The game is now live on Steam .... get it now with a 15% discount for the first week. We also made it on to #EpicFriday - https://twitter.com/UnrealEngine/status/847785174154723329
Get it now on Steam - http://store.steampowered.com/app/552370

As we're going into Greenlight soon, I've been looking at a lot of games on GL to try work out what is good/bad.
I typed out notes while doing that, so I figured I might as well post them here. This is just my personal preference of stuff, but at the very least you'll now know the opinion of some random guy on the internet.
From the amount of crap that actually gets greenlit, I'm not sure any of this is worth it, but there isn't much choice.
Bad Icons
Generally, icons that I wouldn't consider clicking had:
- Just a game title. Which, considering I don't know the game, tells me nothing
- Enough elements (be it text or images) that it is hard to know what is going on
- No information about the game
- Icons that showed obvious min-effort
- Girls in skimpy clothing. My brain automatically assumes that if you need to sell a game with T&A, then it probably isn't a great game.
There are tons of min-effort examples, but these were ones that were bad because of descisions, not due to lack of skill.

Good icons
Icons that interested me enough to click them had one of more of these things:
- A good GIF. Movement is just obviously more noticeable, but it also gives you more time to show some of the game.
- Clear title. I don't know why I wanted them to have the title on the image, but the ones without it attracted me less.
- Clear what is going on
- Shows roughly what style of game it is, or what the game is about
The icon we made for VALA (posted in the VALA thread), satisfies all of these (IMO :P).

Something interesting to note about icons, is that the specs (512x512 and < 1MB) that the GL submission page says are actually wrong. It can be up to 2MB (according to the error message that pops up when you try submit one over 2MB), and can be any (?) resolution.
I also watched tons of trailers. Here are some points I wrote:

- Should go almost immediately into gameplay
- Minimal story elements - this is greenlight, I don't care about the story yet
- 2 minutes is too long. 1:30 is probably around the max it should be
- Don't waste the first 10 seconds showing logo's and such. Do it at the end.
- Music, especially when well timed makes a mediocre trailer into a really good one
- Framerate matters a lot. Lot's of comments and complaints about video with crappy framerate. There really isn't any excuse for it.
- Don't show the game engine name. Why do I care what you made it in?
Trailers I'd consider good:
https://steamcommunity.com/sharedfiles/filedetails/?id=813827321 (this breaks most of what I just mentioned, but the editing is really good)
https://steamcommunity.com/sharedfiles/filedetails/?id=817836436 (I think the music works great here)

Descriptions:
- Big walls of text look crappy
- Splitting description into sections with a graphic heading looks good
- Don't use lots of images, since there are screenshots at the top. But a few GIFs can work
- The type of game (i.e topdown twinstick) should be in the first sentence. It is what shows when you hover over an icon.
Although I think it is too long, and goes into detail that I don't care about (i.e. the credits), I think this description is pretty good https://steamcommunity.com/sharedfiles/filedetails/?id=817836436

Question: How can I add images to a spoilers box so it doesn't take up the whole page? Adding it in spoilers doesn't work.
========
So much fun XD... My team has done a great job with the island and basebuilding - we've got some minor bugs to fix before we release an early Testers build.
Once the test build is live, we'll finish up the save states - clear out more bugs and begin with the bases and resources :D
MAKING PROGRESS! :D
Gameplay Pics:
========

Hi if anyone is using Cocos2d-x, I seriously need help.
When we started our prototype "Who must fall" we were using frame by frame animations and later chose to switch to skeletal animation. You can see below that the framerate before optimization is drawing around 40fps

and immediately after I replace the sprites with skeletal objects, I get a drop in framerate by half which makes the game almost impossible to play.

In the case of above screenshot, I only loaded one spine object, but it seems like it will load every time before it spawns. So is there a way to batch the object so that it only load once? or perhaps is there another way to spawn or delete the objects once its no longer useful. I cant find any useful example online. So can anyone please provide an example.
some anim we made...

@chungking2339, thanks a lot! I would appreciate the help very much. Looking forward to learning from you!

Wow chungking2339 . nice stuff . I think your skills would be great for this game

FaraiV1;46264" said:

I would love to be involved, however, I am still in the process of learning 3D modelling fully (although I feel pretty good in making character models).

