So… recently I’ve been raging about the slow hero release rate, and people started telling me that I have no idea about programming, and how long it takes to make a hero.
So I took the challenge, and I tried to duplicate a hero from HotS (Kael’thas), in a custom WC3 map.
I got a working version of his Flamestrike in about 50 minutes (took me a while to set up all the stuff in the Object Editor, like Kael’thas himself, and to set up the trigger ability… coding it took maybe 25-30min), and a working version of his living bomb (although not tested to its full extent yet, and not yet working EXACTLY as it is working in HotS) in another 50 minutes.
I haven’t done any talents yet. This may slow me down, as I’d need to create the “quest system” on my map (while the devs probably already have it on their “editor”, and don’t need to make it from scratch every time they release a hero with a quest). But I’m fairly sure that I can finish him within less than 48 hours, spent on him.
But do have in mind that I hadn’t worked with the WC3 Editor for about 2 years, so I am slower than I can be.
Also, do have in mind that I’ve studied for an engineer, not a programmer. My programming skills have been developed as a hobby.
The bulk of the work probably comes from fine tuning and testing his talents and numbers.
As this game has painfully shown over and over, even a single misplaced 0,5HP, mana or damage somewhere in the code can completely break that hero and make them intolerable to play against (or with).
And it can take the developers weeks to find out where that misplaced number is.
I am well aware that playtesting is the slowest part of a hero’s development. But people who “challenged” me talked about it as if coding itself was the hard part.
In fact I had forgotten 1 line of code in my Living Bomb, and it was glitchy. I decided to take a bit of rest and just as I closed the editor, I remembered that maybe I missed this line (and when I checked - I indeed had). I added it and it started working.
The time-consuming part of making a hero is conception, modelling, animations, particle design, writing, getting the voice actor, ect. Everything has to look and feel different, fun, and readable. Sure you can remake Kael’thas, but how about you take that same process and make a new hero with a new kit and model from scratch? Like, I’m sure the HotS team has twenty heroes in production, but it takes time to bring them to life.
I’m well a were of that. But all of these don’t take longer time than creating a single unique skin with alternative voiceover. As long as you do the playtesting with a placeholder model, while the actual assets are being developed.
Thing is: I’ve already suggested several hero concepts in the Hero and Skins suggestions. And most of them have complete Talent trees. So I can indeed make 1 of my suggested heroes instead.
As for how long it took me to make these concepts… Well… each case was different, but most often I had some “core idea”, which I wrote down, and I filled in the details (such as filler talents) while writing the forum post. So about 2-4 hours.
I suppose that some of them may be “immature”, as I had suggested 1 hero 2, or 3 times with different kits. They were all centered around the hero, yet they all played very differently. The last one I did was taking into account the difficulty of applying such hero to the game (due to the immense amount of models previous iterations required).
Yeah. They do that. Go watch the Making of Kel’Thuzad mini-doc for a look at their process. They tested him with his WCIII model for most of development. They’re not sitting on their thumbs twiddling all day. What are you even arguing right now?
My point is: coding doesn’t take as much time as people seem to think.
If the devs can release 6 skins + 1 hero every 14 weeks, they can instead release 4-5 skins + 2 heroes for that time.
Thinking that the big part of making a new Hero is coding is ability is childish.
Animation, models, voicing, graphical effect and so on are what really take away enormous amount of time. If the Hero has some new mechanic that will also take away a lot of time because it need to be tested with everything already existing into the game (and even than something may slip through).
On top of that you have to consider that they actually do some internal testing and that they’re always working on several new Hero at the same moment.
No, they cannot. They could when their team was 3x larger than it is now, but they obviously can’t go any faster.
Seriously, what are you arguing? Do you think they’re just… choosing to not add content? They’re too lazy to apply themselves? I don’t understand. I’m sure they can program 600 Kaels in the game in a week. That doesn’t matter at all.
In a recent interview they stated that they are happy with their current release rate. Before it used to be “too fast”, so they slowed down.
