material which are pretty hard to
describe (but cool to look at, I hope xD). That's a similar problem describing
in words the reflection of real metals, yet their reflections looks a lot more
stunning than any other reflection (usually plastic).

That's something I want to try out at least. With a software renderer that's
not a big issue, but doing it right on graphics hardware would be cumbersome.
Well, I can't say to what an extent I can realize it in any final game, if it
even works the way I want it, because spectral rendering is way off from what
GPUs are good at. But with all the modern programmable GPUs one can implement
what's needed while still being fast (at least much faster than on a CPU). ...

tl;dr: spectral reflection + adaptive human perception

Well, the bump-mapping isn't really of any interest to me. I'm not after some
cool bump-mapped graphics or something. That's pretty old stuff, yet cool
on its own (Blinn ftw!). Anyhow, I'm just implementing some ways for
controlling the renderer via texture/function mapping making it easier to
control the renderer and for creating certain cool looks later on.

Next I'm going to implement a 3d texture space moving and rotating with the
objects such that the surface point to be textured is taken out of a volume
texture. This will allow for an infinite amount of cool textures and ways to
control the renderer. Also also want to bump-map it.

I think missiles 3d gaf logo and dev globe is part of his opening salvo for election

I would have some interest, but not this time around. This year will get
pretty though for me, for, I went all-in and now nature is going to calls me
on all frequencies, if you know what I mean.

 

[wilboo]

I finally added the flame weapon. It's still WIP, but it's mostly done. In particular I think the impact point needs cleaned up a bit and made more interesting.

https://gfycat.com/GrouchyYoungAsiansmallclawedotter

The way it works -- Molten shots make impacts that set fire to anything near. Additionally, enemies already tagged by a shot will allow future shots to pass through, though the ones that pass through will be diminished. Currently, diminished bullets don't start new fires, but I'm starting to think they should.

Enemies can also spread fires to eachother:

https://gfycat.com/ThornyWigglyArabianwildcat

Looks really good! Fire weapons are always fun. Are enemies gonna be able to set you alight as well? Not sure if that would work with your setup, but could be an interesting trade-off for a powerful weapon.

 

[missile]

I finally added the flame weapon. It's still WIP, but it's mostly done. In particular I think the impact point needs cleaned up a bit and made more interesting.

https://gfycat.com/GrouchyYoungAsiansmallclawedotter

The way it works -- Molten shots make impacts that set fire to anything near. Additionally, enemies already tagged by a shot will allow future shots to pass through, though the ones that pass through will be diminished. Currently, diminished bullets don't start new fires, but I'm starting to think they should.

Enemies can also spread fires to eachother:

https://gfycat.com/ThornyWigglyArabianwildcat

Ha, that's cool! Burn 'em! What I recognized in your game, giving me some
second thoughts, is the similarity in how the shoots look (elliptical). I
also think there could be some more cool trails for the weapons. For example,
considering the flame weapon, what about letting the flame oscillate a bit
like a wave when moving forwards with the particles close-by / within the wake
start to move in a swirling pattern? I think this will also enhance the
perception of any atmosphere the player is in. And as such the effect my vary
with the set atmosphere (viscosity).

 

[Pehesse]

What resolution are you working in and how tall/big (pixel wise) will your final character sprites be when it's all said and done? After seeing traditional animated HD games like Cuphead and Indivisible I was curious to know what was a great size for image quality, yet decent enough not to take up alot of memory/space/etc.. (I realize there is no perfect size, just curious what works for you or anyone else doing HD traditional animated sprites)

Speaking only for Pacha, to target HD resolution (1900x1080/1900x1200 - 4K is simply unrealistic with my means), my base canvas for drawing/animating is 3200x3200, which I reduce to 1000x1000 before exporting individual images (I like the compression artifacts that occur when reducing the image like this, those who won't will likely use an entirely different workflow/initial canvas size to avoid them). Cropping those images in C2 results in the frames being about 250x550 on average for the character standing motionless (the exact size varies based on the motions done, some require almost the full size of the initial canvas, like the warrior's spinning jump, so there's no single "true" number to rely on here). That's for having a rather large sprite on screen, I could probably get away with much smaller sprites if I had the character take up less screenspace!

On average, under these conditions and with an average of 12-14 frames per anim for simple motions (again, this can vary wildly), an animation requires about 10Mb of RAM according to C2's "estimated memory usage". I'm currently sitting at an overall estimated usage of 1.25Gb with only the main character's anims+some very small side stuff, I estimate the whole game to need about 2.5/3 when all is said and done (again, based on C2's metrics).

