Slaves of Magic - tactical turn-based RPG inspired by XCOM and Guild Wars

[Taka-Haradin puolipeikko]

Tried the demo for a bit.
For some reason mouse is incredibly laggy in tactical layer, makes demo almost unplayable.

 

[sipibaki]

Tried the demo for a bit.
For some reason mouse is incredibly laggy in tactical layer, makes demo almost unplayable.

Sorry about that! Could you post your system specification, operating system, and resolution on which you played it? One thing that quickly comes to mind is that if you have multiple video cards in your system, it is possible that it defaulted on the weaker one.

Edit: Has been solved on the Steam forums, and my guess was right, it defaulted to the weaker card. In the long run, I will try to find a way to force Windows to choose the stronger one, but for now, you can solve this issue by manually setting which video card to use.

 

[Taka-Haradin puolipeikko]

Combat is now incredibly deadly.

And game likes to spawn enemy reinforcements right on top players squad.

 

[sipibaki]

Combat is now incredibly deadly.

And game likes to spawn enemy reinforcements right on top players squad.

We have released a new patch with some bug fixes and balance tweaks. Shields were bugged and did not provide their innate block chance so those units should be more durable now.

On the balance front, I reduced the enemy count slightly and gave more time till the first reinforcement arrive. Should be more manageable now.

The combat is indeed deadly for units with no armor. But shields(now that they work) and armor are very effective in making units survive multiple hits. On the reinforcement front, it is true that it spawns close to the player. That is intended. The player has 1 full round to prepare for their arrival with the knowledge of their exact arrival position.

 

[Camel]

The Steam TactiCon event has started, and our new demo is out and you can check it out on our Steam page.

Good luck with the game!

 

[Jrpgfan]

The portraits are Jeff Vogel level(hopefully they're placeholder) but I actually liked the aesthetics.


I hope it turns out to be good.

 

[Taka-Haradin puolipeikko]

Combat is now incredibly deadly.

And game likes to spawn enemy reinforcements right on top players squad.

We have released a new patch with some bug fixes and balance tweaks. Shields were bugged and did not provide their innate block chance so those units should be more durable now.

On the balance front, I reduced the enemy count slightly and gave more time till the first reinforcement arrive. Should be more manageable now.

The combat is indeed deadly for units with no armor. But shields(now that they work) and armor are very effective in making units survive multiple hits. On the reinforcement front, it is true that it spawns close to the player. That is intended. The player has 1 full round to prepare for their arrival with the knowledge of their exact arrival position.

Yeah.
It's way more manageable now.

 

[Joined on the 4th of July]

and my guess was right, it defaulted to the weaker card. In the long run, I will try to find a way to force Windows to choose the stronger one, but for now, you can solve this issue by manually setting which video card to use.

Was this your own engine or did you at least have access to engine source code? These used to work earlier, but not sure if that's still the case

#ifdef _WIN32
extern "C" {
__declspec(dllexport) DWORD NvOptimusEnablement = 0x00000001;
__declspec(dllexport) int AmdPowerXpressRequestHighPerformance = 1;
}
#endif

 

[sipibaki]

and my guess was right, it defaulted to the weaker card. In the long run, I will try to find a way to force Windows to choose the stronger one, but for now, you can solve this issue by manually setting which video card to use.

Was this your own engine or did you at least have access to engine source code? These used to work earlier, but not sure if that's still the case

#ifdef _WIN32
extern "C" {
__declspec(dllexport) DWORD NvOptimusEnablement = 0x00000001;
__declspec(dllexport) int AmdPowerXpressRequestHighPerformance = 1;
}
#endif

It is my own engine, and funny that you posted this as I just put this solution into the engine today. It will be included in the next patch and hopefully, this still works.

 

[Darth Canoli]

Sorry about that! Could you post your system specification, operating system, and resolution on which you played it? One thing that quickly comes to mind is that if you have multiple video cards in your system, it is possible that it defaulted on the weaker one.

Edit: Has been solved on the Steam forums, and my guess was right, it defaulted to the weaker card. In the long run, I will try to find a way to force Windows to choose the stronger one, but for now, you can solve this issue by manually setting which video card to use.

Does that mean your game will run incredibly slow on integrated video chips?
It certainly doesn't look like it should.
I don't mind this kind of graphics as long as the gameplay is incredibly fast because the engine is highly optimized.

 

[sipibaki]

Sorry about that! Could you post your system specification, operating system, and resolution on which you played it? One thing that quickly comes to mind is that if you have multiple video cards in your system, it is possible that it defaulted on the weaker one.

