Where can I buy AAA prefabs..

I'm looking for AAA quality military buildings (with interiors), military base prefabs, environments and vehicles for a new FPS multiplayer game..

Can anyone point me in the right direction? I've searched the asset store but can't find anything really..

Thanks.

 

Turbosquid.

 

Manufactura K4's about the highest quality I've seen on the Unity Asset Store.

https://www.assetstore.unity3d.com/en/#!/publisher/585

 

Your skill with lighting and post effects.

 

You have a few options:

1) Make the assets yourself. (Your cost here is your time, experience and licensing 3d modelling software)
2) Pay someone else to make the assets for you. (Your cost here is a lot of money and time in finding the right person)
3) Purchase content from the Asset store or any other model sharing store (Your cost here is money, time in finding the right asset for your project and that your game will be using content that other games may also use)

Which option are you interested in?

 

You can't. AAA assets are generally custom. Any asset sold on the store or some other site is generic, and will end up on a dozen games.

 

Yeah, after a LOT of digging around on the asset store I'm fairly convinced that art on there can't make for a complete game.

And Turbosquid is WAY over priced.

Like, absurdly laughably over priced.

If the assets on there were your own, then MAYBE the price is right. But it's not. If you buy off there, Joe-Schmuckatelie can grab it too. And if your game has any success you're giving up all the brand establishment.

Your best bet is hiring an artist.

Post the scope that you're looking to do in job postings. Have a list of assets that you want. Then ask for a bottom line. Whatever you do, don't fall for the whole "I charge (cash amount) per hour" shtick. There's no way to verify if an artist is on the level with you and could easily inflate those hours to no end.

EDIT: Also, why do you want to do a AAA military shooter for (what I'm assuming is) your first Unity game? Wouldn't you rather do something small first to get an idea of all the potential hic-ups?

 

Not_Sure why you feel Turbosquid is overpriced, if you have any examples or .. what would be more correct pricing.

 

Well, it really comes down to the licensing.

Let's use a hypothetical soldier as an example.

A well crafted soldier would take a good artist ~10 to ~15 hours. At $20 an hour, you're looking at ~$200 to $300. That's about what they charge on turbo. Sounds about right, right?

Nope.

For one, for the same cost of getting a prefab you can get a custom tailored model that is exclusive to your project.

But there's more to it than that.

If you pay for it to be made, that asset is yours to do what you want with it. That means using it in follow up games. That means that if your game is successful, you don't have to worry about the asset getting cranked up in price. That means that no one can snag the same model that you used to ride your coat-tails. And that means that it is now your IP and the IP has a value of its own.

Think of how much Mario is worth right now. Imagine if they just snagged that off of some site for a one time licensing site.

To answer your question though, I would only buy the art assets in a bundle to assure completeness and it would need to be for at least half what it would cost for custom models. But more reasonably, about 1/4th. So that ~$200 to ~$300 model, I'd expect to pay more like ~$50 to ~$75.

After all, that money should not correlate to direct compensation to the artist's hourly costs, but spread over multiple purchases plus discounted for its generic-ness and legal limitations.

 

For an asset that takes 15 hours (aka 2 days) to complete, I would agree, $200 would be way overpriced.
But I think your maybe underestimating the amount of hours.

Speaking from experience, I think 2 - 4 weeks would be a more realistic time estimate for a non "hero" character, well crafted. 10k poly and up, including color and normal.

 

Really??

I admit that I am going by some limited knowledge of 3d modeling, and my guess-timates are based off of watching speed modeling videos on youtube.

 

Yeah, I think the modeling, or the 3D concept, could be done that quick,
but not the optimizing UV, baking and texturing.
Also, the amount, as in number, of attached props (aka clothing) is a huge influence on time.

 

Yes right.
But the thing is, you sell it to a lot customers. So ultimately you make more with that trough the quantity of your model.
If you base your price on the pure work hours, what's the point of these models? I mean for that price I could get a custom asset then.

