• Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
  • We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.
Status
Not open for further replies.

OmegaJabroni

Member
Jan 23, 2018
18
This is the first time I've posted on the thread here on ERA about our falling block competitive puzzle game Treasure Stack! The game's being developed in Unity, after getting it's start in Construct 2 in March of last year! I'm gonna drop a few gameplay gifs here to begin with. I'd really like to get some feedback and field any and all questions people have! Thanks for the thread Popstar

Construct 2 Early Gameplay

giphy.gif


Two-Player Versus (Unity)

giphy.gif


Four-Player Versus (Unity)

giphy.gif


If anyone wants to stay current with development and/or announcements etc. Follow us on Twitter @TreasureStack thanks!
 
Last edited:

K Monkey

Member
Oct 25, 2017
278
This is the first time I've posted on the thread here on ERA about our falling block competitive puzzle game Treasure Stack! The game's being developed in Unity, after getting it's start in Construct 2 in March of last year! I'm gonna drop a few gameplay gifs here to begin with. I'd really like to get some feedback and field any and all questions people have! Thanks for the thread Popstar

Construct 2 Early Gameplay

giphy.gif


Two-Player Versus (Unity)

giphy.gif


Four-Player Versus (Unity)

giphy.gif


If anyone wants to stay current with development and/or announcements etc. Follow us on Twitter @TreasureStack thanks!

hey that looks really neat :)
 

HP_Wuvcraft

Member
Oct 26, 2017
4,267
South of San Francisco
This is the first time I've posted on the thread here on ERA about our falling block competitive puzzle game Treasure Stack! The game's being developed in Unity, after getting it's start in Construct 2 in March of last year! I'm gonna drop a few gameplay gifs here to begin with. I'd really like to get some feedback and field any and all questions people have! Thanks for the thread Popstar

Construct 2 Early Gameplay

giphy.gif


Two-Player Versus (Unity)

giphy.gif


Four-Player Versus (Unity)

giphy.gif


If anyone wants to stay current with development and/or announcements etc. Follow us on Twitter @TreasureStack thanks!
That looks effing fantastic.
 
Oct 27, 2017
262
In theory the actual rooms as well, but my experiments into that so far haven't gone so well. I'll need to make another start on that once I've got the dungeon layouts to a place I'm happier with.

I imagine it'll end up as some hybrid where chunks of rooms are reused depending on context. Probably easier to do that with a 2D (or 2.5D) game than survival horror, I guess. I am intrigued as to your approach!
 

xir

Member
Oct 27, 2017
12,577
Los Angeles, CA
This is the first time I've posted on the thread here on ERA about our falling block competitive puzzle game Treasure Stack! The game's being developed in Unity, after getting it's start in Construct 2 in March of last year! I'm gonna drop a few gameplay gifs here to begin with. I'd really like to get some feedback and field any and all questions people have! Thanks for the thread Popstar

Construct 2 Early Gameplay

giphy.gif


Two-Player Versus (Unity)

giphy.gif


Four-Player Versus (Unity)

giphy.gif


If anyone wants to stay current with development and/or announcements etc. Follow us on Twitter @TreasureStack thanks!

oh man, rad. I always loved the promise of the late NES puzzle games and Tetris plus on ps1 with the little guy with the safari hat, but this looks way better. love the idea
 
Oct 25, 2017
6,227
Mementos
My feet hurt from running around for these past few days trying to find people at various colleges to work with on designing a video game. Had to talk to department chairs, get fliers approved, talk to students and I don't have a car so i have to ride the bus and today the bus passes right near my house but wouldn't let me off so I had to go into the next city wait at my mom's job for like 3 hours and then have her drive me home when I could've been home in like five minutes then. I think I found a good few people who are interested. It's just a matter of keeping them around.
 

HP_Wuvcraft

Member
Oct 26, 2017
4,267
South of San Francisco
In theory the actual rooms as well, but my experiments into that so far haven't gone so well. I'll need to make another start on that once I've got the dungeon layouts to a place I'm happier with.

I imagine it'll end up as some hybrid where chunks of rooms are reused depending on context. Probably easier to do that with a 2D (or 2.5D) game than survival horror, I guess. I am intrigued as to your approach!
I'm very interested in this, btw.
 

OmegaJabroni

Member
Jan 23, 2018
18
That looks effing fantastic.

