[Madoc Evans]

Hello,

I'm the lead developer behind the games Exanima and Sui Generis. The games have a low fantasy setting, but are strongly inspired by Medieval Western Europe. We have a very unique and innovative physics based animation and combat system, which has attracted several historical martial arts experts who have influenced our designs quite a lot. Now we try to maintain certain standards of realism and accuracy, even though it is not in fact an historical setting.

I'm hoping that I'll be able to pick your brains on various topics here from time to time as I've often found myself consulting these forums and found them very useful and interesting.

Please note that the game is still in development (though already released in Early Access and in a reasonably complete state) and we're always striving to improve the combat and animations, we do however prioritise the gameplay and its very dynamic nature over realism. This sometimes requires some very "stupid" combat behaviour, but we're still trying to find a middle ground. If you want to see an example of what combat currently looks like in action, take a look at this video:

https://youtu.be/6Kq4lf4W-fE

Here's a screenshot of a couple of in game characters to give you an idea of the graphical style:




Currently we're working on a fleshed out arena game mode. The name says most of it really, you are able to participate in various fights with a variety of combatants and equipment. For gameplay mechanics we need to introdude various "rank" restrictions on equipment, dividing them into 6 tiers. Combatants are similarly ranked based on their competence. At the lowest ranks there will be pretty much peasants in ordinary clothes with crude weapons, at the highest you'll see combatants in full plate armour with poleaxes etc. Rank is not necessarily tied to monetary cost, but rather some estimated effectiveness. Circumstantial factors are a matter of choice.

I realise this is a very artificial way of thinking about things, but such are games, they sometimes require rather silly mechanics. I would particularly appreciate any ideas about what weapons we could put in each rank. We're not very strict with time periods, so far we've included things from 12th to 16th century and some original designs. I'd be more than happy to provide more information if requested, but this post seems to already be getting rather long!

Thanks in advance,

Madoc

[J. Nicolaysen]

Sounds interesting!

[Madoc Evans]

Thanks, definitely some useful ideas and designs there.

Balance is actually one of the statistics we put on weapons. We're looking to overhaul our current weapon system with one that uses procedural elements so that every weapon is essentially unique. They will however be based on many quite strict templates (for the same basic weapon) giving visual consistency and distinct categories. Thing is we're not really sure how to say that one category of sword is worse than another but can still be better or worse than an equivalent one. We want as many types and sub-types as possible, but without someone who knows their stuff screaming outrage.

It might be useful to show some of our current weapons and give a bit more info on how things work. Here's a few weapons:



http://dl.baremettle.com/XAWepnEx02.jpg

The stats we give each weapon are weight (relative), balance, impact, slash, crush, pierce and thrust, as applicable. Balance and weight affect the responsiveness of weapons, how quickly you can bring them to bear for a strike or a parry basically, and how quickly you can accelerate them. The damage stats interact with the physics simulation a lot, the strength and type of collision, and of course armour.

Armour has similar stats of impact, slash, crush, pierce and also coverage and encumbrance. Coverage is just a relative estimate as with armour what you see is what you get in terms of coverage. We already have a procedural system in place for both the appearance and quality of armour. Might as well show some examples of armour too (these have no quality modifiers applied):



http://dl.baremettle.com/XAArmrEx02.jpg

As you can see the stats are quite approximate in their representation, we tried to keep things within realistic ranges and give them realistic effects. The physics simulation plays a huge role in how things actually work. Armour is extremely effective in the game, high coverage full plate makes you virtually impervious to wounding, though you can be gradually beaten down through a sort of endurance system.

We do also have shields. Ranged weapons haven't been added yet, but they will be.

Thanks again. Really appreciate any suggestions as we're often sumbling in the dark, it's hard to know when a source is legitimate or how to gather relevant information.

[J. Nicolaysen]

Can you scale the stats of the player so they get better at the same weapon or armor as they progress? So someone who just got a nifty set of plate armor can't move as well as someone who has done a few rounds with it. But maybe it isn't that kind of game.