 

No no,
I would, personally, charge a lot more than $200 for a custom asset of the same quality (as the models on turbosquid for $200).

A lot of the price your paying for an experienced artist to deliver AAA quality is not the amount of hours spent on the asset, but the amount of hours it has taken to reach AAA quality.

I don't delve too much inside assets for games specifically, but I can tell as much that its not uncommon to bill that amount per hour for animation related content.

 

Just to make that clear, I agree with your overall sentiment Kemonono. You shouldn't undersell your crafting skills. And that is perfectly fine for a freelance job.

But if you sell your model non-exclusively in a shop, you shouldn't ask the same price, you usually charge for a custom one.
Besides that, in my experience with the AssetStore, you sell way more when it's cheaper and that eventually cashes more in than selling the same model for a high price exclusively.

 

Agreed. Of course mileage will vary based on the artist, required quality, style, detail, etc. etc. but 2 days for a hand-crafted character is really short for anything that's worth the effort of custom making.

For starters, if it's worth hand-crafting a character then it's probably worth concepting them first, and that process is likely to take several hours in and of itself at least. If it's a hero character for a significant game then it's going to be far more - potentially weeks in total.

Then it needs modelling and texturing.

And we're talking about a character here, not a statue. So it's got to be rigged and then animations need to be made.

So, looking at the whole content generation process, you're looking at probably at least a week for a simple hand-crafted character, spread over at least three roles.

On the other hand, if they don't need to be unique or hand crafted then options like online content vendors are great. Plus there's stuff like Mixamo Fuse (though even there you're still looking at probably half a day to go through the process - picking stuff from a menu is easy, but that's not the whole process).

 

That's the other thing, 95% of the models are not AAA quality, but do take up AAA resources. And only about 50% are high enough qaulity to be usable at all. Then there's the swaths of stolen properties on the site, which could get you in legal trouble. And finally, getting all your assets of a simular style and quality is almost impossible to finish an entire game.

I'm still saying hiring an artist is the best option.

 

If you can afford it, yes, hiring an artist is a great option. Still, even when working with artists, my first port of call is almost always to see if there's something suitable available off-the-shelf. Even if it needs modification, saving half or more of the artist's time often means an asset pays itself off immediately.

 

Should also be mentioned, due to the title of the thread, there are some companies that specialize specifically selling assets. For instance
http://www.evermotion.org/modelshop/
http://www.rocketbox-libraries.com/index.php/

but might need some TLC to fit inside unity.

 

If you've checked all the typical stores like Turbosquid, Unity asset store and so on, then I guess as a last resort you could check out polycount.com and contact artists there directly and ask if they would be interested in selling you their models. Polycount.com is not a store but there are plenty of artist who have worked in the industry for years.

::Little bit offtopic::
You simply can't get AAA quality by buying from off self stores. AAA quality requires art direction and overall planning. A single asset may be very high quality but it's not going to match with all your other bought assets.

If you're an indie, a mid size studio etc. small entity IMO you shouldn't even try to aim for that realistic CoD/Battlefield/etc. look. You're always going to lose to those big studios with hundreds of devs. And even if you do get high quality in the beginning, its going to take so long to finish the game that its dated by the time you release.
Aim for stylized or somehow unique look that is fast to produce instead. It also helps you to stand out from crowd and makes you memorable.

If you really, really want to aim for realistic then your game's scope has to be limited. Only a few characters, small spaces and so on. After that you hire an art lead who takes care of choosing an art direction and also takes care of that everyone in the team follows that direction. You can hire freelancers to do the bulk work but you're still going to need a capable artist responsible for taking care of the overall look.

 

Why not? Lighting and post is up to a standard where we can easily re-create realistic (or near enough). I don't believe this is rocket science, it doesn't require masses of man-power neither does it require complexity. It's a solid foundation of how 3D models / lighting / effects and level design play into the whole development. I'll agree buying assets will not fill that criterion.

