Welcome to Jester's Trek.
I'm your host, Jester. I've been an EVE Online player for about six years. One of my four mains is Ripard Teg, pictured at left. Sadly, I've succumbed to "bittervet" disease, but I'm wandering the New Eden landscape (and from time to time, the MMO landscape) in search of a cure.
You can follow along, if you want...

Monday, April 25, 2011

Quote of the Week: Not invented here

I'm a big fan of snark, and this week's Quote of the Week, from Herschel Yamamoto on Failheap Challenge, contains a ton of it:
I don't give a shit if they download their engine from the Matrix. Their job is to give us as much cool content as they can per unit dev time, not to sperg about doing something the hard way. You can bore a tunnel with a spoon, and it's really hard, but that doesn't make you a better person for doing it, it makes you a goddamn retard. CCP has the same problem. Given the choice between 72000 hours on Evegategate and a man-week on integrating VBulletin, they always choose to do it the hard way for no good reason. Don't defend not-invented-here :psyccp: stupidity.
Despite being directly mentioned, the topic at hand wasn't the forum mess.  It was the Captain's Quarters test on Duality.  It's clear that everything in it was built pretty much from scratch by CCP.

I spent a good bit of time kicking the CQ around over the weekend, and there's some very good news here: CQ doesn't appear to break anything that we've become used to.  The 30 second and 60 second timers that I mentioned in my blog entry last week are intact.  Despite being "non-optimized" code, the balcony environment seems to load slightly faster than the current hangar environment.  And your menu is right there where you expect it to be.  You can look at your toon while you select the Fittings, Items, or Ships buttons, or the Station Information window for your agents, do what you have to do, and get un-docked again in a twinkling.

So a definite tip of the hat to CCP there.  Rage avoided.  ;-)

The rest, though?  I'm going to have to agree with Herschel.  There's nothing revolutionary or ground-breaking here.  Certainly, there is not enough to justify the effort, time, and money that's been put into it.  CCP is indeed digging a tunnel with spoons, and it shows.  There were so many bugs in the display and function, that I simply couldn't list them all.  Even more than that, the functionality is... clunky.  And I don't think it's because of non-optimized code.  I think the final version is also going to feel clunky, too.  The CQ comes off as a fourth-year CompSci student's attempt to recreate the Unreal engine.  It works.  It would even get a good grade on the final.  But it would leave a true pro tsk'ing at the student's many mistakes and foibles.

And it would leave the other CompSci students grinning and wondering why the nerdy kid spent 650 hours on his project to get an "A" when they spent 120 hours on theirs to get the same "A".

Herschel brings up the expression: not-invented-here.  NIH, it's called, or Not Invented Here Disease.  Those afflicted with this disease will go to the ends of the earth to reinvent the wheel when it's completely unnecessary.  They can't stand to have other people's work in their masterpiece.  CCP has NIH disease in a big, big way.  The forums show this off, of course, and now CQ, but so does the massive amount of work they're doing right now to implement security around the API.  There are licensable products that would let CCP skip all this work, but Heaven forfend they actually use them.

And given how CCP's reinvention of the wheel tends to be second-string work, this leaves me with real fears about their new version of user authentication using those hardware security tokens they distributed at Fanfest.  Are they reinventing the wheel there, too?

It truly makes me wonder how much further along Incarna would be if CCP had just licensed the Unreal engine and started building Incarna off that three years ago.  Had CCP stuck to writing code for their game instead of writing code for the underlying engine of the game, would we be walking in stations already?

I'll probably have more to say about this broader topic in a week or so, because it's important, but I want to collect my thoughts a little.

9 comments:

  1. "Had CCP stuck to writing code for their game instead of writing code for the underlying engine of the game, would we be walking in stations already?"

    Yes.

    ReplyDelete
  2. Actually, maybe not. From my experience with it, Unreal isn't exactly the friendliest beast. It's not exactly a bolt-on. CCP already has an entire engine for their space stuff. Adding Unreal means adding an entirely separate engine, with its own asset pipeline and management structure. If they had planned to do that from the very beginning -- the very, very beginning -- then perhaps. But the more I think about it, the less I think just bolting Unreal onto EVE would work.

    ReplyDelete
  3. Aren't they using Unreal for Dust though? I would assume there would be some cross integration of assets for that project...which might explain why we haven't seen much about the game yet. :P Also might explain why CCP has gone the spoon route with QC.

    ReplyDelete
  4. CCP did not create the 'new' EVE forums from scratch. They took an open-source forum framework and crudely modified it to work with the existing account/API infrastructure.

    Also, I doubt that licensing the Unreal engine, or any other third party engine, and attempting to make it work gracefully with the existing EVE code would have been a good idea. As has been noted above, you can't just put two different game engines together and switch between them on the fly. Furthermore, third-party engines might not have offered the functionality which CCP wanted; for example, I don't think any off the shelf engine offers the degree of character customization available in the new character generator.

    That said, I'm sure we'd actually be walking in stations had CCP started work three years ago using a third-party engine. However, odds are that the final product would not be nearly as good as the current version of Incarna has a chance of being.

    ReplyDelete
  5. Ripard, I just wanted to throw out that this is my favorite Eve Online blog. Second and third go to Fiddler's Edge and Ninveah, respectively.

    Anyway, on to the topic at hand. They need Carbon for World of Darkness. If CCP would learn how to correctly polish features, I think Incarna could be one of the best and most realistic avatar engines on the market. Sadly, I'm not too confident in their ability to do that.

    ReplyDelete
  6. Given that Carbon is the superstructure underlying all current and planned games from CCP, it presumably provides advantages in maintainability, fast creation of dev tools customized for individual games, vastly lowered learning curve when moving devs from one game project to another, and development synergies between game projects. (These potential advantages would apply much more cleanly if they didn't have to deal with Eve's badly written, poorly documented old code that no one left at CCP has a clue about.) I doubt CCP could have accomplished all they hope to with Carbon by licensing third party software.

    And then there is the revenue potential from packaging Carbon up as an offering to other development studios. CCP told CSM5 they had no plans to do that but if it turned out to be lucrative enough or their investors forced their hand, I imagine it could happen.

    ReplyDelete
  7. Don't get TOO focused on Unreal, everyone. It was just an example of a licenseable engine. My main point was on CCP insisting on doing everything the hard way, when there are viable alternatives to doing it that way. ;-)

    ReplyDelete
  8. I think its harder to innovate with a licensed engine. I don't believe we are necessarily missing out on anything, this WiS stuff is kind of a bonus. I think I would actually be disappointed if CCP had licensed out the Quake engine or something for this.

    ReplyDelete
  9. It may be a simple case of 'build or buy' from CCP.

    If they bought an engine similar to unreal engine then it may not be optimised for the MMO environment or it may simply be too expensive as I would imagine any deployment of the engine would require a licence cost to be paid to the supplier.

    I would imagine that CCP plans to use and expand upon the engine themselves for future MMOs either internally (world of darkness, Dust etc) or externally if they plan to sell or licence the engine to other MMO developers.

    Does it make it any less frustrating? No not really.

    Either way I can't think of a better group of players to give it a 'trial by fire' than the EVE players :D

    ReplyDelete

Note: Only a member of this blog may post a comment.