Maybe expand your "finely crafted" weapons and armor and have a higher grade of weapon that looks similar to the lower grade. So a high grade weapon might have a bronze pommel and a blue hilt or something showy like that. Plate could have some design on it at the later levels to show it is better made, less encumbrant (sp?) and better steel. That sort of thing.

I don't know that it's really fair to say this type of sword is better than that type of sword in your type of game. If you have a guy in plate fighting a guy in gambeson, the type of sword won't matter as much. But you could use some of the traditional sword typology to use a Type X and a Type XVa and a Type XVIIIb and if it isn't matched well against the right armor, it suffers in use. Does that make sense?

Also, you can do a lot with polearms I think. Use your traditional rough spear or billhook or flail for entry level, but have a sweet patternwelded spear, or a boar spear or a nice halberd or poleaxe for the higher levels. Things that look better made, but also connote higher level of skill and training to use, in other words, is my general idea for all of this.

And shields...Match a shield with the right type of weapon for it and have a much better efficiency/defence/offense thing. Like a buckler with a type XII or a large round with a type X or a targe with a basket-hilt. Just throwing out ideas. I never had enough time for games, but I can see some neat possibilities here that other games haven't done.

[Tyler Jordan]

people will always trend towards whatever gives the most bonuses. One way around that would be set bonuses. Make equipment sets as put-together themed packages, and if you use it just as is, you get a bonus that would offset mix and match minmaxing. Under-performing weapons may get a bonus when combined with sub-optimal armor, and such. One of the biggest challenges in a game is getting people to use equipment that isn't 'the best for my level', Especially when you have a lot of different equipment, inevitably some of it will go utterly un-used. It's a challenge to design equipment so there isn't an objectively best/worst choice.

[Michael Beeching]

I think the ideal of "best" or "worst" weapon tends to be skewed by artificial parameters, even in reality. I recall a recent TV show, "Forged in Fire," where they had blacksmiths compete with each other to see who could make the best quality weapons with very tight time and design constraints. To test the quality of the weapons, they would of course, test the weapons! Unfortunately, a lot of these tests were biased towards cutting, or some other particular physical motion which does not encompass the whole utility of a given weapon. Thus, you have an artificial parameter in reality - in theory, someone could have made a perfect dagger in that show, but it would have never won, because daggers are not particularly strong chopping tools.

I think you could mitigate some of the "artificial parameters" in your engine if you instituted more in the way of grappling. For instance, I cannot safely fight a swordsman with a dagger from a distance (unless I run away... in which case, I did not even fight!), so I have to close with him. After I do that, I can try and strike him directly, or more likely, I will try and disarm him and throw him to the ground, In this position, it is possible that my humble dagger is now just as potent as his sword was!

Another thing to consider is alternating the use of weapons - swords are a perfect candidate for this. If fighting in armor against another armored opponent, I may want to half-sword, because using the sword as against an unarmored man would do little if nothing against someone wearing heavy armor. In this mode, I'd be doing thrusts and grappling primarily. If I wanted to strike, it wouldn't be a bad idea to allow murder strikes, or rather, holding the blade of the sword in hand and hiting the enemy with the pommel and crossguard, etc.

...So, before just throwing any more suggestions out into the breeze, I think the best thing to do to prevent "best" and "worst" weapons is to mitigate the chances for game mechanics to be skewed by artificial parameters. Easier said than done, of course, but this probably entails diverging slightly from the current physics engine which relies on powerful strikes and footwork to do the most damage. So long as that is the deciding factor in getting the job done in-game, you already know which weapons are best. If that is not satisfactory, then, again, close-in fighting mechanics and multiple weapons-use options may change that. Frankly, I'd love to see armor being used against its own wearer - essentially, in armored grapping, you may come into a situation where you may be able to throw someone (throwing as akin to judo, mind you), hold their arm in a lock, and break it using their own weight against them as the fall.

[Madoc Evans]

There is character advancement in the form of trainable skill techniques, but they don't generally affect statistics. Close combat techniques include things like Riposte, Remise, Fend and Feint, which basically allow instantly transitioning and chaining of attacks and parries etc. Armour skills do include ranks of "Manoeuvering in Armour" which reduce the effects of encumbrance.