Thanks so much for the kind words! We're working hard to make it fantastic!

oh man, rad. I always loved the promise of the late NES puzzle games and Tetris plus on ps1 with the little guy with the safari hat, but this looks way better. love the idea

Thank you! We looked at Wario Woods as a jumping off point. From there we decided to add free-flowing movement and lots of chaos!
 
Oct 25, 2017
6,227
Mementos
Ugh, I think my mother forgot to pay the internet bill. Every time I attempt to go to the SFML website and download their program our internet provider's website comes up talking about payment notification. I don't get it. I can go anywhere else on the internet, but there. I'm supposed to meet people on Discord tonight to start working on a game. It will not be a good look if I don't have it and I'm the one who organized the whole meeting.
 

OniLinkPlus

Self-Requested Ban
Banned
Oct 25, 2017
600
So someone told me Linux is better for making games on. Is that true? I have no experience with it and I'm used to Windows and a lil bit of Mac.
Not really. While for general programming I would urge anybody to work on Linux because Linux has very well crafted development chains whereas Windows... is a nightmare, for game programming I would recommend Windows. This is mostly because Windows has excellent tools like Unity and UE4 which cover the bulk of the game development pipeline. Anything they don't cover is covered by apps that are available on basically all of the main OSes people use.

EDIT: I see in a previous post you're using SFML. In that case, you're not using those pipelines that are well crafted for Windows, in which case yes I would recommend working on Linux. The main issue you'll encounter is cross-compiling, but that is only slightly more difficult to set up than the base development chain on Linux. Just gotta make use of some basic google-fu for whichever distro you use. Worth considering - I'm used to both Linux and Windows, so I'm not considering the time it takes to get comfortable with Linux.
 

Koren

Member
Oct 25, 2017
619
The main issue you'll encounter is cross-compiling, but that is only slightly more difficult to set up than the base development chain on Linux.
So you would recommand going with cross compilation? I develop for both Linux and Windows, and I usually compile on both, which is a hassle, because I never took the time to setup a cross compiler. I should look into it...

Also, isn't Unity virtually fully available on Linux too? I remember reading about 2017.3 coming with a full installer, matching the Windows behavior (haven't tested it yet, probably this summer...)

UE4 is doable too, but afaik with a complete compilation from sources :/ Not the same.

I'm still using SDL2 for games :)
 

Minamu

Member
Nov 18, 2017
1,900
Sweden
In theory the actual rooms as well, but my experiments into that so far haven't gone so well. I'll need to make another start on that once I've got the dungeon layouts to a place I'm happier with.

I imagine it'll end up as some hybrid where chunks of rooms are reused depending on context. Probably easier to do that with a 2D (or 2.5D) game than survival horror, I guess. I am intrigued as to your approach!
Yeah, we're making individual rooms that we connect at the door seams with the help of raycasts to determine what kind of room can fit so they don't clip into each other. I make the general room shape with pro builder in Unity, while others make the assets that I then can place within the rooms. At the moment, the layout is completely randomized so the house could potentially, and very likely, be very weird but that's the nature of random. I've suggested making that issue a part of the narrative, might be easier to swallow then. I'm not super well informed in exactly how the random code works for us, but the idea is to make a sort of AI director like in Left 4 Dead that can change things around.

K monkey: I see you have a mini map, how did you make it? :) I suspect we will need one too.
 

K Monkey

Member
Oct 25, 2017
278
Yeah, we're making individual rooms that we connect at the door seams with the help of raycasts to determine what kind of room can fit so they don't clip into each other. I make the general room shape with pro builder in Unity, while others make the assets that I then can place within the rooms. At the moment, the layout is completely randomized so the house could potentially, and very likely, be very weird but that's the nature of random. I've suggested making that issue a part of the narrative, might be easier to swallow then. I'm not super well informed in exactly how the random code works for us, but the idea is to make a sort of AI director like in Left 4 Dead that can change things around.

K monkey: I see you have a mini map, how did you make it? :) I suspect we will need one too.

This is probably not the most elegant but it works for me :)

pO2T70p.png
cieFd2i.png


Every prefab piece (doors, walls, floors, stairs etc...) has the "legend" graphic which can only be seen by the "map" camera. This is drawn to a rendertexture (the map) and the fog of war code hides it.

Changing floors in the level will just switch between each map texture when brought up. The walking path is just a trail renderer.
 