FaraiV1= If you need any help to learn any tip for unwrapping, modeling, software help, shoot me a line i can help you out if need a hand on learning.
chungking2339@gmail.com

I could help out also , my portfolio
http://joncipriano.daportfolio.com/gallery/510207
been in the industry for 11yrs
chungking2339@gmail.com

ill work to help you out get you on the map, if your interested hit me up at chungking 2339@gmail.com

TL;DR - Dev log of me learning shaders by trying to recreate pics/effects I find online. :)
Over my game dev career. I've come across these black magic programs they call shaders. I'd see something like this

and ask "How do you do that?". The reply would be "it's a shader". In my attempt to become one of these magicians I'd start a shader tutorial series. It'd be blah blah blah runs on the GPU, something something lighting, pixar, blah blah blah lambertian, dot product and you'd end up with

-_______-
I picked up the basic theory. You've got vertices, move them and morph the mesh. You've got pixels, change their color. Oh, and it runs in parallel.
Then I had to write my first production shader. It was for Raptor Polo. We needed to be able to have player colors. There were a ton of frames of animation. The process of splitting the clothes from the orc, grayscaling them and puting them separately was infeasible. So I wrote a shader that did that. This was in Unity3D.

Next shaders were written for a WEB GL. Passed the shader audio frequency data to offset vertices as a song played. Eventually the shader and the music sculpted a sphere into a "rose". And then based on the offset the color of the vertex. That's why its darker closer to the center. Thanks CT mini jams.

Started to get excited about shaders. Next shader was for Semblance. Needed to recreate the lighting streaks like in this concept piece.

This was a fun challenge. Learnt about shader buffers and using more than one pass. Used a "masking effect" to get the light streaks.

Over this period I've been collecting images across the net of shaders I'd like to attempt to make. This thread is gonna be like the art dumps but of my progress learning shaders.

Sigh_Leeeee;45888" said:

I do think there is value in getting good at marketing, and that's something that requires practice.

I think that marketing your game and yourself is now a given when 38% of all games on Steam were released this year!
Building an audience is now as, or even more important than building the game itself :(

https://twitter.com/Steam_Spy/status/804072335997358084

Yowza. This is...not great.
I dunno if the indiepocalypse is real or not, but man, but this kind of supply pressure can't help but cause a collapse in prices.
https://twitter.com/Steam_Spy/status/804072335997358084

Hi All,
I'm excited to announce the launch of the Serious About Games Competition!
Are you serious about games? Interested in the use of games to bring about social impact and innovation? We’re launching South Africa’s biggest serious game competition to engage communities in reimagining their spaces for economic prosperity.
THE CHALLENGE
Apartheid urban planning has led to a highly split society and huge socio-economic challenges across the Western Cape. Let’s create a game that allows poor communities to think innovatively and reimagine their neighbourhoods for economic growth.
Serious About Games is challenging all current or aspiring game developers, digital creatives and designers to use a user-centred design approach to conceptualise a game that addresses the above challenge. The winning team will receive R1 000 000 (yes R1 Million!) to develop the game in 2017.
The Launch Event
When: 16 November, 6.30pm
Where: Bandwidth Barn, 3rd Floor, Block B, Woodstock Exchange
Who should come? All game developers, digital creatives, design thinkers, urban/ civic planning organisations – anyone who is interested in game development or civic planning.
What: We’ll be introducing the collaboration, the game brief, and the competition schedule – all over a drink and a snack.
Click attending on the Facebook event, or follow us on Twitter.
Get your free ticket.
This competition is led by the Cape Innovation and Technology Initiative (CiTi) in collaboration with Interactive Entertainment South Africa (IESA), 67Games, the Cape Craft and Design Institute (CCDI), and supported by the Western Cape Government’s Department of Economic Development and Tourism.
Serious About Games will be hosting a series of workshops and gaming popup events over the next few months, to equip teams with knowledge on serious games, relevant business acumen and allow them to engage with their audience.
Come along on the 16th to find out more!
************
Update, the submission deadline was on Monday, and we have had 17 Submissions in total, which is fantastic for a first time running and the niche nature of the competition. All the entries look fairly solid and there is potential for some excellent games to come from this!
Thank you to everyone from the community that entered, and good luck!
*************
We will be announcing the winner tomorrow at an event at the Bandwidth Barn! I hope that I will be seeing a lot of you there. This is a really exciting point for the game dev community. The backers of the competition seem to be pleased with results of the competition, and I will definitely be looking to make this an regular event!
You can get your tickets online here: https://www.quicket.co.za/events/27351-the-final-sas-biggest-serious-games-competition/#/
**************
Congratulations to @Shanemarks and Renderheads on winning the R1million prize for the Serious About Games Competition!

