Open Group  ·  86 members

• entries
14
212
• views
24954

# August GameDev Challenge Poll

1065 views

## GameDev Challenge Poll   27 members have voted

You do not have permission to vote in this poll, or see the poll results. Please sign in or register to vote in this poll.

It's time to vote for the next challenge. I've listed 10 options in this poll, but I'm going to accept new suggestions in the comments for the next few days.

Wide variety in this one - ranging from more complex game systems to the more simple arcade classics. Which will you choose?

As in past polls, you can vote for as many as you want. Poll closes August 7 with the intent to start the next challenge August 8.

Loot Box means we ignore all the votes and the theme is randomly selected from the list.

Posted (edited)

I find some options to be too similar. Donkey Kong/2D platformer/Contra;  Space Invaders/Centipede. Especially since last challenge there were extra constraints placed on the winning option after voting.

For example, if Donkey Kong were to win, would entries be required to not have side-scrolling? Is that the difference with 2D platformer? How is Contra different to 2D platformer? Cause helicopters instead of nimbus clouds?

Same with Space Invaders/Centipede. If Space Invaders won, you couldn't have Centipede type gameplay? What about Centipede wins and you don't incorporate the dead centipede bodies turning into mushrooms (the thing that makes Centipede Centipede). It's very difficult to make Centipede without making a clone or something very close.

Some options that received very few votes last time could have been removed.

Edited by fleabay

I'm travelling and will kick this one off soon when I'm back home.

Looks like it's a Doom challenge, but what will be the twist?

Did I hear DOOM challenge? Sounds like I will just make time to participate, because that is a must do.

Also @khawk I can't view poll results (is that due to poll being locked)?

Poll results: @Vilem Otte

Thanks, I guess it's time to make something!

1 hour ago, Vilem Otte said:

Thanks, I guess it's time to make something!

Cannot wait ! Gonna model some interesting weapons.  Best of luck with your entry!

15 hours ago, khawk said:

I'm travelling and will kick this one off soon when I'm back home.

Looks like it's a Doom challenge, but what will be the twist?

The most lethal weapons are all timer-based, so you need to set them up beforehand, trigger them, and lure the enemy to them so it explodes when they are near.

19 hours ago, khawk said:

Looks like it's a Doom challenge, but what will be the twist?

Some twist ideas to make it faithful to the original:

• only the CPU can be used, no GPU.
• only keyboard input
• idTech/Carmack was revolutionizing tech, maybe limit the resolution to the original 320x200, but require to use some kind of ray tracing
• maps need to be created with the original editor
• like the original: all enemy need to be scanned from real models, allowing any kind of real life art e.g. photograph of hand painted stuff.
• mode 13h

46 minutes ago, ProfL said:
• only the CPU can be used, no GPU.
• only keyboard input
• idTech/Carmack was revolutionizing tech, maybe limit the resolution to the original 320x200, but require to use some kind of ray tracing
• maps need to be created with the original editor
• like the original: all enemy need to be scanned from real models, allowing any kind of real life art e.g. photograph of hand painted stuff.
• ﻿﻿ mode 13h

All of these are kind of problematic to me...

Using only keyboard, mode 13h or similar - are basically limiting your hardware, and also software (+ languages you can use) - these kinds of limitations are always bad (especially when embedded devices are allowed - which if I'm not mistaken is true).

Limiting resolution, or the way (all) assets are created is somewhat bad - that is basically deciding art direction in competition mainly for people who do programming. Which is going to end in lower number of participants.

I could imagine requiring a software renderer, or requiring a ray tracing for at least some effect. It would, sadly, remove any beginners or people who don't have experience with tons of optimization and rendering.

Most of these would also mean that you need to write completely whole engine, that means either:

• Making competition time too long (6+ months), as most people don't have time to work on this full time due to their real life jobs, long competitions tend to have very few participants though

• Nobody is going to participate - to be fair - I can probably reserve 2, maybe 3 weekends over 6 - 8 weeks to work on this. It is not possible to write whole engine with high performance software renderer (or using specific hardware software - which I might not have used before) and whole game on top of it with art in that time range

EDIT: To not spam... a funny one would be:

- as Doomguy is generally tough guy - participant has to use his own voice to add audio of toughest quotes played when dismembering demons.

Was there a twist on the Frogger challenge? I don't remember one. Why does there have to be a twist?

