Vapourware Codexian Game Development Thread

[TheDeveloperDude]

My fiancee helped me with the color palette and I think this gives me an extra point for the graphics! The colors should blend together better.



Old:

View attachment 34972

New:

View attachment 34973

I skipped a few pages in this thread. This looks familiar though. Do you have a sparkly turd enemy in your game? Are those walls around the level? Stretched textures make me irrationally angry. I think it would look better if your wall texture had the same pixel density as the other textures. The new palette isn't necessarily better or worse. I think the right palette depends on the feeling you want your game to evoke.

Less than 10 items left on my TODO list, then it's testing/balancing time. The game plays like Wizardry IV with some elements from SaGa series. It's somewhat streamlined due to time constraints, but I hope people will like it.

I really like that dithering. That's a fun CRT filter you have going on too. The ability names are a bit childish. Where are you getting your battler sprites? You just taking other images and running them through a contrast filter? The pixel work looks too clean for just that, so I'd imagine you're actually putting some real effort into making them. But then again, I could have sworn I saw a Naruto that you copied/pasted into your enemy lineup.

Any idea how to make this better graphics?

I know only about the Medieval Total War series.
But look this, dots are fighting dots. It looks as a real mess. Yes you can zoom in, but are who using that feature in a serious battle?

You're asking how to improve your graphics? Are you just shitposting here, or do you actually not see any way to improve your graphics? Can you literally not think of anything that would improve the look of your game?

Yes, I am asking.

 

[zwanzig_zwoelf]

I really like that dithering. That's a fun CRT filter you have going on too. The ability names are a bit childish. Where are you getting your battler sprites? You just taking other images and running them through a contrast filter? The pixel work looks too clean for just that, so I'd imagine you're actually putting some real effort into making them. But then again, I could have sworn I saw a Naruto that you copied/pasted into your enemy lineup.

I used character sprites from Toriotto, who did sprites for Javardry, and allowed to use his work in other creations: https://www.pixiv.net/en/users/5887541/artworks?p=1

They are processed through conversion to a 1-bit color palette and downscaling to the target resolution (160x160), which in combination with the CRT filter worked wonders in this case.

As for the abilities, there are just a few cases of humorous abilities when they make sense, e.g. Goblins are supposed to be gross and use underhanded tactics, so why not use that to spice things up once in a while? It's not like you're seeing this stuff all the time.

 

[Nathaniel3W]

TheDeveloperDude go get yourself a subscription to Inkarnate and use that to make a bunch of battle maps. You can also use it to make your castle walls.

Then tell me which medieval strategy game has the best user interface, and then copy it.

 

[Moaning_Clock]

I skipped a few pages in this thread. This looks familiar though. Do you have a sparkly turd enemy in your game? Are those walls around the level? Stretched textures make me irrationally angry. I think it would look better if your wall texture had the same pixel density as the other textures. The new palette isn't necessarily better or worse. I think the right palette depends on the feeling you want your game to evoke.

It is a new project which I started last Monday and submitted on Friday. No. Yes, these are walls. My fiancee also dislikes the stretched textures, I thought about one single color but thought it was worse and then I didn't bother. Especially while playing the contrast was stronger and harder on the eyes (she improved the water in my other game too https://rpgcodex.net/forums/threads...ed-dungeon-crawler.138399/page-6#post-7719602)

 

[Nathaniel3W]

View attachment 34971

Sparkly turd.

 

[Moaning_Clock]

Sparkly turd.

At least it sparkles, eh?

 

[TheDeveloperDude]

I am making a new randomly generated strategy world map instead the all-squares version.

 

[Justinian]

Apparently people have been emailing me asking for keys. Mostly streamers & a couple of curator groups. Some of the replies I sent bounced, and I think it's because some poor sods were pretending to be a streamer to get a free key & gmail nuked them.

Some claim a viewer asked for a review, which I highly doubt given that my game has a couple hundred wishlists.