I'm not exactly sure what you were asking, so I hope these numbers will be of some help!

Hey there

it me again

So i've started my devblog for helvetii, with my very first entry here
https://kevpec.com/2017/01/24/helvetii-dev-blog-1/
It's very exciting for me but also terrifying as an introvert, but I decided to share my road during the dev of this project which is super important to me. And since ya'll are super nice and were very supportive, I thought i'd share with you, maybe you'll find it interesting

Haven't read it yet in full yet, but as a fellow introvert, I can only say: good going! Sharing is tough, but should help in the long run, and makes for a powerful motivator if you make it a habit :-D One small thing, if I may: at the end, the self-deprecation may get a tad bit strong, and feels a bit unnecessary when contrasted with the work you've done so far!

I finally added the flame weapon. It's still WIP, but it's mostly done. In particular I think the impact point needs cleaned up a bit and made more interesting.

https://gfycat.com/GrouchyYoungAsiansmallclawedotter

The way it works -- Molten shots make impacts that set fire to anything near. Additionally, enemies already tagged by a shot will allow future shots to pass through, though the ones that pass through will be diminished. Currently, diminished bullets don't start new fires, but I'm starting to think they should.

Enemies can also spread fires to eachother:

https://gfycat.com/ThornyWigglyArabianwildcat

I was just watching (well, listening to, rather) the latest AGDQ block for Metal Slug+Contra, so this flame shot feels in direct continuity of what I was seeing a moment ago :-D Spread gun when? :-D

I think missiles 3d gaf logo and dev globe is part of his opening salvo for election

Haha, agreed! And even if not, I still think it needs to be part of the next OP! Speaking of next thread, has an OP title been decided yet? If not, in keeping with the current thinking and in reference to the above, how about something like "Targeting 4K ($/y)"

Spoiler

or is that too self-deprecating? :v

 

[missile]

... "Targeting 4K ($/y)" ...

ROFL

 

[Jobbs]

I was just watching (well, listening to, rather) the latest AGDQ block for Metal Slug+Contra, so this flame shot feels in direct continuity of what I was seeing a moment ago :-D Spread gun when? :-D

There's a shotgun already, but it's only much use at very close range.

Ha, that's cool! Burn 'em! What I recognized in your game, giving me some
second thoughts, is the similarity in how the shoots look (elliptical). I
also think there could be some more cool trails for the weapons. For example,
considering the flame weapon, what about letting the flame oscillate a bit
like a wave when moving forwards with the particles close-by / within the wake
start to move in a swirling pattern? I think this will also enhance the
perception of any atmosphere the player is in. And as such the effect my vary
with the set atmosphere (viscosity).

I think I get what you mean. I might experiment with this later.

Looks really good! Fire weapons are always fun. Are enemies gonna be able to set you alight as well? Not sure if that would work with your setup, but could be an interesting trade-off for a powerful weapon.

thank you!

like with most alt weapons, this is gained from a hunter who can use it against you. With that in mind, yeah, he can set you on fire, but it hadn't occurred to me to allow enemies to spread fire to you. I might make it work that way.




I tried making it, like, juicier: https://gfycat.com/NimbleBeneficialCockatiel

 

[missile]

 

[DemonNite]

its a disco ball!

 

[LordRaptor]

I would have some interest, but not this time around. This year will get
pretty though for me, for, I went all-in and now nature is going to calls me
on all frequencies, if you know what I mean.

I regret to say I don't, but I can only follow a fraction of most of your posts at the best of times, lol

I'm still half-convinced you're creating SkyNet or something

 

[TheKroge]

I regret to say I don't, but I can only follow a fraction of most of your posts at the best of times, lol

I'm still half-convinced you're creating SkyNet or something

Seconded. I have no idea what this means, but I'm excited nonetheless.

 

[Kilrathi]

Yeah, this was my first GGJ and I had a terrible time. Our group was comprised of strangers and everything that could go wrong went wrong (completely changed game ideas halfway through, team members left, morale was low etc). Early Sunday morning, I went back to the original idea and attempted to salvage it and we ended up with this:
http://globalgamejam.org/2017/games/newton-defence

That's pretty rough

Our ggj team decided to go all in with vr, using the rift, touch controllers, and a Virzoom bike. It's was an extremely bumpy ride, the game wasn't playable at all till like hour 40.