The twist of no weapons on Side Scrolling Arcade Shooter Challenge was OK because it was a genre, not a game. Doom is a game, not a genre. If you wanted to put a twist on an FPS, you should have called the choice "Old School FPS" instead of Doom.

Posted (edited)

1 hour ago, Vilem Otte said:

All of these are kind of problematic to me...

Using only keyboard, mode 13h or similar - are basically limiting your hardware, and also software (+ languages you can use) - these kinds of limitations are always bad (especially when embedded devices are allowed - which if I'm not mistaken is true).

Limiting resolution, or the way (all) assets are created is somewhat bad - that is basically deciding art direction in competition mainly for people who do programming. Which is going to end in lower number of participants.

I could imagine requiring a software renderer, or requiring a ray tracing for at least some effect. It would, sadly, remove any beginners or people who don't have experience with tons of optimization and rendering.

Most of these would also mean that you need to write completely whole engine, that means either:

• Making competition time too long (6+ months), as most people don't have time to work on this full time due to their real life jobs, long competitions tend to have very few participants though

• Nobody is going to participate - to be fair - I can probably reserve 2, maybe 3 weekends over 6 - 8 weeks to work on this. It is not possible to write whole engine with high performance software renderer (or using specific hardware software - which I might not have used before) and whole game on top of it with art in that time range

Not sure why you are so defensive to my ideas, this is a 'challenge', if you don't want to get out of your comfort zone, what is the point in participating?

I also wonder, which language would not support only keyboard input and what software would not support mode13h style output?

I don't know why the twist has to always be something silly, inducing gameplay. "Make Doom, but you can only jump like a frog". Why can't we for once take what is special about Doom? Art is always limited by hardware, giving people the joy of creating art within the limits Doom had is actually a nice change. Not having to waste resources on creating modern art, but oldschool, can actually empower productivity. (That's why so many indies go for old-school art, it's less time demanding). Otherwise you end up with always the same looks with different game names stamped on it. And honestly, what's the point of "Doom" as a theme if the entries end up being generic shooter with some twist that really could be applied to Duke3D or Serious Sam or ...?

There are software renderers, you don't have to write one yourself. For ray tracing you could even use UE4 if you don't want to write it yourself. On the other side, if you decide to write your own, it's not that problematic with modern CPUs anymore, at this low resolution, you can run simple tracing without 6 months of optimization (how do you come up with that random number? This sounds like "I will make it sound bad with made up, ridiculous sounding facts, cause I don't like this", similar to "Using only keyboard, mode 13h or similar - are basically limiting your hardware, and also software (+ languages you can use)" )

Well, this is a gamedev challenge. You have to program something. Not sure what's wrong with it. Sure scripts from the unity asset store and call it a game, but in this case, why would you even spend time throwing out something generic? For total beginners, that's valid, someone who just learns art or programming, can take the other part, he is challenged enough. But a skilled programmer or artist should not just go for the low hanging fruits. At least that's how I approach "challenges", get out of my comfort zone, make something you haven't done before, learn a little bit, show something embarrassing and be proud you've done it all yourself, no matter how many eyes will bleed due to my coder art. if you just want to do what you always do, you don't need to take part in a 'challenge'.

1 hour ago, Vilem Otte said:

- as Doomguy is generally tough guy - participant has to use his own voice to add audio of toughest quotes played when dismembering demons.

That would be problematic for all the introvert, that don't want their personal stuff thrown out in the public. And maybe for many non-native english speakers as well.

30 minutes ago, fleabay said:

Was there a twist on the Frogger challenge? I don't remember one. Why does there have to be a twist?

The twist of no weapons on Side Scrolling Arcade Shooter Challenge was OK because it was a genre, not a game. Doom is a game, not a genre. If you wanted to put a twist on an FPS, you should have called the choice "Old School FPS" instead of Doom.

Very true, "make Ikaruga, but without bullets" would be quite odd. A non-Doom related "twist" would feel like a lost opportunity for this round. And if there really has to be a "twist", I'd prefer if this challenge was renamed to "FPS" and not "Doom". (Cause I really would love to make some Doom game, but depending on the "twist", I might pass on the FPS. If it was "Doom" but non-Doom, it would feel really sad to me). Maybe "Doom" would get a proper chance another time.

Edited by ProfL

2 hours ago, ProfL said:

Not sure why you are so defensive to my ideas, this is a 'challenge', if you don't want to get out of your comfort zone, what is the point in participating?