This below looks pretty AAA to me and all this really requires is simple geometric shapes, a good understanding of shaders / materials and lighting.

 

It's not whether you can or can not do AAA quality, it's all about time.
You don't really spend less time the better you get (at creating art) you spend more time because you tend to raise your bar exponentially.

 

Not if you've got a deadline you don't, you do the best within the time space allotted. But neither does that make it impractical or impossible.

It's about smart design.

 

Though I do agree there are ways to work smarter,
I still feel obligated to refer to the holy triangle dilemma, time vs quality vs money.
http://www.jonathancrossfield.com/wp-content/uploads/2012/05/venn.png

If you want AAA quality within an indie deadline timelimit, its going to cost.. a lot.

 

Best: You get the highest quality of characters by contracting an artist directly. (~$300 - $3000)
Good: Get the best looking characters you can find from sites on the internet. (~$200 - $700)
Poor: Buy sub-grade characters at disgustingly low prices (~$20 - $40)
Prototyping only: Use free characters. ($0)

This assumes that you are looking for a character, rigged, no (or few) animations, modern quality standards possibly with PBR. You could very likely get more unique, stylized characters at a lower rate depending on details and specifics. That is why the direct contract prices vary so much and will yield the most satisfying results.

$30/hr is cheap for a top notch, full time, freelance character artist. If they do not have a full time job then they are basically at rock bottom price at $30/hr. You could possibly get a lower rate from someone that is staffed during the week, and then working freelance on the weekend for some extra dough but realistically the money spent on "AAA" characters is much more than the highest figure above so expecting to get the same thing super cheap is rather silly.

Also, this is the more business style perspective. A lot of people just want to make cool stuff and think that they might maybe make something cool and turn a profit. In which case, these figures probably seem outrageous because they aren't planning on getting any returns, but they're playing with crayons anyway so it doesn't really apply.

 

Well that's it in a nutshell, if you don't have the manpower then you need the tools. Tools don't come cheap and it limits you to specific tools and engines. The shear amount of art packages we have to use for a 15 man team is silly whilst keep us at the cutting edge in a timely manor. Modo, Maya, Z-Brush, Quixel, Substance, Mari, Nuke etc.

I create artwork and I'm a coder, so I like to dabble in both over the past 15 years or so. The one constant swapping back and forth between a logically laid out game engine and then switching to modelling is how damn un-intuitive 3D packages are, it's like they go out of there way to make them completely abstract and illogical.

Yes I understand a lot of these packages include tools for CGI / runtime animated specials, but still even from basic navigation standpoint they are a pain in the ass.

Point being, I understand why some people don't even want to get involved or try..

 

Interesting thread. My biggest takeaway is good Lord people spend a long time on graphics. 2 weeks or more working on just one 3D model?! I can see putting some time into it but not 80 hours or more per model. At a AAA company with uber man years of time to waste I can see it but for an Indy I'd think 16 to 24 hours would be plenty. The huge time it takes for graphics is one of the main reasons I dropped back to an ultra low res low color retro graphics style for my current project. Even then can take up to 20 minutes sometimes to create the images for a new animated enemy and I feel I am spending too much time on it. Ideally every object would be knocked out in 5 minutes or less so I can get on with the actual game development. I mean there is so much more to the game than just graphics I cannot justify spending so much time on it.

 

It's usually custom made.

AAA quality that is wholly unique and for AAA prices.

Wholly unique means for 1 customer and not resold to others.
AAA prices means customer is willing to pay high prices for top quality.


Our art director is on the Sims 4 credit list.