We pretty much built a paperboy like game where you throw papers and wave at your fellow citizens.

We want to keep working on it and smooth off the rough edges.

 

[missile]

its a disco ball!

Fuck yes! With some environment mapping and some other speculars it should
be just cool. I tried making a 360 spherical rendering into a 2:1 texture
from an arbitrary point in the scene, but for some reason it's rather buggy.

I regret to say I don't, but I can only follow a fraction of most of your posts at the best of times, lol

I'm still half-convinced you're creating SkyNet or something

At the start of august you will know. But has nothing to do with gamedev. It
effects just me but on all accounts. No problem with the post. Most of it isn't
very useful at all unless someone follows a similar track. But I hope my
pictures will tell on their own further down the road.

 

[missile]

... I've been MIA from the site and from this thread ...

Btw; What happend? And what is the state of MF1? Are you kickin sh!t on PS4?

 

[gashikunFLCL]

This was great, I really liked some of the puzzles (though I had to play it by myself which is hard as heck - I can totally see what you were going for with the two-player element). I doubt I could give any feedback you haven't thought of already (the menu selection seems weird, didn't always register dpad pressed - the jumping for player two on Y would kinda slam straight up instead of making the jump arc of player one, etc).

Thanks for playing it.

Just discovered that my teacher put a video of the game on the college youtube channel. It's really hard playing alone with the controller, I play tested the stages with the keyboard and played some of them with the controller, but we veteran players have to show how it's done *smirk*. Since you played alone I don't think you tried out the VS mode, it's a nice break from the main game.
I think I didn't put the InputManager of the dpad in some of the menus beacuse it was a late addition, so I just tested it while playing(one more thing to fix). About the jumping bug we found it while we were presenting the game, I played the first stage and then my partners played the rest while I explained and when it happened the one playing with player two went back to the keyboard, but just to confirm, when this happened you were pressing the shoulder button or the face button?

And again, thanks.

 

[shysaursoft]

but just to confirm, when this happened you were pressing the shoulder button or the face button?

It was the face button (Y), for the lower character on the first puzzle.

 

[Jacksinthe]

Btw; What happend? And what is the state of MF1? Are you kickin sh!t on PS4?

Still at it. Getting more done. Just not feeling much discussion lately. Swapped some meds and I've been off a bit since. Need to swap again but changes take some time. That's about it.

And Raptor already threw down the gauntlet for you so now you must fight to the death with others for king of the next thread. We will lose a few good devs but it's for a good cause

 

[MrHoot]

A new animation for when you summon divinities

AWAKEN MY MASTERS

 

[Makai]

Gitlab is so good. 10gb. Free private repos. Free continuous integration. Built-in ticket system.

 

[missile]

Still at it. Getting more done. Just not feeling much discussion lately. Swapped some meds and I've been off a bit since. Need to swap again but changes take some time. That's about it. ...

Good to hear. All the best, man!

... And Raptor already threw down the gauntlet for you so now you must fight to the death with others for king of the next thread. We will lose a few good devs but it's for a good cause

Down with Lord Raptor! We already lost a couple of good devs in fighting for
our indieland, but we won't give up. Baronet Pehesse nailed it on the wall;
*** no more High Res Work for Low Res Pay ***
*** Targeting 4K ($/y) ***
TO ARMS!

A new animation for when you summon divinities

AWAKEN MY MASTERS

Colors are nice. I actually like the little shiny touch he has. How many
frame will the animation finally have?

 

[Pixels]

Outsourcing art is incredibly expensive.

 

[Minamu]

If I ever manage to carve out some time, I want to dive seriously into UE4 and try to make something.

I've used UDK off and on over the years, and messed briefly with UE4. Are there any references, guides, intros, summaries, tutorials, etc. that are highly recommended? Or should I just dive in and look up stuff when I need it like usual?

What tools have you been using lately? Just curious I'm in the same position as you, and last time I tried to get into UE4, there was a massive playlist on youtube with like 400 official videos you could plough throw. It contained both a complete game(s?) you could code along with the videos and content regarding every single feature and button in the UI, all from Epic themselves with updated annotations as new engine versions came out. I'd link it to you but I'm at work. Should be real easy to find though.

 

[mStudios]

Outsourcing art is incredibly expensive.

Learn how to draw. It's cheaper.

