RedFury123

Join Date 07/2010

Thoughts from an ex-game developer

Hi Brad,

Just want to start this post on a positive note by saying that I believe Elemental can be great. I've been following things for a while and even made a couple of passionate posts about things I'd like to see improved.

I was a designer/programmer for a games company for 6 years from the late 90's to mid 2000's. Worked for a smallish company making RTS and turn-based games not too unlike your own - some here might have even heard of some of them. I also did AI programming for these strategy games so my hat is off to you there. You're a much better AI programmer than I was, I'll say that. It's one of the hardest things to get right and people crucify you when you don't. My AI's ranged from 'reasonable success' to 'epic fail' but nowhere near the level you've reached in the past and will reach again. You have my utmost respect in that area.

Anyways, just a couple of quick points I wanted to bring up from my own experience. This might be slightly releveant or totally not at all.

1) After a long, intensive development cycle working on a game, the fun starts to become the "working on the game" rather than the game itself. You start to become invested and attached to the choices you've made and the systems you've created because you are proud of what you've done and you feel that they're fitting in with your overall vision. Much care needs to be taken to seperate this from what others who casually pick up your game and give it a go would find logic, interesting, flavoursome, strategic and FUN.

One distinct feeling I remember after having my head buried deep in the system design and coding of a game for 2 years, once the game finally nears completion, is - "oh my god. I have no idea if this is actually FUN for the average person to play!" That probably sounds really weird to most people here but I think anybody who has gone through a full game dev cycle and shed blood, sweat and tears over it will probably know what I mean.

What I'm saying is, I think its easy for devs to get so wrapped up in the day-to-day workings of the games they're involved and emotionally invested in that you can lose the forest for the trees very easily. Sometimes I wonder if this has happened a little to you with Elemental because, since (bugs aside) most people still seem to think that (design wise) this game is currently very promblematic and has a long ways to go, you constantly quote that is "the best game you've ever released." I've played some of your other games Brad and as a person who just loves strategy but is far less emotionally invested than you, I just don't agree.

I hope you've got a couple of people close to you that you *really* trust that have the same vision for the game as you but are not nearly as tied up day-to-day in it or invested emotionally in all of its design choices. You can use such a person/people as a barometer as to how good the game really is. I'm talking about someone who digs the style, grand concept, strategy and content but will give the game a shot once a week for a few hours and then give you honest feedback on what works and what doesn't. Beta testers could possibly fulfil this role, but there are all too many of them and opinions can vary wildly. Either take a step back yourself or start listening to others whose judgements you fully trust who can do that for you.

2) I want to make it clear that I"m referring to game design here, not problems with the new engine. Also, this is a description of the way we used to do things and how they seemed to differ from Elemental - not a "you should have done it like this!" rant although regrettably that may be the way it comes across.

Beta phase and release strategy. I have never seen a release strategy like yours before. Clearly it worked for your previous games like GalCiv2 and kudos to you for that, but honetly I don't know how. Yeah maybe things have changed a bit in 6 years but in the company I worked for, your "Betas" are called "Alphas" - its where you're weighing up game systems against each other, keeping the same basic systems but messing quite a lot with how they work. I would typically be doing this around 6-4 months before release. An example of an ALPHA task during this timeframe would be fundamental changes to the city building mechanics and resources mechanics. The type of changes that I was amazed to see happening 4 weeks out from elemental release.

Once your alpha is feature complete, all your system functionality and behavior is locked down, then you go into "Beta" where you don't mess with core game systems, its just bug squashing and balance, balance, balance. This is actually the phase when you really, properly *play* your game the most and discover where all the fun lies, and what areas could do with extra injections of fun. We'd normally start this process about 3-4 months before release so that by the time gold came around we were in good shape. Of course you are still going to have bugs because testing with a QA dept of 16 different machines can't possibly conpete with the millions of combinations out there, but after 3 months of beta polish, bug-squashing, tweaking and balancing they'll be down to a minimum.