Stats on weapons are really not very important, and players pick up on this pretty quickly. Because of the accurate physics, the shape of the weapon can be more important than its stats, and each one plays very differently. Players develop distinct playstyles and preferences in weapons, there are unlimited ways to play the game and people tend to do whatever works best for them. Two handed swords are very popular, especically amongst new players, because of their balance and reach and the fact that the entire length of the weapon is effective against lightly armoured opponents even without much force. Other weapons can have distinct advantages though, some weapons can quite easily throw or pull people to the ground if used correctly, or are more effective against armour, shields are also used a fair bit with a variety of weapons and personal preference is definitely a factor.

Footwork and body movement are indeed crucial to landing powerful blows, but this is by no means the only way to play. You can focus more on defence and timing, you can try to outmanoeuvre your opponents in various ways. Against weapons with long reach for example it's very viable to get inside someone's range and actually stop their swing with your body. We're also introducing further manoeuvres such as thrusting with any weapon and bashes / shoves.

We can dynamically switch to a half-swording grip and we'd like to try some things with using the same weapons in different ways. Grappling is very difficult to do without precanned animations and they don't fit the kind of fine grained control we want over every action, we avoid the typical press-one-button actions. We are also introducing the ability to manually pull your weapon back from a swing to pull people down with polearms or do draw cuts etc.

Either way we'd like to provide combatants with a good variety of weapons at any rank, ones that support different fighting styles. We're currently thinking of this mostly as the progression from say a cudgel to a flanged mace, a bill to a poleaxe, a rusty knife/dagger to a nice arming sword. It's quite hard however to imagine what intermediate steps might be and how to introduce new weapons or qualities gradually across so many ranks meaningfully.

[Bram Verbeek]

The game sounds very nice, I'll surely pick it up. From the video's, it did look like it was very cut centric, are thrusts not yet implemented, or did I miss them?

[Madoc Evans]

Thrusts are mostly implemented but we haven't put them in release builds yet as we're still working out some details of the targeting system. The idea with thrusting is that you can target accurately in 3D space and this is a little tricky when your thrusts are more predictive rather than having a real target. We'll add them soon, but we're still working on a lot of different aspects of the game so our priorities might not always make sense from a player's perspective.

[Bram Verbeek]

Nice! That it is not yet shown does not mean a lot, it is in development. When I have time, I will look it up, probably buy into both games. I have had my fun with mount&blade, but this seems a lot more involved.

[Madoc Evans]

Yes, involved is right. Be warned that this is not an easy game to pick up. Animations are not precanned, but a result of your inputs, physics and your character's virtual muscles. It takes coordinated and well synchronised inputs to control your character smoothly and with intent. This is very unusual for a game and new players sometimes don't understand why their character ends up stumbling around and flailing weakly. The controls are simple but quite unique and they do take muscle memory and therefore practice and experimentation. May not be everyone's cup of tea. Usually though people do have fun even while they're getting their arses handed to them. There's certainly something very satisfying about landing that devastating blow when it's you that's in control rather than the game doing it for you.

[Lafayette C Curtis]

Seems like you got a pretty good thing going on here. Your characters tend to emphasise cuts rather than thrusts and wide swings rather than tighter, more precise ones, but these are perfectly acceptable compromises for the video game medium since the tighter and more realistic movements of actual medieval sword techniques actually look pretty boring (and I'm speaking as somebody who studies them too -- they're nice to practice and spar with but not really that good to watch when you're not actually involved in the fight). Have you read Guy Windsor's book on this very subject? He -- obviously a much more experienced practitioner than I am -- has an anecdote where he was invited to demonstrate some medieval longsword techniques from Fiore de'i Liberi and everyone found that the techniques were too small and subtle to be effectively or convincingly represented in video game animations. So too much realism can be a bad thing!

http://guywindsor.net/blog/books/swordfightin...l-artists/