Cazzeris
Guest
http://www.irontowerstudio.com/forum/index.php/topic,7187.0.html
It's mostly about the art costs, but it includes some bits regarding system design.
This must be the studio which made the plasma gun art: http://www.rmory.net/
It's mostly about the art costs, but it includes some bits regarding system design.
Vince said:We started working on AoD back in 2004 which was a long time ago, so long, in fact, that I’ve completely forgotten what it feels like to start from scratch, when you need everything but have nothing. To say that it’s overwhelming is to say nothing at all. If anything it was easier the first time around when our naïve belief that we can do it in two years shielded us from lasting mental harm. It’s not the same when you know that 4 years development cycle is the best case scenario, if everything goes smoothly.
So what does one need to make an RPG?
- An engine would be nice. Fortunately, it’s not my department so I’ll scratch it off my list and pretend that it will magically appear one day. So far, the plan is to use Unreal 4, but if we run into trouble we can always use Torque (the AoD engine), aka oddly comfortable plan B.
- Setting, story, locations, quests, characters – the writer's domain. If AoD’s 600,000 words are any indication, it’s a big project that would probably take at least 3 years out of 4, if not all 5 out of 4, but as long as I’m half a step ahead of Oscar, it’s all good.
Right now I’m working on the 'foundation': history, mutiny, early factions, post-mutiny events, schisms within factions, etc. Basically, the past events that explain why things are the way they are and how "we" got there in the first place. It’s important to see such things clearly when working on quests and leaders who don’t spring out of nowhere but are the product of their age and milieu. Overall, it’s been about hundred years since the mutiny and more than three hundred years since the ship was launched.
- Systems. While the exact details aren’t important yet as all systems will keep evolving throughout the development, gaining more depth and complexity with every iteration, we need to get the 'foundation' right the first time, then build on top of it. More on that later.
- Art. Surprisingly, the biggest problem right now is art. We need a lot of art, all kinds, shapes, and sizes. No, we don’t need it right now, but all these tasks take months and years, so we need to start planning now or we’re going to miss the train.
It feels like a Tetris game where you have to put all these different pieces together and make sure we get everything we need and in a timely manner. Mazin (our artist) is truly exceptional and there is very little that he can’t do well, but he’s working part-time. Even if he weren’t, he can’t do everything as the list is too long, which means that we need to hire freelancers.
That’s where it quickly gets prohibitively expensive. Quality takes time (days) and nobody will work for days for $25. Now we know that the game will still look crappy indie because we don’t have the manpower (the Witcher had 80 people, the Witcher 3 250 people, we have one artist/designer which explains why AoD didn’t snatch a single “Best Graphics 2015” award), so if we can’t deliver awesome visuals, we HAVE to deliver awesome details like portraits, icons, intro art, models, etc.
So what do we need?
- Logo
- Interface
- Intro, Menu, Promo art. Each piece takes 3-4 weeks, so that alone is a year-long project IF the artist is committed and dedicated. Sadly, that’s a big IF so it might take 2 years IF it goes well. When it comes to freelance artists, your mileage varies greatly.
- Visible Things That Must Be Designed:
- Weapons
- Mechs & Turrets
- Anti-Riot Droid aka your faithful companion
- Shuttles
- Mutated critters
- Space Suit and other trendy clothing items
- Gadget effects when deployed
- Non-generic locations. For example, the cargo hold is visually interesting (looks amazing in Excel!), but it's built from generic parts (containers, cranes, etc). A place like the Bridge, for example, or the Neo-Church, or the Breached Hull :have spacesuit – will travel: requires a concept artist’s touch (weeks of work per location).
- Misc; ideally a lot more things should pass through the Concept Art department, but we have to be realistic and never ever bit more than we can chew on.
- Portraits. AoD has 57 portraits. Let’s say we’ll go with 60 portraits because they really do add so much to the dialogue screen. 3-4 days per portrait, that’s 180-240 days right there.
- Inventory Icons. AoD has 252 unique items (not counting different variations like bronze, iron, steel, etc using the same models and icons). We don’t have a full item list yet but I’d be surprised if we end up with less than 200 items, so we’ll have to split the items between two artists to make sure it’s done by 2018.
So far we've finalized weapons, armor, and gadgets. We started with 81-90 weapon icons but reduced them to 66 for cost-related reasons, which is still more than AoD’s 40 weapons plus uniques. It takes 3-4 days per icon (including the back-n-forth design phase), so 6-8 months of work with a dedicated artist, about 2 years of work with a busy artist who’s working with multiple clients, and that’s just the weapons.
This brings us to the issue of costs and art budgets. Any work, be it a portrait or an inventory icon, that takes a couple of days of work would end up costing you $100-300 per item. For example, when I was looking for a portrait artist for AoD, I was quoted $80-100 for a black-n-white portrait, $200-250 for a color one, which explains why you rarely see high quality portraits in games. It's just too expensive. If not for Mazin's generosity, we wouldn't be able to afford them either.
So 66 weapon icons would cost us anywhere from $6,600 – if we’re lucky and find a guy willing to do the top tier work for the low tier price – to almost 20k, which is a lot of money. It's very easy to understand how game budgets quickly grew out of proportions and to the point where you have to play it safe to get your investment back and make a few bucks on top of it.
It’s tempting to invest more, to hire more artists, but I know of three indie studios that either ended up deep in debt or failed to make any money to continue after releasing RPGs that seemingly did well and I’m not in a hurry to join them. I'll go over the indie "business model" in the next AoD update.
- Models (weapons, objects, etc) : Same story, there’s only so much we can do on our own because modeling guns and objects take time from Oscar and Ivan (our animator). So again, you outsource - you pay a lot and drain your coffers faster than you can blink, so it requires an approach so careful and balanced it would make Sawyer proud.