Either way, a lot of unnecessary effort when I just gave a free key to anyone who asked.

In related news, my game is basically done. Probably still need to deal with bugs, and maybe fancy up a couple of places that I kinda phoned in (particularly the Library where all the metaupgrades are), but right now I don't plan to add any more features or options unless something really good comes to mind.

 

[zwanzig_zwoelf]

Apparently people have been emailing me asking for keys. Mostly streamers & a couple of curator groups. Some of the replies I sent bounced, and I think it's because some poor sods were pretending to be a streamer to get a free key & gmail nuked them.

There's a 99% chance that key requests via e-mail are from scammers, so I'd suggest using Keymailer if you want to get some coverage on YouTube.

That said, if you can create a trial version of your game and generate tons of keys for this particular version and just send them to potential scammers, you can probably get some free advertising by generating butthurt.

 

[The_Sloth_Sleeps]

I was able to fix sprite draw ordering. It becomes a problem when sprites are bigger than a single isometric tile. It could be worked around if only character movement was stuck to the grid, but actually I prefer seamless movement by a whole lot.

View attachment 34840

This video was very informative:

I also use monogame. I came up with my own way of doing it which is rather trivial.

I create a depth encoding layers. Simply take your object texture, and draw over it where you want precedence of depth.
You can also combine encodings, to decide a precedence of depth e.g. if I touch this and that combo, then under, useful for protrusions like arms. The end result is decided and drawn on a pixel by pixel basis.

I made things a bit harder by having a sprite shadow. I later switched to a circular shadow which looks nicer and is easier to work with.
I show a quite complex situation here:


I could get it better actually by adding further layer encodings. But its a bit of a tradeoff between time and effort, is it worth it. Not really.

 

[The_Sloth_Sleeps]

Working on a level editor for a tactics game

Nice graphics! Like a professional Ultima 8 look (Regardless of it's gameplay I always had a soft spot for the graphics in that game).

Looks really nice. Monogame is a good choice.

 

[Twiglard]

Added Z layer as property of each chunk. I haven't implemented topological sorting static sprites though, which is necessary for everything rendering in the correct order. Maybe most of the errors can be masked and I won't have to bother with it. But these links might be useful to you.

• https://gamedev.stackexchange.com/q...-map-draw-sort-with-z-buffer-and-anchor-point
• https://forum.gamemaker.io/index.php?threads/smart-isometric-drawing-order.57356/#post-347325, https://shaunlebron.github.io/IsometricBlocks/

It's possible to sort them using O(n2) topological sort. But that requires more VBO uploads each frame (or not batching draw calls [1]). Right now static parts (non-animated scenery, floor, walls) are only uploaded when the map is edited.

I also found this regarding 2D lighting.

• https://github.com/mattdesl/lwjgl-basics/wiki/2D-Pixel-Perfect-Shadows
• https://slembcke.github.io/SuperFastHardShadows
• https://www.gamedeveloper.com/programming/fast-2d-shadows-in-unity-using-1d-shadow-mapping, https://github.com/nemestniy/WGA_TEAM_8/blob/master/1DShadowMapDemo.rar

[1] Batching isn't done yet, it's going to need a maximum-size 3D sprite atlas to avoid binding textures. Too early in development.

I also use monogame.

The video is lacking in CornflowerBlue, disturbingly enough.

 

[The_Sloth_Sleeps]

Added Z layer as property of each chunk. I haven't implemented topological sorting static sprites though. Maybe most of the errors can be masked and I won't have to bother with it. But these links might be useful to you.

• https://gamedev.stackexchange.com/q...-map-draw-sort-with-z-buffer-and-anchor-point
• https://forum.gamemaker.io/index.php?threads/smart-isometric-drawing-order.57356/#post-347325, https://shaunlebron.github.io/IsometricBlocks/

It's possible to sort them using O(n2) topological sort. But that requires more VBO uploads each frame (or not batching draw calls [2]). Right now static parts (non-animated scenery, floor, walls) are only uploaded when the map is edited.

