Sign in to follow this  
Norman Barrows

Does releasing in-house tools lead to better tools through user feedback?

Recommended Posts

Brain    18906

I guess this depends on how specialised the tools are.  If they could only be useful for making your game, then you're unlikely to get many users to offer feedback. If they're more generic people will download.

 

Perhaps you could open source your toolkit?

 

This has the benefit of having others add to your tools without you having to do the work, and cuts down on the support you are expected to offer with it too...

Share this post


Link to post
Share on other sites
MattSutherlin    1210

The short, obvious answer is... yes.

 

More users leads to more potential use which leads to more potential feedback.  You aren't guaranteed to actually get more feedback, or for that feedback to be worthwhile if you do get it, but you'll increase the odds.  And that can be a very good thing.

 

The other end of the spectrum is what you're opening yourself up to.  If you're writing a custom engine, you're probably targeting the one specific game that you're making with it.  That means that you're likely taking shortcuts in places that you find acceptable to the workflow and final functionality you know you need for your game.  You might also make some design choices in your game based on your known tool/engine limitations.  Your final game seems perfectly functional and stable because you intentionally never create scenarios you know would effectively be broken for the player.  As soon as you publicly release your toolset, you are showing the world the flaws they wouldn't otherwise ever know about.  Users are going to try to set up scenarios that end up being broken, and won't necessarily understand why.  Your game and engine might instantly feel less stable to your users.  Is that something you can live with?

 

As a personal anecdote, I was working on an engine with a publicly available toolset a few years ago.  I was working on a new visual effect, got it implemented, did some optimization passes on it to make sure it would run acceptably within the game.  As part of the code review on it I was asked if it'd work on 5000 overlapping entities.  It wouldn't, but I had made sure that it'd work with a set of overlapping entities that was still far beyond the unit cap in the game.  That should be fine, right?  Nope!  Because some user could set up a mod with our toolset and easily overlap 5000 entities.  Would that be insane to do?  Yes.  Did we need to try to support it anyway?  Absolutely.  So, I had to de-optimize a little bit to make sure that our users had a better tool experience.

 

So, I'm not saying you shouldn't.  I almost think you should.  You just need to be prepared for what's to come if you get the best case scenario and people actually use the tools you release.

Share this post


Link to post
Share on other sites
Norman Barrows    7179


Perhaps you could open source your toolkit?

 

This has the benefit of having others add to your tools without you having to do the work, and cuts down on the support you are expected to offer with it too...

 

 


The short, obvious answer is... yes.

 


So, I'm not saying you shouldn't.  I almost think you should.  You just need to be prepared for what's to come if you get the best case scenario and people actually use the tools you release.

 

the tools in question are my game dev library and cscript macro processor.  both are generic in nature.

 

past versions of the library have been a retail product.

 

The guys who made diamondware sdk were interested in Cscript, but i was too busy with SIMSpace at the time to pursue it.

 

i'm considering releasing the library with full source, and releasing the cscript macro processor exe. i'm not sure about releasing the CScript source code, as that might lead to multiple versions of the scripting language it uses. I think i'd like to retain control of the language definition -  at least for now. but the cscript source is pitifully simple, a tokenizer, and a big switch statement based on the first token in a line. the exe would include the language definition in the docs, so anyone could write their own cscript compiler if they wanted to.

 

sounds like maybe i should release the tools.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this