 

This is pretty accurate. I'll also say that sometimes, the option of "upgrading" assets with issues works pretty well - often a model will not have PBR textures/rigging/LODs, but as long as the actual base model + diffuse texture are good, then you can really bump up quality.
As an example, I can turn this into this with B2M, and I'm not an artist by any stretch of the imagination, I couldn't model a stick (also, note the terrible looking low poly helmet in that screen).
AAA? Probably not, and you've got to remember that 1 'AAA' model plonked into your game may not look good at all - you've got to get your lighting/image effects/general visual style/etc right first. Some recent AAA games like Hardline and the new Rainbow 6 have come out with fairly unimpressive guns dotted around them though, so who knows.

 

Why should you?

I love the video you posted but in my opinion it goes to the category of scaling down your scope that I mentioned. You could make a narrative driven game with relatively small spaces for example and still keep up that quality bar. If you want a narrative game with multiple recognizable characters, you've already going to use significant portion of your man hours on sculpting characters. Scale that up to have a scene with a crowd and you'll also have start to think about rendering tech too. For smaller scoped games its fine.

The example vid looks fantastic. It's maybe 40 square meters of beautiful art. But how many seconds of gameplay would you get out of that scene in a typical FPS? You pass by the room, shoot 2 bad guys in the room and move on. Your ratio of time spent creating the room and time players spend there it is poor. Of course I don't mean it's not like you can't re-use any of the assets. You can and you should. But there is a point when you re-usability starts to show.

Plus the vid looks great because it has great lighting which in my opinion is 90% of beautifulness of any 3D scene I'd exchange 2 weeks on artists working on scene lighting / tuning scene mood than have them spent 2 weeks painting scratches to metals. AAA team can have both.

I suppose my point is to ask why do you want to achieve realistic AAA quality look in the first place? Are you somehow going to do things so much better than bigger studios with fraction of their resources? Why is it a goal you want to achieve anyway when you know you don't have the resources to get past mediocrity?

Which successfull indie/small/mid size team's games comes up to your head when you think of AAA quality realistic graphics? I can't think of many. Did realism help them somehow? Compare that to how many small team's games pops up to your head when you think of somehow unique visual art style.

There are some cases where it appears that some larger entities are thinking this too. For example Sunset Overdrive ended up going againts the realism route. It has unique style that is easily recognizable, open world city that looks great and was according to their GDC talks relatively inexpensive to produce. They produced an open world game and had total of 8-12 environment artists. But I have to say that they were smart in many other ways too so was not really just the visual style.

 

Again it's all about tools and logic, you create a base model and sculpt variations of it and scale to size. Then use UE4's ART tool to automatically rig, use FaceFX for speech. It takes 36 hours to create a model, but six to twelve to modify variations of them.

Our game is enormous compared to that scene and with the tools we use it's pretty simple to get it right:

(Blockout > Atmosphere > Export scene to modeller > add detail > export > apply materials > vertex paint decals)

In some cases sure, if all the game looks the same then it can cause fatigue. But on the other hand having slight variations in meshes causes enough of a disparency / discrepancy to remove that challenge. Also pretty lights and shiny things, really help.

DING DING DING, someone gets it, lighting is probably the main reason it all looks good. If you're painting scratches by hand for two weeks you're doing it wrong, you'd use Quixel to automate scratches and effects. The point? Well competition for one, a lot of Indie's are doing some beautiful stuff in UE4 and I can link TONS of examples.

Which is great, people whether others want to believe it or not, do consider graphics and "immersion" as quite a large part of the experience, whilst I'll always believe artwork comes second to gameplay and plot. It's not an area you want to lack in when you're working in a professional fashion.

Plus all in all, it's not THAT hard to do. Sometimes AAA artwork sucks as well, ever seen skyrim? The modeller in me screams when rocks look like badly UV mapped octagons, but I understand there has to be some give. Did you know that ONE person fixed ALL the artwork mistakes in Skyrim? Makes you think doesn't it?

You don't know if we have the resources to get past mediocrity and you never will.

It's about balance, I wouldn't expect a team to be able to build there own engine and then furthermore be at the direct cutting edge of graphical fidelity implementing and creating new ways to improve above and beyond. An indie can't do that, we don't have the money or funding and probably in some areas the skillset.