I also found this regarding 2D lighting.

• https://github.com/mattdesl/lwjgl-basics/wiki/2D-Pixel-Perfect-Shadows
• https://slembcke.github.io/SuperFastHardShadows
• https://www.gamedeveloper.com/programming/fast-2d-shadows-in-unity-using-1d-shadow-mapping, https://github.com/nemestniy/WGA_TEAM_8/blob/master/1DShadowMapDemo.rar

[1] At least except assigning different depth values to individual vertices of the floor, wall, scenery quads.
[2] Batching isn't done yet, it's going to need a maximum-size 3D sprite atlas to avoid binding textures. Too early in development.

I also use monogame.

The video is lacking in CornflowerBlue, disturbingly enough.

So your man there, if he is wielding a very long pike in front of him. How would the system perform?
I imagine you have to split the pike up into several sprites or...?

 

[Justinian]

Apparently people have been emailing me asking for keys. Mostly streamers & a couple of curator groups. Some of the replies I sent bounced, and I think it's because some poor sods were pretending to be a streamer to get a free key & gmail nuked them.

There's a 99% chance that key requests via e-mail are from scammers, so I'd suggest using Keymailer if you want to get some coverage on YouTube.

That said, if you can create a trial version of your game and generate tons of keys for this particular version and just send them to potential scammers, you can probably get some free advertising by generating butthurt.

Any advice on final release? Should I do a bunch of advertising first and then release, or try to get some wishlists/reviews before publishing?

 

[zwanzig_zwoelf]

Apparently people have been emailing me asking for keys. Mostly streamers & a couple of curator groups. Some of the replies I sent bounced, and I think it's because some poor sods were pretending to be a streamer to get a free key & gmail nuked them.

There's a 99% chance that key requests via e-mail are from scammers, so I'd suggest using Keymailer if you want to get some coverage on YouTube.

That said, if you can create a trial version of your game and generate tons of keys for this particular version and just send them to potential scammers, you can probably get some free advertising by generating butthurt.

Any advice on final release? Should I do a bunch of advertising first and then release, or try to get some wishlists/reviews before publishing?

Get as many wishlists as you can. Successful indies tend to talk about 10,000+ wishlists being an optimal number prior to launch.

Advertising should be started long before the game is out, and not be the last thing you do before release.

As for reviews / coverage, I'd suggest sending the keys about a week or two prior to the final release to give people time to familiarize themselves with your game and write reviews / make vids / run streams / etc.

 

[Twiglard]

So your man there, if he is wielding a very long pike in front of him. How would the system perform?
I imagine you have to split the pike up into several sprites or...?

I'd have to split it, but the real challenge is doing it in a generic fashion, so that it could be used declaratively and without too much hassle for all large weapons.

 

[Justinian]

Apparently people have been emailing me asking for keys. Mostly streamers & a couple of curator groups. Some of the replies I sent bounced, and I think it's because some poor sods were pretending to be a streamer to get a free key & gmail nuked them.

There's a 99% chance that key requests via e-mail are from scammers, so I'd suggest using Keymailer if you want to get some coverage on YouTube.

That said, if you can create a trial version of your game and generate tons of keys for this particular version and just send them to potential scammers, you can probably get some free advertising by generating butthurt.

Any advice on final release? Should I do a bunch of advertising first and then release, or try to get some wishlists/reviews before publishing?

Advertising should be started long before the game is out, and not be the last thing you do before release.

 

[The_Sloth_Sleeps]

So your man there, if he is wielding a very long pike in front of him. How would the system perform?
I imagine you have to split the pike up into several sprites or...?

I'd have to split it, but the real challenge is doing it in a generic fashion, so that it could be used declaratively and without too much hassle for all large weapons.