 

[Blizzard]

What tools have you been using lately? Just curious I'm in the same position as you, and last time I tried to get into UE4, there was a massive playlist on youtube with like 400 official videos you could plough throw. It contained both a complete game(s?) you could code along with the videos and content regarding every single feature and button in the UI, all from Epic themselves with updated annotations as new engine versions came out. I'd link it to you but I'm at work. Should be real easy to find though.

I haven't been doing game development in quite some time. I've used UDK, messed with Unity, messed with Blender, messed with various Game Maker and similar tools, and created my own 2D engine in C++.

 

[missile]

... I ... messed with Unity, messed with Blender, messed with various Game Maker and similar tools ...

That's a mess! xD

 

[correojon]

I´ve been making slow progress in the last weeks, though nothing that can shine in a gif, just rewriting the collision engine and making some tweaks to the core of the level editor. I´m getting really burned with the level editor, it´s taking A LOT of time and I feel like if I ditched I could´ve already made 3 or 4 levels for the game. Instead I don´t have crap to play a year after starting the game The hope of the editor making level creation much faster in the future is what keeps me going on, the truth is that it´s more flexible and solid than it´s ever been now. I think that in a week or so the engine may be back on the point I was at September when I had a working basic demo and had to scrap almost all of the level editor core code.

I´m using GMS1 and am a bit proud of the collision system I´ve come up with. I still need to test it under stressful conditions with many enemies and moving platforms to make sure performance is stable, but it´s looking very promising so far. If anyone wants more details please ask

Speaking only for Pacha, to target HD resolution (1900x1080/1900x1200 - 4K is simply unrealistic with my means), my base canvas for drawing/animating is 3200x3200, which I reduce to 1000x1000 before exporting individual images (I like the compression artifacts that occur when reducing the image like this, those who won't will likely use an entirely different workflow/initial canvas size to avoid them). Cropping those images in C2 results in the frames being about 250x550 on average for the character standing motionless (the exact size varies based on the motions done, some require almost the full size of the initial canvas, like the warrior's spinning jump, so there's no single "true" number to rely on here). That's for having a rather large sprite on screen, I could probably get away with much smaller sprites if I had the character take up less screenspace!

On average, under these conditions and with an average of 12-14 frames per anim for simple motions (again, this can vary wildly), an animation requires about 10Mb of RAM according to C2's "estimated memory usage". I'm currently sitting at an overall estimated usage of 1.25Gb with only the main character's anims+some very small side stuff, I estimate the whole game to need about 2.5/3 when all is said and done (again, based on C2's metrics).

I'm not exactly sure what you were asking, so I hope these numbers will be of some help!

Those numbers are insane! I´m struggling here with simple 32x64px animations and you are easily making stuff 6 times bigger with more frames (and better). Kudos to you, sir.

Haha, agreed! And even if not, I still think it needs to be part of the next OP! Speaking of next thread, has an OP title been decided yet? If not, in keeping with the current thinking and in reference to the above, how about something like "Targeting 4K ($/y)"

Spoiler

or is that too self-deprecating? :v

Lol I love that OP title. May I suggest a smal addition?
GAF Indie Game Development Thread 3: Lazy Devs Targetting 4K ($/y)

This year will get
pretty though for me, for, I went all-in and now nature is going to calls me
on all frequencies, if you know what I mean.

I think you are either:

• Spoiler

  Having a child (big congratulations if so!)
• Spoiler

  Having to go through an ancestral ceremony where you´re left on your own in the middle of a forest armed only with a pasta strainer and must survive for a month to become a man. And the pasta strainer is cursed.

 

[FlashbladeGAF]

Cropping those images in C2 results in the frames being about 250x550 on average for the character standing motionless

I'm not exactly sure what you were asking, so I hope these numbers will be of some help!

Yeah I guess I could have worded it better.
I just wanted to know the pixel size of your (idle) sprite (for your main character).

Most 2D games are either "pixel" or HD puppeteer/flash style (Vanillaware), so seeing that you choose an HD traditional animated style had me curious to your method and the technical pros and cons (outside of drawing every frame) that picking that style brings.

I read that you were using Construct and (if I'm not mistaken) that you were close to maxing out the size limit with just the main character alone. Is that because the system assumes you are making a game for mobile? If Construct is limiting, why do you use it instead of another program or is it just the program you happen to be familiar with.

 

[HandsomeCharles]

Haha, agreed! And even if not, I still think it needs to be part of the next OP! Speaking of next thread, has an OP title been decided yet? If not, in keeping with the current thinking and in reference to the above, how about something like "Targeting 4K ($/y)"

Spoiler

or is that too self-deprecating? :v

That is a ridiculously good title

 

[missile]

...