OmegaJabroni

Member
Jan 23, 2018
18
This is probably not the most elegant but it works for me :)

pO2T70p.png
cieFd2i.png


Every prefab piece (doors, walls, floors, stairs etc...) has the "legend" graphic which can only be seen by the "map" camera. This is drawn to a rendertexture (the map) and the fog of war code hides it.

Changing floors in the level will just switch between each map texture when brought up. The walking path is just a trail renderer.

This is all randomly generated? Looks pretty cool! Nice work.
 
Oct 27, 2017
262
Yeah, we're making individual rooms that we connect at the door seams with the help of raycasts to determine what kind of room can fit so they don't clip into each other. I make the general room shape with pro builder in Unity, while others make the assets that I then can place within the rooms. At the moment, the layout is completely randomized so the house could potentially, and very likely, be very weird but that's the nature of random. I've suggested making that issue a part of the narrative, might be easier to swallow then. I'm not super well informed in exactly how the random code works for us, but the idea is to make a sort of AI director like in Left 4 Dead that can change things around.

Ah, cool. So all the rooms string together, or is there a hallway they're built around? I think it wouldn't be so hard to implement rules about which rooms can go near which other rooms, but that depends on your approach of course.

Trying to make the layout actually feel natural, like things are laid out in a coherent way a human might have done, is pretty tough. I've been working a lot on a) what the goals are for my dungeons in terms of a satisfying layout and b) how to represent that in a way that a program can generate and test for.

I don't know much about AI, but in the past I've experimented with a genetic algorithm approach where a large number of dungeons is generated at random, then the best ones "mate" until over time the best possible dungeon evolves. Unfortunately Zelda-inspired dungeons are quite specific in their layout, so a method like that which begins with random generation didn't work out so well... the best dungeons it found were always kind of shitty. I might go back to that for other elements of the game though.

K Monkey: Very impressive procedural dungeons! I'd love to know more about how you did it. (Can you tell this stuff fascinates me? Haha)
 

OniLinkPlus

Self-Requested Ban
Banned
Oct 25, 2017
600
So you would recommand going with cross compilation? I develop for both Linux and Windows, and I usually compile on both, which is a hassle, because I never took the time to setup a cross compiler. I should look into it...

Also, isn't Unity virtually fully available on Linux too? I remember reading about 2017.3 coming with a full installer, matching the Windows behavior (haven't tested it yet, probably this summer...)

UE4 is doable too, but afaik with a complete compilation from sources :/ Not the same.

I'm still using SDL2 for games :)
I mean, I personally would recommend cross compiling, but again I'm used to Linux as I've been using it for about 10 years now and probably a little biased! The whole development chain is just... so much easier to handle on Linux thanks to the package systems that Linux distros use for updates and installation. MinGW is even available!

I do remember there being experimental work on porting Unity's editor to Linux but I switched over to Godot before I saw what became of it. Very good to hear that it's apparently a thing now!
 
Last edited:

Koren

Member
Oct 25, 2017
619
I don't even have Windows on my home computers anymore (just on laptops), and I fully share your feelings about the tools.

It's just that I only crosspiled for ARM and the like, and it usually is a bit akward to setup.

I should try to compile with MingW on Linux (though I can hear cpp_is_king shouting it's Visual, clang or bust ;) )
 
Oct 25, 2017
3,686
Yeah, I've coded a lot in my life (professionally as a career and also hobbyist stuff) and my goal is always to do things in a way that are clean and don't have the "bandaids," but sometimes you realize it's impossible to do something completely clean and elegant and those "bandaids" are needed. Also, very generally speaking, there's now *WRONG* way to do something. It's just possible there's a better way lol.
You had me up until the bolded part. I've been programming over a decade and I'm almost CERTAIN there are wrong ways to do things (including the ways I've used for personal projects). :P
 

_Rob_

Member
Oct 26, 2017
606
I've found as of late, that players really respond to "life" just happening around them. It seems to work even more so if these semi-random events pose some sort of hazard. As such I've decided the natural next step is to have two little imps riding around on a runaway carriage... of course!

(excuse the WIP visuals!)
giphy.gif


Edit: Here it is in-game, thankfully it didn't take nearly as long as I thought it might!