Yes this is what I talk about. Finding a general solution is quite difficult, especially for different size or large sprites, or narrow cooridoors etc. With my approach however I can simply paint the sprite depth as I please, that is the sprite depth is decided on a pixel by pixel basis.

 

[Twiglard]

I create a depth encoding layers. Simply take your object texture, and draw over it where you want precedence of depth.
You can also combine encodings, to decide a precedence of depth e.g. if I touch this and that combo, then under, useful for protrusions like arms. The end result is decided and drawn on a pixel by pixel basis.

Maybe switching to 3D models will be useful at some point? There are permutations over character rotation, armor, weapon type, animations, and character's sex.

I'll be looking into writing a shader making 3D models more bearable at some point.

I show a quite complex situation here:

In the video, your character doesn't animate fast enough compared to how fast he's moving.

But the tileset looks amazing.

 

[The_Sloth_Sleeps]

I create a depth encoding layers. Simply take your object texture, and draw over it where you want precedence of depth.
You can also combine encodings, to decide a precedence of depth e.g. if I touch this and that combo, then under, useful for protrusions like arms. The end result is decided and drawn on a pixel by pixel basis.

Maybe switching to 3D models will be useful at some point? There are permutations over character rotation, armor, weapon type, animations, and character's sex.

I'll be looking into writing a shader making 3D models more bearable at some point.

I show a quite complex situation here:

In the video, your character doesn't animate fast enough compared to how fast he's moving.

But the tileset looks amazing.

I will never use 3d to achieve what I want. Yes the art is amazing and I cannot take any credit for most of it, which is the work of this incredible artist https://ember-atelier.tumblr.com/

 

[Tavernking]

I am working on one of the greatest indie RPGs of all time (at least from my perspective). Since last year I've been working on it in secret and think I shouldn't announce it until there's a demo so that people have something to bite their teeth into from the moment they hear about it. Compared to my last project where it was all news news news and very little playable demos until it fizzled out and went on an eternal hiatus.

Is it better to start advertising it early or should I continue with the plan of keeping my head down and focusing on the game until there's a demo to play?

 

[Zanzoken]

I am working on one of the greatest indie RPGs of all time (at least from my perspective). Since last year I've been working on it in secret and think I shouldn't announce it until there's a demo so that people have something to bite their teeth into from the moment they hear about it. Compared to my last project where it was all news news news and very little playable demos until it fizzled out and went on an eternal hiatus.

Is it better to start advertising it early or should I continue with the plan of keeping my head down and focusing on the game until there's a demo to play?

You definitely don't want to wait until release to start marketing, however you need to approach it strategically.

Keep in mind I have released zero games so far, but here is what I would do in your position.

Spoiler

Pre-announcement

You shouldn't announce the game before you are confident that you have a firm grasp on the scope, systems, and production design (art, music, etc). Players do not react well to things like cut features, wholesale changes in art design, or anything that results in the game that ships being significantly different than their first impression. And under no circumstances should you announce a release date until you are 100% confident you can reach it.

A good example of what NOT to do is Mechajammer fka Copper Dreams, where they went so far as to launch a Kickstarter where (with the benefit of hindsight) we know they actually had no fucking idea what the game was. Unsurprisingly, after years of delays and sweeping changes, the developer burned through all their credibility and goodwill, and the game flopped hard (it was also really, really buggy at launch).

Announcement

Once you think you're ready to announce, you don't need a demo, you just need a good trailer to show that will start getting players hyped about the game. Marketing is all about teasing and building anticipation, NOT trying to prove to players that you have a good product. Even if your game is good, I think a demo will end up hurting your sales.

Have your Steam page, social media, and website (listed in order of importance) ready to go on announcement so that people can immediately start to follow you and wishlist your game. There are plenty of good tips out there on how to properly set up your Steam page for the most success.