But what I would expect to see is AAA quality scenes with what we have available to us, sure we will never push the boundaries of what's possible fidelity wise. But that's OK...

 

I don't think you should be trying to compete with "AAA" graphics. These games have hundreds to thousands of artists and engineers working on them for 3+ years, with decades of experience, using the most cutting edge software. People will be willing to buy games that don't quite measure up to those graphical standards when they're priced right. Making a game with not quite AAA graphics and selling it for $60 probably won't work out. But selling it for $10-$15, people will feel its a better value. Especially if its a fun game (which should always be the main focus).

But I do agree with the guy who's saying lighting and composition make a HUGE difference. It's very true, especially now that Unity is using GI and the standard shader. These things can compeltely transform even the most basic models and textures. If anything, I'd say textures are more important visually than model complexity.A simple model with fantastic texture work won't look any lesser than a "AAA" model.

 

Hundreds to thousands? Decades of experience? Quite a bit of an exaggeration there.

 

When I think "AAA" graphics, I think Rockstar, Ubisoft, EA, etc. All of which have studios numbering in the thousands, with highly experienced engineers and artists. GTA 4 and 5 each specifically had full-time teams of over 1,000. Assassin's Creed, CoD, etc. also have similar size teams

 

@ShadowK Lots of good points and I agree with most of them. Smart asset production gets a small team much further.

I still think that even if we now have more sophisticated tools to produce realism doesn't mean that it makes realism is a goal we should automatically be aiming for. Most of these production tools are just as usable in a more abstract presentation anyway.

Generally speaking, I think it makes sense to take a deep breath and have a look look at your design. What do you want to emphasize? What do you want the player to experience? The general goal of the game. And then think of visual ways to support those goals.
If your design absolutely needs to have realistic look, then go for it. It's justified and if your scope is small it is something you can achieve these days. But often it seems that realistic approach is simply chosen because it comes to most of our minds first. In my opinion it should be the last resort considering the costs. Hence the original post you quoted.

Absolutely agree. Immersion is really fascinating subject on its own. Depending on what your goal is, it can make sense to choose realistic approach for immersion reasons. But it has to be justified. It's definitely not as simple as realism → immersion. Realism can easily work in the opposite way too.
There's a really interesting book called “Understanding Comics: The Invisible Art” by McCloud which covers a part of how we as humans think of icons/abstract presentation versus realism. The book itself is about comics but most points that the author makes are universal to any visual art. He covers topics like how eliminating details amplify what is left and enables viewer to focus on what is actually important. And how different levels of abstraction greatly influence viewer's identification of objects and characters (abstract presentation = you, complex = someone else). It's great read for anyone interested in any visual art form.

 

Most of those studios are not developing one single game but rather multiple games side-by-side. Grand Theft Auto V being practically the only exception to that rule with all of Rockstar's studios working on it at once. Normally those other studios are producing their own titles though with Rockstar North being the GTA developer.

The lead developer has even stated, mentioned in the site below and others as well, that the game having so much detail over the previous required so many more developers. This indicates that GTA IV had a much smaller team.

http://www.develop-online.net/studio-profile/inside-rockstar-north-part-2-the-studio/0184061

Additionally these companies are typically developing their own engines. Simply glancing over Unreal's staff shows that it takes considerable effort and staff to develop and maintain an engine. Yet you do not need to make your own engine to qualify as "AAA".

 

I think some people and OP overvalue assets found in AAA games. If you take a closer look on these games, often you will find broken UV mapping, texture positioning etc. Why AAA games look AAA is mostly because of composition and I can easly find some great packages on Asset Store that could fit in many of best selling games. Don't want boast here but I'm sure my weapons pack could fit in :]

 

Thing is, it seems many don't understand that it's not really an artwork issue but it's a technological issue. I realised this from reverse engineering arch viz setups in commercial offline renderers. Any muppet can put some primitives together in a scene and apply a vector 3 mask and noise, then export it out to V-ray and it'll look amazing.