From what I can tell, this beta stage for Stardock was called "getting ready for gold" and happened about 2-4 weeks before release. Again, this strategy has clearly worked for you in the past but I think I"ve seen enough game dev to know that its *very* different to the way most development houses would choose to do things if they had a choice (like you guys do) rather than being pushed around by publishers like most small devs. I just can't fathom how you can gel a whole bunch of independently tweaked game systems that are only finalized and come together two weeks before release into a really full game in that timeframe. In my experience created the balance and bringing the fun out of most games takes a lot longer than that - because, and I hate to use this cliche, but in the truly great games i have played what has made them truly great is that the whole is far superior to the sum of its parts. You don't get this without a LOT of love and care in the true 'beta' period of the game.

So, thats it really - i hope I haven't come across as being too harsh or even hateful. Thanks for listening (or not .. I know how many posts are made on these forums.) I really hope Elemental will become the game everyone wants it to be. I know you love Elemental like your baby, but I just get the gut feeling that you might need to step back and accept some home truths about its current *design/balance* (rather than just its engine/bug state) for the fun in this game to truly come out and shine. I sincerely hope that happens as trust me, this game has the potential to be far, far greater than anything I ever contributed to.

elemental

Locked Post +3 Karma

Advanced Search

Watch this post

Do not email me updates for this post Email me updates for this post

Successfully updated karma reason!

Nick-Danger
Blaze of Glory
riadsala

Do they have commercial compatibility-validation (CV) houses out there that have built up a huge array of platforms and contract out the validation of people's software? That sure seems highly commercially viable to me. Heck, if I wasn't doing well already, I'd be going to the Venture Capitalists and pitching my idea right now.

Successfully updated karma reason!

I thought this too when I saw how SD did their "beta" process. It seemed a lot like an alpha version when you were still deciding game mechanics, but considering that they have always been open to ideas from other gamers, I didn't think that was such a huge problem. On the other hand, the lack of beta stability/balance testing seems like the major culprit. Of course, hindsight is 20/20, and from what Brad has been posting, it seems like he and others at SD have figured it out as well.

Successfully updated karma reason!

Tom Chick perhaps said it best in that Brad (and through his example, the rest of SD) is a programmer. He does things from a programming perspective. This influences how he approaches things compared to how a designer or artist might and how all three disciplines would work together. You can kind of see this same thing in other studios with studio heads that are active developers - iD/John Carmack, Maxis/Will Wright (well formerly), Firaxis/Sid Meier.

So yeah, if they can put on some strong designer googles, things should be okay. Perhaps never as good as it could have been, but not much to be done about could have beens to begin with.

Successfully updated karma reason!

Interesting you think that. I've been thinking the exact opposite: NOT strong programming discipline. I'm not fascist when it comes to coding standards--and believe me I have met PLENTY of code fascists--but there are lots of little things you can do to keep your code sane which don't impose anything.

I've never designed artsy games/programs myself, but I have built multithreaded, event-driven UI's. I think the discipline needed here is to well-think out your design BEFORE your software people spend way too many late nights on it. It's demoralizing to jerk them around. You say you want one thing, they spend 2 days and 2 nights getting you what you wanted--and then you come in later, 30 seconds and a hand wave, and now you don't want it anymore. Then several months later performance appraisals come along, raises are given--and the software guys don't get any credit for those late nights because somebody threw out all their work with a single hand wave.

You need to show the developers more respect than that, and make sure you've really thought through your design first. Debate it over with other folks (including your developers--actually, especially your developers). In some cases you can rapid prototype it in a RAD environment. Heck, I'll use my web browser to prototype an interface. It doesn't mean you won't still throw it out even after all that, but at least show some reasonable effort not to jerk your programmers around.

Successfully updated karma reason!

Kamamura_CZ

Join Date 03/2006

+34

The confusion about the alpha/beta phases can be easily clarified: we, the customers are part of the beta testing team

Not that it must be necessarily a bad thing - in-house beta testers may suffer from "collective blindness" - the actual userbase may produce much detailed and complete feedback.

Paradox routinely does the same with their games - they are released mostly with gamebreaking bugs (EU3, or HOI3), but after several months/years of polishing, the games begin to shine.

The important thing is the commitment to post-release support - which is there in case of both Paradox and Stardock. Mind you, I would not buy a game in this state from, for example Creative Assembly, but here I trust the game's problems will be eventually sorted out.