Edit: Has been solved on the Steam forums, and my guess was right, it defaulted to the weaker card. In the long run, I will try to find a way to force Windows to choose the stronger one, but for now, you can solve this issue by manually setting which video card to use.

Does that mean your game will run incredibly slow on integrated video chips?
It certainly doesn't look like it should.
I don't mind this kind of graphics as long as the gameplay is incredibly fast because the engine is highly optimized.

Well, higher fidelity does mean increased system requirements compared to our old build. There are 2 component that contributed to this.

- High-resolution animated sprites. Because of the way video cards work they store texture without compression (okay, there is a kind of compression that works on video cards but the result was very ugly when we tried it out). So that increased our video RAM usage. The game will have a minimum requirement of 2 GB of video ram because of this. But this limitation is basically the same as far as I saw for games using similar HD sprites (for example project Zomboid, or the age of empire HD remake), so there is not much to be done here except drastically reducing our sprite sizes.

- The other cause of higher work on the video card is the per-pixel dynamic light calculation. Probably there is some optimization that is still possible there, but it is not hard to see why that would increase the workload significantly on the video card compared to before.

Still, the weakest integrated video card we tried out the build was intel HD graphics 630. It could run the game on 30 FPS in 1920x1080 resolution (and lowering the resolution reduces the dynamic light workload by a lot). So relatively new integrated cards can still play it just fine and anything better than that with at least 2 GB will have no problem running the game with 60+ fps. As I said, there might be some optimization headroom here which could improve it a bit, but realistically the game will require some sort of a video card, no way around it with dynamic light calculations.

One thing I was thinking about to still support people with very low-end systems is to let them turn off dynamic lights and force every mission to be sunny. This could work because as for now lights have no gameplay relevance. Of course, if I ever give gameplay relevance to it that would cause problems.

 

[Darth Canoli]

One thing I was thinking about to still support people with very low-end systems is to let them turn off dynamic lights and force every mission to be sunny. This could work because as for now lights have no gameplay relevance. Of course, if I ever give gameplay relevance to it that would cause problems.

Sounds pretty good to me, even when it doesn't affect performances like in M&M 6, I've always hated weather effects, in M&M 6, for example, snow makes the pixel hunting (exploration) part quite tedious.

Being able to disable dynamic lights and weather effects should be a default option for every game featuring it.

And if you ever implement gameplay effects, just add a weather icon somewhere and link the gameplay effects to that icon instead of the weather effect itself, which could be disabled at will.

 

[sipibaki]

One thing I was thinking about to still support people with very low-end systems is to let them turn off dynamic lights and force every mission to be sunny. This could work because as for now lights have no gameplay relevance. Of course, if I ever give gameplay relevance to it that would cause problems.

Sounds pretty good to me, even when it doesn't affect performances like in M&M 6, I've always hated weather effects, in M&M 6, for example, snow makes the pixel hunting (exploration) part quite tedious.

Being able to disable dynamic lights and weather effects should be a default option for every game featuring it.

And if you ever implement gameplay effects, just add a weather icon somewhere and link the gameplay effects to that icon instead of the weather effect itself, which could be disabled at will.

Actually, another idea that came to me is that making it per tile lighting should be relatively easy, only calculate the light for the middle point of a tile, then use that color for the whole tile. So in practice, it would work like the light on the original XCOM. That is a lot less taxing on the video cards plus that keeps the lighting information if I ever want to give gameplay relevance to it. In addition, no need to remove nighttime missions in this case, as the light sources still lit up the map.

While the dynamic lights look a lot better in my opinion, for people with low-end systems this could make the game playable. And it shouldn't be too hard to provide both options for the player.

 

[Darth Canoli]

Actually, another idea that came to me is that making it per tile lighting should be relatively easy, only calculate the light for the middle point of a tile, then use that color for the whole tile. So in practice, it would work like the light on the original XCOM. That is a lot less taxing on the video cards plus that keeps the lighting information if I ever want to give gameplay relevance to it. In addition, no need to remove nighttime missions in this case, as the light sources still lit up the map.

While the dynamic lights look a lot better in my opinion, for people with low-end systems this could make the game playable. And it shouldn't be too hard to provide both options for the player.

You'll get no argument from me by taking features from UFO, it's the best tactical game ever created.

 

[udm]

This could work because as for now lights have no gameplay relevance. Of course, if I ever give gameplay relevance to it that would cause problems.

Please say lights will eventually have gameplay relevance!

 

[sipibaki]

This could work because as for now lights have no gameplay relevance. Of course, if I ever give gameplay relevance to it that would cause problems.

Please say lights will eventually have gameplay relevance!