It's sheer basics, look at this example explain to me anything artistically complicated about it?

This is Just a bunch of rectangles with a basic material on it, but it looks gorgeous. Why? Well gorgeous post effects, PBR and Photon mapping. If you could screw up the actually artwork in that scene, I'd just give up.

It's well beyond what's is rendered in current AAA visually and it comes down to this, performance. It's one thing to make a game look beautiful and it's another to make it run well across multiple platforms which contains very weak hardware.

In larger games, baking starts to become impractical if not impossible. You don't want to release an 80GB game where 60GB of that is lightmaps, so you then look to real-time and run into performance issues with global illumination. Not only is the quality low the impact is high, things all of a sudden start becoming complicated.

So personally I did reduce the size and complexity of our scenes due to this sort of headache. I also changed the setting of the game to take advantage of things like SSR.

AAA would throw money at a problem like this until the issue went away, we can't waste the time or the money so we re-factor..

Don't get me wrong, making artwork is hard and extremely difficult dependant on the game you're trying to achieve. Realistic openworld RPG is a fools errand even for some AA / AAA companies, so there are a lot of factors to this.

You can't just blanket statements like "you shouldn't do this". Because a game can be big, it can look great and it can be done by small teams. But it requires a certain direction and methodology that frankly for beginners is too much to handle..

 

Is this turning into another "What is AAA and what is not AAA?" thread?

 

Forces pulling from the center of the earth again.

 

Always. You can't have a discussion about how to make AAA without a working definition of AAA. You can't have a working definition of AAA without strangers arguing on the internet.

 

Hmmm nope, it probably has more to do with Jaffa cakes and Unicorns than it does with what's AAA and what isn't.

 

I think generally, Jaffa cakes are kind of indie, while Unicorns are rare, premium horned horse. Unicorns are pure AAA.

Jaffa cakes aren't even really cakes even though they qualify, so they couldn't possibly be AAA.

 

Your assumption at $20 an hour is actually really low. $20 an hour at full time is $41,600 a year. This is a rather low salary for artists. Also you have to consider that if contract work is your living, you won't have work all the time, you have to pay for your own insurance (depending on country), taxes can be different, cost of the tools (Zbrush, Max or Maya or Modo, photoshop etc.). So realistically take what an average artist makes in the games industry:
http://www.gamasutra.com/view/news/221533/Game_Developer_Salary_Survey_2014_The_results_are_in.php

Which the average artist salary is ~$74k. Now per hour this is $35 an hour. Factor in everything else and most artists wouldn't lift a finger for $35 an hour for contract work (if they've worked in the industry) unless they really need the money. You are realistically looking at $50-$70 an hour for contract work for an artist. So a single 10-15 hour asset can cost you on the low end $500 and on the high end $1050.

 

Jaffe cakes are a realistic goal. Unicorns, not so much. No matter how big your team, you still end up with a horse which has a horn glued to its head.

 

Pshhh! but they don't "get it" my tinfoil hat says, I'm talking about dividing the square root of a Jaffa cake * a Unicorn. It's about being cleverer. So you end up with a JaffaCorn..

Seriously how could anyone miss that?

 

Since you said JaffaCorn, in that order, I would assume that its the rubbish bit mostly with some AAA on the end. If you said CornJaffa, it might be the first level is amazing but the rest are crap.

 

Anyone can pick one part of something and say it's easy, because that's true. But you need more than arch vis to make a game. For most games it's not one room, it's dozens of rooms over dozens of levels (and that's ignoring the idea that you might want to go outside), with characters, props, animations, effects, scripted events, gameplay elements, etc. etc. etc.

 

You could make a AAA escape the room game.

 

I think that'd be beyond most people. The art assets alone would be difficult.