Alright - so let me elaborate. My reaction to points proposed by you were from the experience of someone who has already hosted multiple events (and not just in gaming). Generally statistics says that the longer the event, the less amount of participants there will be. Also if you look into past challenges, you will see that even here on Challenges the amount of participants isn't high.

To see basic statistics - compare amount of successful participants on Ludum Dare (which is a weekend jam) to Lambda Cup in Run Think Shoot Live (which is technically a yearly cup with multiple maps you create - just few people participate). Increasing scale of the project - in both, time and complexity, will decrease number of participants.

Now, I have used myself as example in how much time I can spend on the project - this was a MISTAKE on my side. I should have used an example of this - you have someone working full time job (i.e. 9-5). Therefore over the course of 8 weeks, he can work on 16 days at most (i.e. the weekends), assuming he won't do anything else and fully commits to the project.

You can argue whether he can work after he returns from work, or such ... which is a speculation and will differ for each individual. Someone has a pet, someone has family, someone has another hobbies.

Now, sorry for going a bit off topic in next paragraph (especially to moderators), but I believe I have right to defend myself here ... while I do believe your intentions were to motivate me - and I have already had my fun and experience with high performance ray tracers while on university (both - my Bachelors and Masters were about them), and I do know that working with such is enjoyable. I simply have to oppose the "get out of your comfort zone" here.

Without going too much into personal and work details (as most of it is classified) - game development is just a hobby for me. Stopping my company/work completely for 2 months straight and denying any support which I'm legally bound to provide, would cause serious problems (and not just economical and legal).

2 hours ago, ProfL said:

﻿ I also wonder, which language would not support only keyboard input and what software would not support mode13h style output?

I already gave you a hint - embedded devices - if I'm not mistaken doing a mobile game running on Android or iOS is a valid entry, using 13h there would be a major pain. Also, for same devices requiring keyboard entry, while possible, would be quite painful.

All I want to do is recreate a Doom looking game with my own graphics and not be bogged down by too many requirements.  I've been coding for a long time, but I'm also so time constrained that if I have to worry about too many technical details I will never finish. I also use these challenges as an opportunity to create graphics, music, and sfx all on my own.

I hope the twist wont be a deal breaker. 😮

@ProfL Please give a beginner with 5 lessons C#, and 2 Unity tutorials also the option to participate in this friendly dev challenge.

The challenges represent different things / goals for different people. For some it might be recreating the exact spirit of the original in mode 13h, for others it might be an opportunity to make their first game, others to practice / learn a new 3d engine with some friendly bouncing of ideas etc. I agree that putting in too many constraints only serves to decrease the audience interested in taking part. For most mere humans, completing a game, *any* sort of game in these timescales is a challenge in itself!

It also appears that regularly people doing the challenges have restricted themselves to certain tech (e.g. GoliathForge's homemade 2d engine for side scrolling shooter, DexterZ's froggy hop), and that is great to challenge yourself, but imo there's no need to restrict everyone to the same approach.

On 8/11/2019 at 6:55 AM, Alberth said:

@ProfL Please give a beginner with 5 lessons C#, and 2 Unity tutorials also the option to participate in this friendly dev challenge.

No one and nobody would ever constrain you from participating. Neither C#, nor unity is a crime. Take part for the sake of learning new things. It might be your 6th, most valuable lesson.

On 8/11/2019 at 1:28 AM, Vilem Otte said:

I already gave you a hint - embedded devices - if I'm not mistaken doing a mobile game running on Android or iOS is a valid entry, using 13h there would be a major pain. Also, for same devices requiring keyboard entry, while possible, would be quite painful.

Both statements are simply not true. Mode 13 is so simple, that doom was ported to many many devices, exactly for that reason. Doom on Android?

There is a list: https://en.wikipedia.org/wiki/List_of_Doom_source_ports

And just a week ago, Doom on Switch was released, which is a port to Unity3D, running on Switch.

Just don't be scared due to false assumptions. Mode 13 is nothing more than drawing pixel on a 320x200 screen with just 256 colors. You can do that in any language, with every API and any engine you want. Saying this is too complicated is like saying an apple for dinner is too complicated, cause people want simple pasta-cheese.

Anyway, the terms are set: " the game must recreate the Doom experience and style of gameplay"

1 hour ago, ProfL said:

Just don't be scared due to false assumptions. Mode 13 is nothing more than drawing pixel on a 320x200 screen with just 256 colors. You can do that in any language, with every API and any engine you want. Saying this is too complicated is like saying an apple for dinner is too complicated, cause people want simple pasta-cheese.

