Does anyone out thre work with game/software development who can provide any insight on what would be the actual “limit” (and if there is such thing) of the current WoW engine in terms of visual capacity? I mean naturally it gets better every expansion, but I was wondering how far they could push it/update it, comparing to some modern games, as example.
I know Blizzard wants to preserve the cartoonish-like style that WoW has had since it’s inception, it’s tied to it’s identity, and it would be impractical to revamp everything at once, but I regardless, don’t think I little bit elevated realism/dark fantasy would hurt.
i’m a software developer, and i’ll agree with annastasi here. i can’t give you any specifics because wow runs on an engine developed in-house and not on commercial engines like unity or unreal, which are much easier to examine. i suspect wow’s engine is capable of rendering more, eh, “realistic” objects, and that the more cartoony look is a deliberate design choice.
Expanding on this, because it’s an in-house engine it could mean one of two things:
1.) they have much more flexibility in customizing it and modernizing it however they want as they own the source code completely versus going with an out-of-the-box solution like Unity;
or
2.) they struggle making the slightest changes to it because of poor internal documentation about the engine since inception and a lot of the skilled labor that created it/worked closely with it have migrated off to other projects and/or companies
There is no such thing as limit, most people think engine as something like game console where PS1 can only achieve PS1 level of graphic.
Game Engine doesn’t work that way (especially for one made inhouse), if Blizzard tomorrow want to double the amount of tris of every WoW model they have, they can do that, if they want to develope/change the preexisting shader to look like Unreal, they can do that, or if they want to import Overwatch’s net code, they can do that.
The reason why they don’t do that is for accessibility, the average pc specs will keep getting better each year and Blizzard will adjust to that.
I feel like most of the things you’ve attributed to engine limits are stylistic decisions from the art team. However, given their minimum specs tend to be extremely modest, you’re only likely to see incremental enhancements in graphical fidelity over time (especially since those specs have to last the duration of the expansion, since they don’t tend to change them mid-expansion for obvious reasons).
I have seen realistic models beign imported in-game with patches so You are mostly right. At least on the model side of things, Illumination and photorealism would be a different kind of worms, I think
One of the FEW things that keeps me playing THIS game is the cartoony graphics.
When WoW looks like ESO, I’ll go play ESO instead given that overall its a much better game.
They can update the skins of just about anything in the game to look more ‘real’. The models wont be much better, but they can take real pictures of tree bark and leaves and certainly make it look more real.
And that would literally be the end of my time playing WoW.
Im still here hoping they’re going to give up the petulant games…but without the game looking like it does now, Im out. That’ll just be far too much for me.
If they make it like ESO, I may as well just play ESO only.
Of course the code is a mess. It is over 20 years old and has had dozens if not hundreds of developers working on it in those decades. In 2000, it started out as a modified version of the Warcraft 3 engine. Since then they have changes and tweaked it millions of times. There are probably hundreds of classes and objects that wouldn’t be necessary in modern development. This doesn’t even touch the backwards modeling in the game were almost everything is actually just a a pile of 2D textures and not truly a 3D model.
It would be even worse to try and change it to a new engine… Or maybe people want the next expansion to be released in 2027 or 2030.
WoW is probably trying to maintain a careful balance of looking as good as possible without out-teching customers that do not have the best hardware/internet options.
I think an example of problems that can arise if you consider Wildstar. It used the same type of cartoony graphics and while my PC at the time had zero issues running WoW, it bombed trying to handle Wildstar. I was lucky to get 20fps playing out in the world, and any form of instanced play was impossible.
I’m not a tech so I presume that although Wildstar looked like the same style of game, it had a lot more going on under the hood than WoW did.
All models, in 2004 WoW and in Today’s wow are “truly 3D models”.
It was just less taxing on the average 2004 computer to have very, VERY primitive models for trees, which are still present in some zones in the current live game.
Any codebase running for 17+ years is going to have accumulated a pretty substantial amount of technical debt, huge codebase that nobody fully understands, missing tests, legacy broken code with no time to fix, core code where the maintainer has left and it now can’t be touched in case it breaks, all which slow down development.
You can see this with bizarre bugs like Kael’thas baldness. Nobody obviously meant for that to happen, weird things happen with millions of lines of code though:
This is one of the reasons a rewrite or sequel would probably speed up their development massively. All the accumulated spaghetti code could be tossed. They could also build in modern technologies like raytracing, global illumination and physically-based rendering from scratch instead of trying to retrofit it, which never works as well. They could also make all gear sets fully 3d modelled, instead of pasted on textures, which is another holdover from the 2004 original codebase.
They weren’t though. That is why it took until Cataclysm to allow flight on the original continents and why flight still isn’t allowed in the Draenei and Blood Elf zones. Cities and some mountains were originally built as a “Hollywood Set” style that couldn’t be viewed from the sky.
Some video cards already have the version of unreal you can use it but it is limited and not good for playing the game due to frame limitations . Got more people and it could be a process problem.
I don’t think anyone wished to crash their system after paying a lot of money on it.
Yep. It’s easy to forget if you’re in a gaming/tech bubble but your average person’s machine looks more like “$600 Walmart special minitower from 2016 with a GTX 950” and “beaten up 2014 Macbook Pro from college” than it does newish, high end, or midrange anything, especially nowadays when you have to give two limbs, a kidney, and your firstborn to get a graphics card.
Only Blizzard devs could say and honestly they may not even know what it’s breaking points will be in future given how old and cobbled together in places it is. They do run into restrictions now and then that prompt them to either change the engine so they can continue on or abandon the idea as they can’t justify the work to make the changes to the engine. There may be a point in the future, especially if WoW grows less profitable and begins to contract that they may run into a situation where they can’t afford to fix the engine and can’t afford not to fix the engine so they just stop development. By contrast if they ran a commercial engine from another company like unity or unreal there’s no reason to suspect those engines won’t be supported for potentially decades to come so they would in that sense be in a more stable position. On the other hand they’d also have the drain of the costs to license such an engine which would mean WoW would be unprofitable a lot sooner potentially.
The limits aren’t really visual. It’s not like processing power as someone else noted. It’s more what can the engine do with models and the underlying math. And importantly does the engine and assets you’re using it with overwhelm the equipment your customers are using to play your game?
They’ve definitely said number 2 regarding the engine being in places held together with spit, tape, and sacrifices to old gods. At this point I’m sure many of them are unhappy with it. There could be some who feel that number 1 applies to a degree but I’m not sure. At this point replacing the engine would be a major undertaking, besides the fact that it would mean expensive licensing costs to the developers of that engine which would reduce profits.
Yep. It would reduce costs, speed up development massively. The problem is it’s a huge upfront investment for a long-term benefit and for a game as old as this corporate may not be at all interested in retasking teams to make a new engine and implement it and the associated high costs and drain on profits. They think about quarterly results, not about how to get the best use out of money and dev teams over the next 4, 8, 10 years.