GreatWenCarriage.png
 
Last edited:

jahasaja

Banned
Oct 26, 2017
793
Sweden
I've found as of late, that players really respond to "life" just happening around them. It seems to work even more so if these semi-random events pose some sort of hazard. As such I've decided the natural next step is to have two little imps riding around on a runaway carriage... of course!

(excuse the WIP visuals!)
giphy.gif


Edit: Here it is in-game, thankfully it didn't take nearly as long as I thought it might!

GreatWenCarriage.png
Looks great. And I think you have a really good point that having things in the background that the player can interact with makes the game more alive.
 

Minamu

Member
Nov 18, 2017
1,900
Sweden
I've finally managed to take some screenshots of my projects xD What's a good image host site I can use? imgur is blocked at work.
 

Minamu

Member
Nov 18, 2017
1,900
Sweden
Thanks, yeah with some google translate, I got in :)

Here are some of the maps (3 of 5) from Holy Sheep (still WT). In case some of you didn't follow this development on the old site, Holy Sheep is an online multiplayer cat and mouse game for up to 4 people. Each player class have defensive and offensive abilities that affect the other players negatively. The goal is to collect the Holy Sword which when used to strike a player turns them into tiny sheep. Last man standing wins. We used to be two programmers and one designer, but one of the programmer got a baby recently so we're basically just two guys these days. This ongoing GGJ project actually turned two years old yesterday:
Sacrifice has improved so much, and it's fun to shock my teammates with the new looks whenever they see it xD The first version was all green and very tiny, but the idea of two floors to play on has remained since the start. This close to final version is basically a space station above a strange planet, with towering skyscrapers surround the center piece. This is how the map used to look way back:
https://i.imgur.com/W0VE4QM.png

I hope this picture isn't too dark, it looked brighter when I captured it yesterday. Pyramid still looks fairly similar to its original concept, but like with pretty much all levels, it's been opened up to not have huge 30 meter high walls all around the play area. There's a small ravine around the level and obviously an Egyptian theme all around it. A volcano of sorts randomly shoots up glowing rocks that roll along the ground a couple of times per minute.

Our Battleground map has changed quite dramatically over the years. Like the picture below shows, it used to be super blue but it made more sense to connect the level and its name to our Ghost class so now it's entirely black and white in tone. The rain clouds above the level come and go, which is a happy accident due to bad particle system settings, so it stays :D The rain could probably use some work, but it and the very new lightning strike at least have sound effects now, something no other map has. I'm very happy with the way the lightning turned out, randomly striking the level every 3-6 seconds with sparks flying wherever it hits. This is a very early layout of the level:
https://i.imgur.com/ZIIy2Qm.png

***

Some pictures of my Skyrim mod environment. I'll link them because the file size became higher than I anticipated:
http://abload.de/img/wiptavernskyrimsaso1.png
The idea is to make the tavern keeper much more interactive of course and involved in a puzzle riddle, and add some better mood lighting.

http://abload.de/img/wippracticeskyrimf4sf4.png
This is a tavern where people can practice their bow and arrow skills. I've placed some pickable arrows randomly in an attempt to make the environment less sterile and more believable.

http://abload.de/img/wiplibraryskyrim8ssgs.png
A library with another puzzle that will eventually lead the player to a completely different dungeon with a different kind of gameplay. I still need to code a functioning NPC for this part. The bookshelves all have some random books in alphabetical order and you will need to find a certain object somewhere to figure out what and where something is missing.

***

Here are some pictures of our currently untitled randomized horror game. These environments are obviously very early work in progress:
This area is our first vertical slice demo in unlit mode. The level was put together manually by our programmer for a showcase.

The second room in the bottom right corner of the first picture. Mood lighting test.

We don't have a lot of assets so far so I had to make due with what was available hehe. I tried making a small ritual shrine with the teddy bears close to a dining table.

Another teddy bear scenario. This room in particular is just an S shaped corridor, only lit with tiny candles here and there. I've placed some bookshelves to create a sort of safe space to read a good book in. Oute items for pickup can spawn between the candles in the middle, and hopefully something scary could happen down the road with the teddy, who knows :)

I've made a 100+ different rooms over the last few months, I just need to get the assets to make them feel lived in, and get more variety in the tapestry etc, hehe.
 

M-66

Member
Oct 27, 2017
68
sh_5cf0ce87.png

Doing some 4-bit (15 colors plus BG/transparent) palette spritework for fun. Here's some experiments with different player sprite sizes for a roughly 5-6 head character.