Regular updates

You don't want to announce and then just disappear for months. Once you put the game into the public eye, you need to maintain a steady flow of content so that you keep reminding people and expanding your audience. I would aim for a monthly development update, plus occasional snippets of something cool to show. Have several months of this content planned out ahead of time before you announce, and then stay a few updates ahead so that you never miss one or are scrambling to put something together last minute. I would honestly aim to keep this practice going all the way through the post-launch.

Remember the purpose is to build excitement, not to vent your spleen about every little detail, so focus on things players will be excited about and make them want to play the game. Another thing that should not be a main point of focus, but something you sprinkle in here and there, is your own personal story. As a solo indie developer you are the ultimate underdog, and some people will respond to that and want to support you more as a result. Don't lose sight of the fact that the goal is to sell your game, not yourself, but if you can get people invested in you along with your game then that will pay off.

Kickstarter

Eventually, you want the game to be feature complete and have the first part of the content implemented (like the first major town, quests, etc). Now you are ready to go to Kickstarter. Even if you don't need money to finish the game, Kickstarter is still a good idea for a number of reasons. The people who pledge to your Kickstarter are your first true stakeholders in the game, the seed that will hopefully grow into a large audience.

You can find a lot of info out there on how to run a successful Kickstarter so I won't go into that here, but you should plan carefully and dedicate 100% of your time to the Kickstarter while it is live.

Backer beta

Not long after the Kickstarter is done, launch a backer beta. This is the first time people are getting hands on the game, so the feedback will be coming fast and furious. You want to spend this time purely on bugfixing and community engagement. Dedicate a couple of weeks to this, then close the beta and incorporate all of the lessons learned. If for some reason there is some kind of major flaw in your game's design that you previously hadn't identified, now is the time to fix it.

You can do multiple backer betas if you need to, but the goal is to get the game's systems and early content to as close to a finished state as possible. By now you should also have a good estimate of how long it will take you to finish the game, so you know an approximate release date. But first you want to take advantage of Early Access.

Early Access

Set the EA release far enough in advance that you can get involved with media, streamers, etc and make a strong marketing push before it hits. This is your game's debut and first exposure in the wild, so you want it to make as much of an impact as possible.

Hopefully by now your game has a lot of steam (pardon the pun) behind it and the EA release hits. Ideally this is the point where you can afford to quit your day job and work on the game full-time. Make it clear on the Steam page where the game is in terms of content completion and how long it will reasonably take to finish.

You will need to spend more time fixing bugs and incorporating feedback, but hopefully a lot less than before since you already got at least one pass at this with the backer beta. People are more tolerant of issues in an Early Access release, but keep in mind from this point forward your game is being judged in the open marketplace. Everything you release from here should be as polished and production-ready as possible.

Finish the game

Now the goal is simple: finish the game. After each major content update, earmark time for bugfixing and increased community engagement. Try to make it so each update doesn't break people's saves if at all possible.

Once you can see the finish line for having all of the content implemented and polished, set a final release date. Make another marketing push like you did for the EA launch. Go as big and as wide as you can, because this is the climax of the entire development cycle and the best chance you will get to squeeze as much juice out of this peach as you can.

Release

Now you release! Do everything you can to promote the game and reach as many people as possible. Hopefully it goes well and now you're rich!

Post-release

Keep fixing bugs and releasing patches for as long as it makes sense to. Release on consoles, or make a deal with a publisher to port it for you. Start pre-production for the sequel.

tl;dr You just need a trailer to announce the game and start advertising, not a demo. Then it goes Announcement --> Kickstarter --> Backer beta --> Early Access --> Full release.

 

[Moaning_Clock]

Is it better to start advertising it early or should I continue with the plan of keeping my head down and focusing on the game until there's a demo to play?

I think the stuff Zanzoken wrote is good if you go the Kickstarter/EA route but I never did do that. There are some contrarians who say that wishlists get stale if you take too long from the first announcement until the release. I simply don't know. Both routes can work but the announce and release and do everything in 2 months route is way harder if you didn't do it already.