1&2. Yeah, I understand: elevation itself isn't really important, and if that gets in the way of actually playing, then by all means, don't do it. But I feel like there are many games (e.g. RTS, hack 'n slash) that have solved this (i.e. (1) camera is from above, not necessarily completely top-down (which may better fit a 3D game anyway), (2) gameplay that plays on what's largely a flat plane, (3) still reads as background, so isn't so loud as to impede on other gameplay information, and (4) still nice-looking, and doesn't look flat).
A good recent example, I think, is Helldivers (which has a lot to study actually, including gorgeous, juicy explosions, screen shake and post fx). It's hard to find a screen shot that's plain and doesn't have explosions XD

3. Adding camera shake doesn't have to mean giant nausea-inducing shaking. It does depend on the game (I think camera shake can have its own personality), but in a game like Nuclear Throne, which has very punchy visuals and audio, the camera shake seems to stop in like 0.01-0.2 of a second. Even if it shakes for just 1 frame in a 60 fps game, and you probably don't consciously spot it, you feel it, and when it's not there it just feels less powerful, and you lose one channel of feedback. It doesn't have to be all or nothing either. Big explosions can have big camera shake. Small explosions can have small camera shake. The vast majority of bullets may not need any at all. Some weapons, especially ones that shoot fast-moving, infrequent projectiles, could benefit from a 1-frame shake the moment they fire. Shake can be weighted so that you don't feel any if things explode far enough away from your character (or the centre of the screen, though these seem to be the same thing).

And I just watched a gameplay video of Dead Nation, and it does actually have camera shake (although admittedly very, very seldom). e.g. 9 seconds in

It's really just the punchy feeling I'm trying to describe. Nuclear Throne may do that with camera shake. Dead Nation may not do it with camera shake, but then does it instead with 1-2 frame muzzle flashes. A game like One Finger Death Punch has almost no camera shake, but still feels juicy because the actions and camera snap instantly with every move (and in retrospect I suppose you could argue that a camera snap in that case kind of acts like a camera shake anyway). Personally, for an action-oriented game, I'd probably try and create that punchy feeling in every channel available to me (including -- or especially -- audio). (If I'm using multiple methods I'd probably tone it down proportionally.)
I dunno if I'm overemphasising this. I imagine there may be other things that are higher priority. I do think camera shake is pretty low-hanging fruit though; adding it in the first place is (usually) almost trivial, and I think most of the time goes into picking the right amount per item/event to make it feel good.

In an effort to be less crap with social media things, we've recently set up a facebook page. Like us on there, or follow us on twitter for some sporadic updates of our adventures making things for games.

I guess data visualization could be procedural generation. Here are some experiments I've done with using audio frequency data to generate 3D terrain.

Zaphire;42338" said:

00:35 seconds was the best part. (Archer in the cabin)
That part looks really beautiful with smooth animations

Agreed, that was the best part for me too.

Shift the tiles around and and match them up. Soko Match is a cross between Sokoban and a Match-3 game. You can match 3 or more tiles in a row or column with the same color or number. Can you get to 15 matches?
Play on itch.io: https://francoisvn.itch.io/soko-match
Ludum Dare entry: http://ludumdare.com/compo/ludum-dare-35/?action=preview&uid=23363
So I made a game for Ludum Dare 35 this weekend. Yay! The venue at the Bandwidth Barn was great for Saturday, and I worked on it a bit more on Sunday, but I would have liked to spend more time on it. I think there are some interesting interactions going on, but I think the space hasn't been properly explored yet. If I have some time I might spend another day or so exploring the design space, but I'll probably wait till the Ludum Dare judging is finished at least.
So if you have a moment, check it out and see what you think. Any interesting ideas for design space I should explore?

Antoine Routon - Distilling A Franchise: A Lara Croft GO Postmortem

So, links to cool free content people want to watch?
Here's the 1ReasonToBe panel that Tsitsi and Sithe utterly rocked!
Alexis Kennedy's (Failbetter Games, Sunless Sea) talk about narrative - This is my best talk of GDC.