I think it's an interesting time for pixel art, since to my eyes, 240p canvas resolution doesn't really cut it these days. Targeting a high-quality pixel aesthetic (think late SNK era) for modern displays requires more detailed sprites than it did back then for the same amount of screen coverage.

As an example, the Metal Slug main player sprites were roughly 36ish pixels tall and took up about 16% of the height of the screen at 224p. A commonly accepted "modern" pixel canvas size would be 640x360 (2x pixels at 720p and 3x at 1080p), So to get the same height coverage you'd want sprites around 58-60 pixels tall.

There are other considerations as well when choosing screen coverage size. In OwlBoy for example, I notice that they still hew closer to the old-style dimensions for the player sprite (also about 36-37p), while using a modern canvas size. To me the sprites feel just a teeny bit too small, though the game has a long-range/shooter aspect, so the smaller sprite size is most likely a benefit, giving the player better situational awareness (can see threats further away)?

If the above sprite was for a keyboard + mouse aim game you'd expect to engage targets from longer distance so probably the smaller sprites would be better. It's hard to go back to something like the 48 pixel version after the lovely detail possible in the 72 pixel version haha. The 60 pixel version would probably be the final compromise.

The true test would come with animation. More detail/pixels means lots of additional work since the artist is now responsible for considerably more pixels per frame. One would probably need to adopt a much more "clustery" approach to coloring than I did here. Indeed even the MS sprites have considerably less AA despite also having 15+1 colors. While the individual frames for the player sprites aren't often that detailed, Nazca made up for it with exquisite animation and a wonderful freedom in terms of warping/rotating the character through implied 3d space for each frame (very little of just copying the same head for the character for all 12 frames of the run cycle for example).
 
Last edited:
Oct 27, 2017
262
Minamu Looks like you're working on a bunch of interesting things! Love the visual style of Holy Sheep, although it's not so much my kind of game. The survival horror environments look suitably creepy (love the creative use of teddy bears, haha) and at a glance the house layout seems pretty authentic... as in, not so much less eccentric than the classics :P Would love to play this one day.
 

mikehaggar

Developer at Pixel Arc Studios
Verified
Oct 26, 2017
1,379
Harrisburg, Pa
You had me up until the bolded part. I've been programming over a decade and I'm almost CERTAIN there are wrong ways to do things (including the ways I've used for personal projects). :P

Well, like I said, very generally speaking... lol

I've found as of late, that players really respond to "life" just happening around them. It seems to work even more so if these semi-random events pose some sort of hazard. As such I've decided the natural next step is to have two little imps riding around on a runaway carriage... of course!

(excuse the WIP visuals!)
giphy.gif


Edit: Here it is in-game, thankfully it didn't take nearly as long as I thought it might!

GreatWenCarriage.png

This looks fantastic! And yes, making the environments feel alive and lived in goes a long way!

sh_5cf0ce87.png


Doing some 4-bit (15 colors plus BG/transparent) palette spritework for fun. Here's some experiments with different player sprite sizes for a roughly 5-6 head character.

I think it's an interesting time for pixel art, since to my eyes, 240p canvas resolution doesn't really cut it these days. Targeting a high-quality pixel aesthetic (think late SNK era) for modern displays requires more detailed sprites than it did back then for the same amount of screen coverage.

As an example, the Metal Slug main player sprites were roughly 36ish pixels tall and took up about 16% of the height of the screen at 224p. A commonly accepted "modern" pixel canvas size would be 640x360 (2x pixels at 720p and 3x at 1080p), So to get the same height coverage you'd want sprites around 58-60 pixels tall.

There are other considerations as well when choosing screen coverage size. In OwlBoy for example, I notice that they still hew closer to the old-style dimensions for the player sprite (also about 36-37p), while using a modern canvas size. To me the sprites feel just a teeny bit too small, though the game has a long-range/shooter aspect, so the smaller sprite size is most likely a benefit, giving the player better situational awareness (can see threats further away)?

If the above sprite was for a keyboard + mouse aim game you'd expect to engage targets from longer distance so probably the smaller sprites would be better. It's hard to go back to something like the 48 pixel version after the lovely detail possible in the 72 pixel version haha. The 60 pixel version would probably be the final compromise.