Of course their team was cut down in size, but it was also their conscious decision to reduce hero release rate, and they are not at their highest hero release rate.
So I’m saying that they should increase the hero release rate to at least 7-8 heroes per year, even if they have to sacrifice few skins for it.
I think I agree with the OP on the spirit of the thread. Can the devs go faster still? Yes, they can. It’s possible. Small teams can still achieve considerable results in a short time.
Should they? I don’t know. Deep down the hero release slowdown is related to them cutting costs as even if your developers do the lion’s share of work what consumes the most money on hero releases is the portion you don’t spend to your core employees. Voice actors, artists and so forth. While you can kinda solve some in-house and Blizzard actually has a history of literally doing everything like that at one point (everyone knows who voices Thrall, for example) I doubt it’s sustainable now.
Speaking of sustainability, if you push your already emancipated staff further then it’d only lead to burnouts. Cautiousness in this regard is fine and businesses worldwide realize the necessity of this. So it is fairly reasonable to reduce the total workload of the staff to a third when their numbers similarly degraded.
Well, this thread highlights something, and that is:
That you really have no idea about programing
If want to do the challenge of how hard is to make a new hero i recomend getting Blender3D, choose a random hero from any Blizz universe that is curently not in Hots, and build it…everything…
Walk cycle, Dance, Taunt, Ability animations, FX for every skill it has, textures.
Than if thats done, get Unity engine, import all the hero files and start programing that:
mouse buttons move command/attack comand
Q W E R D buttons use the right animation and FX in correct order
And if that is set up, u can start working on the Hitbox and Dmg boxes to function correctly.
And ofc the bugfixes
And when that is done show it to Blizz forum so everyone can tell how Overpowerd, busted abomination of a you just made and they don’t like it cause it’s not their favourite character
So back to the drawing board, GL HF
PS: IK it’s trolling but i just had to write something with my afternoon coffee
My concern is that they wouldn’t be sacrificing skins for this, but reworks. Although there are a few glaring counter-examples, for the most part, the reworks over the past two years have been fantastic, and I would hate to lose those. Personally, I am perfectly fine with fewer heroes if they manage to make the entire roster viable and relevant, and change things when needed.
You created 1 Hero in vacuum and you had to do only the abilities (no talents) and it took you an afternoon.
Now imagine the same when you need to code in a way that it won’t interfere with existing code, Heroes, plus you need to do talents, and the model, animations, particales and effects syncronise them and you had to create the idea and playtest a lot of things.
Sure, the coding may not be as timeconsuming as the ppl you argued with thought, but it’s still probably harder in reality than what you think based on your “test” and even if the coding is not the problem and the slowing power, thinking that the devs can just speed up the release is delusional.
Especially Whitemane’s rework! Just ask hellobg what he thinks of it. He, with his level 100+ Whitemane, made it perfectly clear that they ruined Whitemane, and she is no longer a viable hero
Anyways…
Most reworks don’t require additional assets to the game, so mostly the coders are involved in them. While hero releases, apart from coders, require the art team to also work for that hero.
In the OP I pointed out that coding is the easy part, while skins are the part that take the most amount of time. So it’s logical that they’d be releasing less skins than less reworks.
How many heroes do you think truly need a rework right now? Ones that are not confirmed to be getting ones (Diva, Tassadar).
I haven’t “updated my rework list”, but I’d say that very few heroes are in NEED of a rework, and a just a bit more could use a rework.
The WC Editor has this built-in interface, where you can just select the animation a unit plays when they use an ability. It’s pretty basic in the WC edition, but if it has it, then that surely suggests that the devs would have something similar in their “editor”, and they wouldn’t need to manually code the animations played every time they add a new ability.
But even if they had to do that, it’s not such a big deal, as it only adds 1-2 lines of code to the ability.
The developers are split in different teams. Coders work on code, animators work on models an animations. It’s not as if the same person does all the work by himself. Coders also use the already made models by the art team, or they use placeholder models until the actual assets are complete.
I’m presenting the “problem” from a coder’s perspective.
you just made and they don’t like it cause it’s not their favourite character 