Sorry, I have to react once more here. No, you simply can't run mode 13h on Android without EMULATING it, which will result in poor performance. Not to mention ugly stretching on the screen.

Again, why going through the pain of making something - that is clearly going against your hardware?

1 hour ago, Vilem Otte said:

No, you simply can't run mode 13h on Android without EMULATING it, which will result in poor performance.

I didn't want to intervene, but purely as a graphics guy, I'm curious, why can't you? What exactly would cause poor performance when you'd transfer images of 64000 bytes? What exactly would involve the  "EMULATING".

1 hour ago, Krypt0n said:

What exactly would involve﻿﻿ the﻿  "EMULATING﻿﻿﻿"﻿.﻿

So, mode 13h was introduced in 1987, and widely used in games at the time. The original interface between computer and monitor was VGA, at the time it had only one 256-color mode with resolution of 320x200 and that is mode 13h.

It is, as you correctly mentioned 64000 bytes of video memory you can write into.

Now let's dive a bit into software - how do we set it? Like this (I will use NASM like syntax):

mov ax, 13h
int 10h

Next, we can access video memory directly at address 0xA000 (i.e. range 0xA000:0000 - 0xA000:F7FF). Like (this sets pixel in location [3,3] to color 9 from color palette - which is light blue):

mov es, 0A000h
mov di, 960
mov al, 9
mov es:[di], al

This level on access is simply not possible on Android at all. You have to emulate the behavior - most likely you will have to create 320x200 buffer in memory and lookup table (for palette - from index to actual rgb color). This buffer can then be used as texture and rendered through OpenGL ES to screen for example. Any way you do it, you will have to emulate this behavior - as hardware and software (mainly operating system) simply doesn't allow you to do this.

Of course it isn't hard, and there are different ways to actually do it - some faster, some slower.

As for performance, if you go the true old school way (i.e. naive emulation) - you will be writing every single pixel, ideally 60 times a second. That is 3.84 MB of data written each frame into that buffer (which you will have to display somehow - I.e. update texture data on GPU with your buffer and display it - 60 times a second). It is not that much indeed.

Of course you can directly render into 320x200 render target with fragment shader simulating palette. Performance hit of doing this is extremely small.

EDIT: Small edit, for performance - if you would decide to go the first way (i.e. write into memory and then transfer), doing it pixel by pixel may actually be too slow on slower embedded devices (especially if you write the code directly in Java).

I re-watched the video on Brutal Doom posted by @ProfL and it is not technically DOOM as I remember it (this one has 'mouse like look'), and it was actually using common configuration for moving around (left hand moves up/down and left/right ... right hand points the camera). This alone supports my statement that keyboard controls would be a major pain on some devices.

This being said, I believe I will find some time to invest into making a competition entry, and maybe even @ProfL will be surprised about what I'd like to do. Hopefully I will manage to find enough time to work on it.

Posted (edited)

I think it was mentioned before, mode 13h rendering only means you render with 320 by 200 pixel with 256 colors. An example of how you would set it up in Java/Android/C#/XNA/"every high level" : https://gamedev.stackexchange.com/questions/38373/how-to-emulate-mode-13h-in-a-modern-3d-renderer

Similarly other retro-render techniques, like "Mode 7", "Pixel Art", "Tile Based", only mean how you prepare the output, it does not mean to emulate the hardware. For example, there are plenty of tile based games on Android, although there is no Android phone with tile based hardware.

But maybe you unintentionally proof yourself right. Someone who has no idea of what Mode 13h is, will most likely be overwhelmed of it or even if they'd try, they'd spend more time setting up something super complicated, than working on the game.

Have fun everyone, do it the way you can. You have 2 whole months of time, don't hesitate to try silly things you've never done before, your biggest limit is that tiny voice inside that makes you give up before you even try.

You don't have to clone that, but it's interesting to know how things were done before maybe some of you were even born. IdSoftware always made these solutions that nobody believed would be possible and at early times, people called it even crazy.

Edited by Krypt0n

13 hours ago, Vilem Otte said:

This being said, I believe I will find some time to invest into making a competition entry, and maybe even @ProfL will be surprised about what I'd like to do. Hopefully I will manage to find enough time to work on it.

I'm looking forward to see your game! I will try to surprise you as well 😎

## Create an account

Register a new account