MMO Development Lesson #10
Be careful when “planning ahead” technologically. It’s no secret that epic MMOs take a long time to make–longer than all other genres, on average. So, be extremely careful when you decide to base your system requirements on your predictions of technology’s future. The reality is, technology changes at a variable pace. For example, determining that a 4.5GHz PC with 2GB of RAM and a GeForce 9 series card will be the standard in 3 years is, at best, a potential misjudgment and, at worst, completely ignorant and erroneous. If you want your game to be accessible to the widest range of people, you need the broadest range of system requirements, which leads to the lesson for next week. 
Maybe its a good idea to develop the game for DirectX15 so microsoft can use it to sell Windows2011?
[...] MMO Development Lesson #10 [...]
Game design should be, from the ground up, scalable, whenever possible. You should design for a “reasonable” rig spec right now… go to Dell, Gateway, HP… spec out a $2000 PC (with monitor) and design for that. In 3 years, that is likely to be standard or less than standard with the way PCs trend.
It would be nice if a game company made a game engine that looked great at the required specs and only got better as you turned on more features. Sadly, most games look great with all the features turned on and start looking unfinished or crappy as you turn features off.
I remember bringing up the point of how EQ2 seemed designed for a spec that never materialized (kinda like what Ryan posted) in the official EQ2 forums. The game really seems to depend on CPU horsepower more than the brawn of the graphics card. The CPU manufacturers, however, decided to go the multi-core route and give the GHz race a rest for a while.
I got flamed for that post, stating that I obviously knew nothing about software development, that there was no way to predict what was coming in the future, so you code for what you expect to have, or code for what you have now and produce something that looks not as spectacular as it could (with WoW being given as the example of that).
I guess I should say that more clearly… games should be designed to scale up… not scale down. Scaling down almost never works because it means you started with a product that won’t run and are trying to make it run… if you go with a scale up model, you make something that runs and will run better if more is available.
I maintain that WoW’s graphics are the reason it is so popular and the reason it’s not seen as “nerdy”.
I’m more partial to the graphics they [Blizzard] built into the Outlands than the original stuff… much less cartoony — although cartoony is fun. If a developer would come out with an MMO which had graphics along the lines of what we’ll hopefully see in the FPS/RPG Project Offset (if it ever hits the shelves) or the Crysis engine, I would be completely enthralled.
I’m also a big fan of scalable FX — it’s too bad that you couldn’t actually base the game dynamics themselves on the capabilities of something like the physics add-in card (forgot the name of it), rather than just cooler FX (e.g. realistic knock-back is only cool if it applies to every player).
Planning for expansion isn’t just a client-side issue, either. There’s all of the tasty provider infrastructure to consider. [Go go gadget grid?] It’s important to come up with a capital investment plan taking into account server growth vs. client growth, distributed content delivery (if applicable), obsolete servers, etc . — everyone in the biz & several of those who aren’t could probably tell you this if asked outright, but thinking about “scalability” is more than a 45 minute meeting one afternoon. It’s almost a project in itself (and very much appropos in this topic of “planning ahead for technology” I think).
How many examples can we think of where “a potential misjudgment and, at worst, completely ignorant and erroneous” decision (either in client specs, or in infrastructure) resulted in a lackluster rollout or “failed” product/company?
I agree I think designing for what is the best now if your game isn’t going to be out for a few years would be the best idea as it will probably be the low in when the game does release, but with technology always changing it does seem like it is sort of like roulette.
Why not design on what the best is now but in that design process add in the ability to upgrade the game engine with newer graphics engines/technologies. Look at EQ1 when it came out it was pretty cool for the time it came out as time went on people wanted it to look better so they kind of upped the engine and made the luclin models. which made it look a little better… why not plan for that and let the game grow as technology grows?
Of course I could just be talking out my rear :p
EQ is a decent example… except that the Luclin models were so ugly!
It’s absolutely a good idea to plan ahead an know that you will be upgrading the visuals as time goes on if:
1) Your game is intended to go on for more than a few years without any type of replacement (unlike Guild Wars, for example).
2) Your game does not have timeless graphics (unlike World of Warcraft, for example).
A game like Vanguard should plan to upgrade visuals as time goes on if those two rules are followed.
The best for whom, though? Based on the system specs of the programmer? The modeler/artist? The end user?
I think there is a very high danger in building your engine (and this is doubly so for an MMO engine!) around the idea that you’ll be upgrading the core graphics pipeline/rendering system as time goes on. The planning going into this (what happens to those who don’t upgrade? What happens to those who meet sys. req. upon release, but first patch kills their ability to play? What happens if there is a missed bug (never happens, of course
))? You just shut out every player who has been patched) type of event would just be maddening. It can be done, of course, but does the end justify the journey?
Upgrading the visuals is one thing (if you structure your rendering system right to begin with), but when you start upgrading your rendering system during the middle of your games life cycle, I think you just may be asking for unneeded headaches.
But then again, I too may be talking out of MY rear.
at worst, completely ignorant and erroneous. If you want your game to be accessible to the widest range of people, you need the broadest range of system requirements, which leads to the lesson for next week. [IMG
] posted by Ryan Shwayder @ 22:00 Comments Off
How about the idea approached something like this…
Create and design a game for what the standard (high end) gaming rig is now. Create a steady timeline/story for a few expansions over about 2-3 years then instead of upping the engine or adding more expansion plan for a sequal to the game with total revamp of game to update graphics / playability improvements. Much like the FF series. The story progresses but so do the graphics etc. But with the sequals have each player that maxed levels/skills or what ever system you had in place in the previous game able to roll thier character over into the new game with added benefits thus making it feel like natural progression.
Does that makes sense?
One problem games always have to contend with as time goes on is if they want to honor the rule of “dance with the date you came with”.
City of Heroes ran fine on my PC… when City of Villains came out, they wanted to upgrade the engine for CoV, but needed to apply it to CoH as well… I didn’t buy CoV at the time, but I stopped being able to play CoH. I cancelled right then and only came back recently after buying a new PC. They lost me as a customer for the time in between.
Yeah I see your point Jason. Well if the graphics upgrades are purely cosmetic could the graphics upgrades only be applied if you bought the expansion much like it was done with the Luclin expansion? Only problem I see with that would be when adding armor sets etc in you would have to make sure it fit all different character models old and the like.
This one makes sence “One’s first step in wisdom is to kuesstion everything – and one’s last is to come to terms with everything.”