I would recommend: Announce it when you are sure that you'll finish it and have something to show. Not finishing stuffs occasionally is not a problem but there is a bad habit that some people have of announcing and abandoning projects.

The worst outcome getting demotivated because the early feedback of one rando was weird/bad and being more likely to abandon because of it.

regarding the last point

 

[Tavernking]

I am working on one of the greatest indie RPGs of all time (at least from my perspective). Since last year I've been working on it in secret and think I shouldn't announce it until there's a demo so that people have something to bite their teeth into from the moment they hear about it. Compared to my last project where it was all news news news and very little playable demos until it fizzled out and went on an eternal hiatus.

Is it better to start advertising it early or should I continue with the plan of keeping my head down and focusing on the game until there's a demo to play?

You definitely don't want to wait until release to start marketing, however you need to approach it strategically.

Keep in mind I have released zero games so far, but here is what I would do in your position.

Spoiler

Pre-announcement

You shouldn't announce the game before you are confident that you have a firm grasp on the scope, systems, and production design (art, music, etc). Players do not react well to things like cut features, wholesale changes in art design, or anything that results in the game that ships being significantly different than their first impression. And under no circumstances should you announce a release date until you are 100% confident you can reach it.

A good example of what NOT to do is Mechajammer fka Copper Dreams, where they went so far as to launch a Kickstarter where (with the benefit of hindsight) we know they actually had no fucking idea what the game was. Unsurprisingly, after years of delays and sweeping changes, the developer burned through all their credibility and goodwill, and the game flopped hard (it was also really, really buggy at launch).

Announcement

Once you think you're ready to announce, you don't need a demo, you just need a good trailer to show that will start getting players hyped about the game. Marketing is all about teasing and building anticipation, NOT trying to prove to players that you have a good product. Even if your game is good, I think a demo will end up hurting your sales.

Have your Steam page, social media, and website (listed in order of importance) ready to go on announcement so that people can immediately start to follow you and wishlist your game. There are plenty of good tips out there on how to properly set up your Steam page for the most success.

Regular updates

You don't want to announce and then just disappear for months. Once you put the game into the public eye, you need to maintain a steady flow of content so that you keep reminding people and expanding your audience. I would aim for a monthly development update, plus occasional snippets of something cool to show. Have several months of this content planned out ahead of time before you announce, and then stay a few updates ahead so that you never miss one or are scrambling to put something together last minute. I would honestly aim to keep this practice going all the way through the post-launch.

Remember the purpose is to build excitement, not to vent your spleen about every little detail, so focus on things players will be excited about and make them want to play the game. Another thing that should not be a main point of focus, but something you sprinkle in here and there, is your own personal story. As a solo indie developer you are the ultimate underdog, and some people will respond to that and want to support you more as a result. Don't lose sight of the fact that the goal is to sell your game, not yourself, but if you can get people invested in you along with your game then that will pay off.

Kickstarter

Eventually, you want the game to be feature complete and have the first part of the content implemented (like the first major town, quests, etc). Now you are ready to go to Kickstarter. Even if you don't need money to finish the game, Kickstarter is still a good idea for a number of reasons. The people who pledge to your Kickstarter are your first true stakeholders in the game, the seed that will hopefully grow into a large audience.

You can find a lot of info out there on how to run a successful Kickstarter so I won't go into that here, but you should plan carefully and dedicate 100% of your time to the Kickstarter while it is live.

Backer beta

Not long after the Kickstarter is done, launch a backer beta. This is the first time people are getting hands on the game, so the feedback will be coming fast and furious. You want to spend this time purely on bugfixing and community engagement. Dedicate a couple of weeks to this, then close the beta and incorporate all of the lessons learned. If for some reason there is some kind of major flaw in your game's design that you previously hadn't identified, now is the time to fix it.