The true test would come with animation. More detail/pixels means lots of additional work since the artist is now responsible for considerably more pixels per frame. One would probably need to adopt a much more "clustery" approach to coloring than I did here. Indeed even the MS sprites have considerably less AA despite also having 15+1 colors. While the individual frames for the player sprites aren't often that detailed, Nazca made up for it with exquisite animation and a wonderful freedom in terms of warping/rotating the character through implied 3d space for each frame (very little of just copying the same head for the character for all 12 frames of the run cycle for example).

These look great! I think the 60 pixel height sprite would work perfectly fine for a modern game, although if you were using a 640x360 native canvas you would have a large viewport--meaning a lot of the environment around the player would be visible. We ended up going with a 480x270 pixel canvas for Bushiden as we felt too much of the surrounding environment was visible/exposed using 640x360 with our sprite sizes and tile sizes. The only drawback there to me is that we can't scale evenly to 720p. Anyway, it all becomes a balancing act and you are smart to be thinking about this now and doing tests.
 

_Rob_

Member
Oct 26, 2017
606
sh_5cf0ce87.png


Doing some 4-bit (15 colors plus BG/transparent) palette spritework for fun. Here's some experiments with different player sprite sizes for a roughly 5-6 head character.

I think it's an interesting time for pixel art, since to my eyes, 240p canvas resolution doesn't really cut it these days. Targeting a high-quality pixel aesthetic (think late SNK era) for modern displays requires more detailed sprites than it did back then for the same amount of screen coverage.

As an example, the Metal Slug main player sprites were roughly 36ish pixels tall and took up about 16% of the height of the screen at 224p. A commonly accepted "modern" pixel canvas size would be 640x360 (2x pixels at 720p and 3x at 1080p), So to get the same height coverage you'd want sprites around 58-60 pixels tall.

There are other considerations as well when choosing screen coverage size. In OwlBoy for example, I notice that they still hew closer to the old-style dimensions for the player sprite (also about 36-37p), while using a modern canvas size. To me the sprites feel just a teeny bit too small, though the game has a long-range/shooter aspect, so the smaller sprite size is most likely a benefit, giving the player better situational awareness (can see threats further away)?

If the above sprite was for a keyboard + mouse aim game you'd expect to engage targets from longer distance so probably the smaller sprites would be better. It's hard to go back to something like the 48 pixel version after the lovely detail possible in the 72 pixel version haha. The 60 pixel version would probably be the final compromise.

The true test would come with animation. More detail/pixels means lots of additional work since the artist is now responsible for considerably more pixels per frame. One would probably need to adopt a much more "clustery" approach to coloring than I did here. Indeed even the MS sprites have considerably less AA despite also having 15+1 colors. While the individual frames for the player sprites aren't often that detailed, Nazca made up for it with exquisite animation and a wonderful freedom in terms of warping/rotating the character through implied 3d space for each frame (very little of just copying the same head for the character for all 12 frames of the run cycle for example).

I agree with mikehagger, the 60 feels like a fair compromise between size and detail.

This looks fantastic! And yes, making the environments feel alive and lived in goes a long way!

Thanks, and it really does; suddenly just being near the road feels treacherous when you hear the hooves in the distance, in fact:




giphy.gif

Ignore the inverted stereo... it turns out I've had my speakers flipped over for months, what a Muppet!
 
Last edited:

jahasaja

Banned
Oct 26, 2017
793
Sweden
sh_5cf0ce87.png


Doing some 4-bit (15 colors plus BG/transparent) palette spritework for fun. Here's some experiments with different player sprite sizes for a roughly 5-6 head character.

I think it's an interesting time for pixel art, since to my eyes, 240p canvas resolution doesn't really cut it these days. Targeting a high-quality pixel aesthetic (think late SNK era) for modern displays requires more detailed sprites than it did back then for the same amount of screen coverage.

As an example, the Metal Slug main player sprites were roughly 36ish pixels tall and took up about 16% of the height of the screen at 224p. A commonly accepted "modern" pixel canvas size would be 640x360 (2x pixels at 720p and 3x at 1080p), So to get the same height coverage you'd want sprites around 58-60 pixels tall.

There are other considerations as well when choosing screen coverage size. In OwlBoy for example, I notice that they still hew closer to the old-style dimensions for the player sprite (also about 36-37p), while using a modern canvas size. To me the sprites feel just a teeny bit too small, though the game has a long-range/shooter aspect, so the smaller sprite size is most likely a benefit, giving the player better situational awareness (can see threats further away)?