But I strongly agree with the original poster - it looks like the devs became so immersed in their work so they stopped seeing the forest for the trees.

Successfully updated karma reason!

Blaze of Glory

Join Date 04/2006

+27

Paradox routinely does the same with their games - they are released mostly with gamebreaking bugs (EU3, or HOI3), but after several months/years of polishing, the games begin to shine.

Put that is the problem. PC games are dying because they come out buggy and it seems most folks these days won't wait for six months or a year. They will be on to the next thing. As the saying goes, you only get one chance to make a first impression.

While there may be a core group that will give that kind of leeway of paying for something and waiting months or a year for it to be what they expected, most people will not.

As to the OP, a very well written piece.

Successfully updated karma reason!

Quoting tetleytea,

reply 4

I think Elemental suffered from a strong programming discipline overwhelming the design and art aspects of the game. It's not bad in itself but it is bad, as you mention, when it starts to take away from some of the other important things in the game.

Interesting you think that.   I've been thinking the exact opposite:   NOT strong programming discipline.   I'm not fascist when it comes to coding standards--and believe me I have met PLENTY of code fascists--but there are lots of little things you can do to keep your code sane which don't impose anything.

I've never designed artsy games/programs myself, but I have built multithreaded, event-driven UI's. I think the discipline needed here is to well-think out your design BEFORE your software people spend way too many late nights on it.   It's demoralizing to jerk them around.   You say you want one thing, they spend 2 days and 2 nights getting you what you wanted--and then you come in later, 30 seconds and a hand wave, and now you don't want it anymore.   Then several months later performance appraisals come along, raises are given--and the software guys don't get any credit for those late nights because somebody threw out all their work with a single hand wave.

You need to show the developers more respect than that, and make sure you've really thought through your design first.   Debate it over with other folks (including your developers--actually, especially your developers).   In some cases you can rapid prototype it in a RAD environment.   Heck, I'll use my web browser to prototype an interface. It doesn't mean you won't still throw it out even after all that, but at least show some reasonable effort not to jerk your programmers around.

I'm not talking about code. I'm sure the code could be better or worse. I'm talking about the design aspects of the game and the lack there of. I don't think there were enough designers or design mindset. The UI, mechanics, and presentation are built for experts; if you know how the game works or are willing to think like a programmer, the game works. If you're looking to -learn- how the game works and think like a player, the game design is obtuse and labyrinthine about things. Right now the overall design of the game is that it's a game made by experts of the game for experts of the game.

For instance, there's no real good reason to split the inventory screen from the stat screen much less require dropping in and out of menus to reach them. This makes perfect sense from a programming standpoint (or maybe not; at any rate, it's not -wrong- from a programming standpoint) but it's horrible from a design standpoint. It's horrible for non-experts trying to find and figure out what they need.

Take the 5 tile limit for cities and resources as well as how it's relative to the nearest city tile. It's not mentioned -anywhere- in game and it works differently than other 4X games. The assumption in the game is that you just kind of figure it out and know. It's like garbage collection in Java or semi-colons in C/C++. It's just one of those things you just know after one or two mentions without really considering that for someone who -doesn't- know has no clue at all what's going on and why.

This applies to the game as a whole; outside of the programming issues, things are designed as if the game were a debug/game development tool set not a game UI that creates a nice gameplay experience.

Successfully updated karma reason!

Quoting Blaze of Glory,

reply 6

As to the OP, a very well written piece.

Agreed.

Successfully updated karma reason!

Blaze of Glory

Join Date 04/2006

+27

Quoting Gwizz41,

reply 9

quoting post

Hi Brad,

Déjà vu. Brad has already spoken numerous times in threads started and posts made!

Whew, thanks for posting that, saved me an hour of reading the forums...

Successfully updated karma reason!

Very interesting read.

For me, the question is, can the game be patched up and fixed. Brad has talked about 'surgery' being needed, I am in complete agreement. And I agree that the lack of a unified direction/vision is quite evident, and has hurt the release. It *may* be that the game cannot be patched up, they could get the various systems already in place to play nice with each other and better relate to each other, but would that result in a fun, engaging game? We don't know. Nobody does, including the devs.

Successfully updated karma reason!