As I said, I want to leave the option open to make it relevant at some point, but it is not exactly high on the priority list so to speak.

 

[Taka-Haradin puolipeikko]

 

[udm]

So I've been playing the demo and enjoying it. Was a bit disappointed to find that it uses 2AP system rather than granular APs, but I imagine it'd break a lot of the game if you were to convert it.

Will post more impressions after playing it some more.

Also, will there be indoor battles? The system seems like the perfect match for tight CQB.

 

[sipibaki]

So I've been playing the demo and enjoying it. Was a bit disappointed to find that it uses 2AP system rather than granular APs, but I imagine it'd break a lot of the game if you were to convert it.

Will post more impressions after playing it some more.

Also, will there be indoor battles? The system seems like the perfect match for tight CQB.

Glad you are enjoying it! Yes, the 2 AP systems will remain at this point. But I will try to make up for it with the number of skills letting you circumvent in some ways.

Most of the normal missions will be outdoor as you see in the demo. There are big buildings that can be generated like the church though which can create tight CQB if the fights take place there. But the basic idea is that most of the time the player will be able to teleport their units at the start of the game in such a way that they will be in combat or at least very close proximity to some enemies.

Special missions like taking out an enemy stronghold probably will be fully indoor though, so occasionally there will be completely indoor battles.

 

[Galdred]

The portraits are Jeff Vogel level(hopefully they're placeholder) but I actually liked the aesthetics.


I hope it turns out to be good.

Have you thought about using midjourney, Dall-e2 or stable diffusion for the portraits? From my epxerience, they can end up costing a lot if you plan to have a large roster, and AI seems to be relatively competent to draw them, especially in low definition.

 

[sipibaki]

The portraits are Jeff Vogel level(hopefully they're placeholder) but I actually liked the aesthetics.


I hope it turns out to be good.

Have you thought about using midjourney, Dall-e2 or stable diffusion for the portraits? From my epxerience, they can end up costing a lot if you plan to have a large roster, and AI seems to be relatively competent to draw them, especially in low definition.

That's not a bad idea in itself, but my girlfriend (the graphic designer) is actually very excited about the portrait generation part and wants to do it herself. I just really haven't given her the time to properly get into it yet. It will be better than the current ones I promise.

 

[Galdred]

The portraits are Jeff Vogel level(hopefully they're placeholder) but I actually liked the aesthetics.


I hope it turns out to be good.

Have you thought about using midjourney, Dall-e2 or stable diffusion for the portraits? From my epxerience, they can end up costing a lot if you plan to have a large roster, and AI seems to be relatively competent to draw them, especially in low definition.

That's not a bad idea in itself, but my girlfriend (the graphic designer) is actually very excited about the portrait generation part and wants to do it herself. I just really haven't given her the time to properly get into it yet. It will be better than the current ones I promise.

Generating them (from different parts?) is very tricky, though, unless you only add accessories. I tried to do it that way in my game, but the artists rebelled.

 

[sipibaki]

The portraits are Jeff Vogel level(hopefully they're placeholder) but I actually liked the aesthetics.


I hope it turns out to be good.

Have you thought about using midjourney, Dall-e2 or stable diffusion for the portraits? From my epxerience, they can end up costing a lot if you plan to have a large roster, and AI seems to be relatively competent to draw them, especially in low definition.

That's not a bad idea in itself, but my girlfriend (the graphic designer) is actually very excited about the portrait generation part and wants to do it herself. I just really haven't given her the time to properly get into it yet. It will be better than the current ones I promise.

Generating them (from different parts?) is very tricky, though, unless you only add accessories. I tried to do it that way in my game, but the artists rebelled.

Well, the will is there, so we will see if we can handle it. But that is exactly the reason we have not yet done it because it will take a considerable amount of time to test out what is feasible for us, and figuring out the gameplay came first. But we will get there.

On another note, the discussion about making an integrated video card-friendly lighting model might be a moot point. I was messing around with the shader, planning how to implement it when found a pretty big bottleneck. After restructuring that part of the shader our integrated intel HD graphics 630 manages to do 60 FPS on HD resolution (previously 30) so I gravely underestimated how much optimization headroom was there. That improvement should be enough to push most weaker integrated graphic cards into the 30 FPS range. The next patch for the demo will include this change.

 

[sipibaki]

For those of you who have a lot of time on your hands, there was a live interview + let's play of the demo done by kormakrtv with me on the Tacticon Steam event, and it has been uploaded to Youtube. It was the first time I did something like this, but it was fun!


Besides that, we are working on the multiple-base system I wanted to implement for quite some time now, there is a very early look at how we imagine it on Twitter:

 

[sipibaki]