If the above sprite was for a keyboard + mouse aim game you'd expect to engage targets from longer distance so probably the smaller sprites would be better. It's hard to go back to something like the 48 pixel version after the lovely detail possible in the 72 pixel version haha. The 60 pixel version would probably be the final compromise.

The true test would come with animation. More detail/pixels means lots of additional work since the artist is now responsible for considerably more pixels per frame. One would probably need to adopt a much more "clustery" approach to coloring than I did here. Indeed even the MS sprites have considerably less AA despite also having 15+1 colors. While the individual frames for the player sprites aren't often that detailed, Nazca made up for it with exquisite animation and a wonderful freedom in terms of warping/rotating the character through implied 3d space for each frame (very little of just copying the same head for the character for all 12 frames of the run cycle for example).

Great information! I am doing a pixel art game with characters between 50 to 100 pixel height but with less detail than your sprites. However, since I try to make quite smooth animation (every character have 70 plus frames) more than 75 percent of the development time is spent on the animations.
 

M-66

Member
Oct 27, 2017
68
We ended up going with a 480x270 pixel canvas for Bushiden as we felt too much of the surrounding environment was visible/exposed using 640x360 with our sprite sizes and tile sizes. The only drawback there to me is that we can't scale evenly to 720p.

Checked out your game; looks sweet. What do you do for 720p? 3x and crop? I still hold a candle for 720p because my home machine is a potato. I imagine supporting 720p will soon be as quaint as supporting mode 13h and I should maybe get with the times haha.
 

2+2=5

Member
Oct 29, 2017
971
Hi guys, can anyone point me to some guide on how to design levels visually? There are lots of level design tutorials but they mostly focus on the "functional" side of it, i would like to know if there are tips, workflows to follow, things to avoid etc while visually designing an environment.

If i can i would like to suggest some resources:
-Krita: GREAT professional-level free paint program, it's awesome for pixel art too
-InkScape: free vectorial graphics program, i suggest it because for people who isn't really good at drawing vector graphics could help doing the basic design of a character for example, exporting like pixel art(even at different resolutions) and continue editing in a pixel art(or paint) program
-normal map online: it creates instantly normal maps(and other maps)
-LMMS: a free DAW(music program ala music maker), download soundfonts, sounds and VST you like and you can do basically anything
-sunvox: a free modern take on music trakers
 

mikehaggar

Developer at Pixel Arc Studios
Verified
Oct 26, 2017
1,379
Harrisburg, Pa
Checked out your game; looks sweet. What do you do for 720p? 3x and crop? I still hold a candle for 720p because my home machine is a potato. I imagine supporting 720p will soon be as quaint as supporting mode 13h and I should maybe get with the times haha.

Thanks! We just aren't currently supporting 720p. The only thing that bothers about that is a potential Switch port, but there's plenty of development time left to find a solution. The PC version offers a variety of resolutions (all even scales of the native res) and then we offer a "full screen" option which will scale the game to whatever resolution your monitor is. Only a monster would use that option if their monitor resolution isn't an even scale of the game resolution though lol. 960x540 on a 720p screen is acceptable in my opinion. And who knows, maybe we end up using a 640x360 native res. It would only take me 20 minutes or so to implement that.
 

Minamu

Member
Nov 18, 2017
1,900
Sweden
Hi guys, can anyone point me to some guide on how to design levels visually? There are lots of level design tutorials but they mostly focus on the "functional" side of it, i would like to know if there are tips, workflows to follow, things to avoid etc while visually designing an environment.

If i can i would like to suggest some resources:
-Krita: GREAT professional-level free paint program, it's awesome for pixel art too
-InkScape: free vectorial graphics program, i suggest it because for people who isn't really good at drawing vector graphics could help doing the basic design of a character for example, exporting like pixel art(even at different resolutions) and continue editing in a pixel art(or paint) program
-normal map online: it creates instantly normal maps(and other maps)
-LMMS: a free DAW(music program ala music maker), download soundfonts, sounds and VST you like and you can do basically anything
-sunvox: a free modern take on music trakers
Can't help but would like to see this as well. What kind of functional stuff have you found?
 
Status
Not open for further replies.