You can do multiple backer betas if you need to, but the goal is to get the game's systems and early content to as close to a finished state as possible. By now you should also have a good estimate of how long it will take you to finish the game, so you know an approximate release date. But first you want to take advantage of Early Access.

Early Access

Set the EA release far enough in advance that you can get involved with media, streamers, etc and make a strong marketing push before it hits. This is your game's debut and first exposure in the wild, so you want it to make as much of an impact as possible.

Hopefully by now your game has a lot of steam (pardon the pun) behind it and the EA release hits. Ideally this is the point where you can afford to quit your day job and work on the game full-time. Make it clear on the Steam page where the game is in terms of content completion and how long it will reasonably take to finish.

You will need to spend more time fixing bugs and incorporating feedback, but hopefully a lot less than before since you already got at least one pass at this with the backer beta. People are more tolerant of issues in an Early Access release, but keep in mind from this point forward your game is being judged in the open marketplace. Everything you release from here should be as polished and production-ready as possible.

Finish the game

Now the goal is simple: finish the game. After each major content update, earmark time for bugfixing and increased community engagement. Try to make it so each update doesn't break people's saves if at all possible.

Once you can see the finish line for having all of the content implemented and polished, set a final release date. Make another marketing push like you did for the EA launch. Go as big and as wide as you can, because this is the climax of the entire development cycle and the best chance you will get to squeeze as much juice out of this peach as you can.

Release

Now you release! Do everything you can to promote the game and reach as many people as possible. Hopefully it goes well and now you're rich!

Post-release

Keep fixing bugs and releasing patches for as long as it makes sense to. Release on consoles, or make a deal with a publisher to port it for you. Start pre-production for the sequel.

tl;dr You just need a trailer to announce the game and start advertising, not a demo. Then it goes Announcement --> Kickstarter --> Backer beta --> Early Access --> Full release.

Thanks, sounds like a good plan. I guess it's just my own biases at work. 'Are you telling me people seriously get interested in a game before they can even buy it?'. I never follow upcoming games (apart from BG3 because I like sharing videos to my wife who's keen to try it).

 

[Zanzoken]

I think the stuff Zanzoken wrote is good if you go the Kickstarter/EA route but I never did do that. There are some contrarians who say that wishlists get stale if you take too long from the first announcement until the release. I simply don't know. Both routes can work but the announce and release and do everything in 2 months route is way harder if you didn't do it already.

I used to not be a fan of Kickstarter either, but after doing some more research on it I think the pros outweigh the cons.

Pros
- Steam will provide keys for every Kickstarter backer, free of charge. Given that Kickstarter's fees are 8 to 10% and Steam's fees are 30%, this is a big deal for your bottom line.
- Some people on Kickstarter will voluntarily pay more for your game than the asking price. Sometimes a lot more. This is essentially free money, just be careful not to get too crazy with rewards and all that.
- A backer beta gives you a captive audience to focus test and QA your game before it hits the public marketplace on Steam.
- Kickstarter itself is a promotional tool that will draw attention to your game.

Cons
- Steam reviews left by Kickstarter backers don't count toward your game's review score. This is a significant drawback considering your backers are the most invested in your game, and are likely to leave positive reviews.
- Launching and managing a Kickstarter requires a non-trivial amount of time and effort.

I don't pretend to have all of the answers either, of course. This stuff is not an exact science by any means.

Thanks, sounds like a good plan. I guess it's just my own biases at work. 'Are you telling me people seriously get interested in a game before they can even buy it?'. I never follow upcoming games (apart from BG3 because I like sharing videos to my wife who's keen to try it).

Well, a slight variant on the approach I described is you could hold off on announcing the game until you're ready to launch the Kickstarter, and then do both at the same time. If you wait to go to Kickstarter until you've got a playable demo / vertical slice, then you can go straight from the campaign into the backer beta if you want. Then you're selling something tangible, unlock most Kickstarters which are just promises.