- Animations. AoD has over 500 animations, but that’s mostly melee. Firearms and cover mechanics can easily double that number.
^ It's not the final name but it's one of the "finalists". My first choices was Adrift but it's already taken (a Steam space game creatively called Adr1ft). The Sunless World isn't the best name but it works. The ship IS an entire world to its inhabitants and 'sunless' is a fairly good and accurate description. There are some books with the same name (one's from 1967, the other is an e-book) but I don't think that anyone will claim we're trying to capitalize on their fame, although you never know. Anyway, until we launch a website, nothing is final.
Back to the systems. Now that you have a good idea of our workload, let's talk about two key systems: combat and stealth. I'll do my best to organize my thoughts and present them in a semi-coherent fashion, so bear with me.
Melee-based combat makes a lot of things simple in 3 key areas:
So, logically, combat with guns should have higher THC in general, higher mobility, no dodging bullets, no damage modifiers, which means that fast attacks, normal attacks, and aimed attacks will do exactly the same damage, which means the player would want to use the fastest attack unless there’s a strong penalty, but we’ve just decided to keep THC relatively high.
- Melee attack can be dodged or blocked (or parried or deflected), thus you easily miss even if you’re standing next to the guy you’re trying to kill. The famous duel in Rob Roy would have been very different if both duelists had SMGs.
- Damage dealt can be modified by effort (i.e. fast or power attacks).
- Combatants can just stand there and trade blows all day.
The obvious conclusion is that we need grazing, cover, and a wide range of attack types:
Another thing worth mentioning is that firearms, unlike fairly straightforward melee weapons, have very different designs, so each category (Pistols, Shotguns, SMGs) will have 3 subcategories. For example, Shotguns will have One-Handed Shotguns (sawed off and ‘Mare’s Leg’ style shotguns), long, heavy-barrel shotguns, and shotguns with revolving cylinders. So even if you choose to specialize with a single weapon class, you will have plenty of tactical options within this class.
- Let’s start with grazing. I wanted to implement it in AoD but we were out of time and the fast attacks were basically grazing attacks, doing a lot less damage. The ranged combat is perfect for it.
Let’s say you have 80% THC (to-hit chance). You roll the dice and as luck would have it, you’re 1 point short but the binary miss-hit system doesn’t reward your near excellence and treats it as you weren’t even close. So, we’ll change that and go with 4 roll 'ranges': miss, graze, hit, critical hit.
This will give us some flexibility with damage ranges and allow you to trade damage for THC.
- Cover is another way to lower your THC without raising eyebrows and explain why the combatants take 4-5 turns to kill each other. We don’t want to place cover objects everywhere, so we’ll go with energy shields you can place in front of your character (i.e. you throw a 'gadget', it generates an energy shield in front of you (not around you) which absorbs X amount of damage and makes it harder to hit you while you’re hiding behind it).
Naturally, once you decide to go with gadgets, why stop with one? Why not have gadgets modifying every available battlefield stat?
- Depletable energy shield (absorbs x damage)
- Reality distortion field (THC penalty against you)
- Optical illusion a-la Total Recall (chance that enemies will target the illusion)
- Cloaking field aka Stealth Boy
- Stasis field (holds enemy, no damage can be dealt)
- Brainwave Disruptor (don’t leave your home without Psychic Nullifier)
As mentioned previously, expect 10-12 gadgets with 3-4 upgrade levels. Earlier I was toying with the idea of energy armor but the energy shield idea is better as it ties you down, creating tactical opportunities for your enemies. Then you’ll have upgradable synthetic armor offering different degree of protection against melee, projectile, and energy attacks.
- Attack types
- Three basic attacks are Fast aka unaimed (increases your Graze roll range and cuts the Hit roll range in half), Normal, and Aimed Attack (doubles the Hit roll range, cuts in half the Graze range). You can use these attacks with any weapons.
- Class- and weapon-specific attacks like Short, Long, and Wide Burst for SMGs (some SMGs would be more suitable for wide bursts whereas others would be more suitable for long burst; however, these attacks would be available to all SMGs) and certain shotguns; Fanning for revolvers, Double Shot or Full Broadside for multi-barrel weapons.
- "Tactical" attacks like Suppressive Fire and various Interrupt, Attacks of Opportunity Reaction Shots
Last but not the least is the focus on mobility. In AoD we didn’t want a melee opponent chasing you all over the map, so your movement rate was slow (2AP per square). In the CSG we want mobility play a large role, allowing the enemies (and encouraging the player) to move around, flank and flash you out.
TB Stealth feature list:
Well, that's about it, folks. If you have questions, ask. If not, wait 'til the next update. Only 55 updates to go.
- Your actions will generate 'disturbance' (the word noise doesn't work with visual disturbance caused by cloaking), modified by your stealth skill and the guards’ perception. The disturbance meter will show both the disturbance you generate while in stealth mode and disturbance you detect (cloaked or hidden enemies).
- You determine how fast you move while sneaking (2:1, 1:1, 1:2 – AP to square ratio), the faster you go, the more noise you generate.
- If you manage to get to a guard from behind without alarming him and triggering an interrupt attack (like a shotgun blast in your face), you can kill him using Melee weapons, your skill level will determine whether you succeed or fail. If you fail, we switch to TB combat.
- The guards will have 3 states; each state will increase the detection radius, representing alertness.
- unaware of your presence – do nothing
- suspicious – investigating
- alarmed – hunting down and summoning reinforcement, will get an interrupt attack on sight
This must be the studio which made the plasma gun art: http://www.rmory.net/