I think you are either:

• Spoiler

  Having a child (big congratulations if so!)
• Spoiler

  Having to go through an ancestral ceremony where you´re left on your own in the middle of a forest armed only with a pasta strainer and must survive for a month to become a man. And the pasta strainer is cursed.

Thx man! Yes, I'm becoming a father!!! Six month to go... Holy!

 

[Pehesse]

Those numbers are insane! I´m struggling here with simple 32x64px animations and you are easily making stuff 6 times bigger with more frames (and better). Kudos to you, sir.

It's really two different things, though - a bigger canvas doesn't mean harder, in fact I'd argue the opposite :-D You have to make every pixel count, whereas I can afford lots of approximation!

Thx man! Yes, I'm becoming a father!!! Six month to go... Holy!

Oh! Congrats :-D :-D

Yeah I guess I could have worded it better.
I just wanted to know the pixel size of your (idle) sprite (for your main character).

Most 2D games are either "pixel" or HD puppeteer/flash style (Vanillaware), so seeing that you choose an HD traditional animated style had me curious to your method and the technical pros and cons (outside of drawing every frame) that picking that style brings.

I read that you were using Construct and (if I'm not mistaken) that you were close to maxing out the size limit with just the main character alone. Is that because the system assumes you are making a game for mobile? If Construct is limiting, why do you use it instead of another program or is it just the program you happen to be familiar with.

Ah, I see!

Well, in my case it's really an aesthetical choice: as far as I know, there are few to no pros on the technical side to use that animation technique (and most others are designed expressedly to cover for its faults) - it's longer, it's harder, it costs a lot more... I just find the aesthetical gain to completely offset those cons and make it ultimately "worth it", but it's obvious I couldn't do this in any kind of traditional industrial setting! Very few "standard industrial" mid-size studios can afford it (in fact, I can only really think of Lab Zero these days), and I only can because I work solo, on my own schedule every day with little in the way of short-term external deadlines, and where the main pressure comes from my source of funding (french welfare system, with oversight by the people in charge of my situation) - meaning if they decide to cut me off at any point, I have nothing left, as I have barely enough to make it month to month as it is. (To be clear, I'm not complaining, far from it: just trying to answer your question as detailed as I can to underline that it's not a choice to be made lightly, and that I made it because I believe it's one of the few things I *can* do and bring to the table at this point).

Regarding C2, I've talked about it a few times and I may have been unclear: there is no actual "max memory size limit" that I know of, but the engine had a memory warning when going over an arbitrary size (1Gb of estimated RAM usage), which is what I have repeatedly complained about. After enough complaining, though, C2's devs have very recently removed this warning and the associated hassle (many thanks to them!!) meaning I'm free to stuff as many assets as I want in C2... at least, so far! Maybe I'll run up against an actual wall later down the line :-D

I'm using C2 because it's the only engine I know of that is explicitely designed *not* to rely on written code by the user. Even Game Maker expects users to know and use GML after a while, and I simply cannot learn/use proper programming syntax (spent enough years hitting my head against that specific wall to know that's a major issue for me). If I'm to work solo, C2 is the only engine I *can* use, as everything about it is accessible through visual building blocks (Unity and Unreal have some "visual scripting" plugins I know of, but in my experience, they're still very limiting and don't allow for use of the whole engine's capabilities, like C2 does).

So it's basically a "make do with what I have" situation, where I'm lucky enough that C2 can do what I want well enough! :-D
But it's true that it's primarily designed with mobile games in mind, so one has to watch out for issues specific to bigger projects that might pop up: if you intend to use C2 for your own ends, look into the trial version/other games (most notably: Yonder, Klang) that have been done with it first to see if it suits your needs!

EDIT: shoot, top of the page, I was hoping someone else would've posted something while I was typing that novel. Well uh, have one of the latest animations:

(one of the last for a while, actually, as I'll begin cleaning/coloring all of the sketches so far in the coming days and for a likely big part of the year, so updates might get a bit scarce).

Also, I hadn't posted the finished version of the illustration I had posted prior, so here it is as well for those curious/interested:

 

[desu]

Thx man! Yes, I'm becoming a father!!! Six month to go... Holy!

Congrats!

 

[timetokill]

Congrats missile!!

 

[missile]

 

[Dascu]

Awesome news missile.

 

[MrHoot]

What engine are you using for your kid ?