What is the best game engine to use for a mmrpg? What is the easiest to use?

Bigworld! www.bigworldgames.com
However, it would, i''m sure, cost an arm and a leg to license it

The one that you make yourself... :D

Don''t license one. They will gut you in royalties. Think that 8% isn''t that big? That basically means that they own between 25 and 50% of your company because that''s what percentage of what would be the profits they will be getting. Might as well get your game done and convince a bank to fund a buyout of them.

solinear, i guess you refer to Quazal''s royalty for Eterna. Their rationale is to take money out of the datacenter''s pocket by reducing server-side bandwidth. A lot of flexibility is built into it to decentralize most real-time data updates while keeping client/server for security-sensitive attributes... If this work, then they split 50-50 the cost reduction, hence about 8% royalty. Personnaly, I prefer to licence tested middleware than do it myself...

BTW, they are open to negociate a fixed price if they fail to deliver on their promise!

Actually I wasn''t thinking of anyone in particular, it seems to be a pretty regular percentage though, with some asking for more (closer to 10%) and others asking for less (down to around 5%).

I can understand why some companies might want to use middleware, I just don''t like the idea that another industrious team might be essentially working on nearly the exact same product as you are, with the exact same engine. Plus the issues about royalties being a large percentage of your overally profits, if not closing in on 50% of them.

I outlined MMOG costs somewhere else and I think that anyone who actually does the research will realize that bandwidth costs are not so significant as to really warrant giving away a relatively large percentage of the company though. Bandwidth costs are usually less than 10-15% of your overall costs (personnel being the vast majority of them, close to 65%, with insurance, software, equipment and lease taking up the last 20-25% of your costs).

I''m not saying that it''s not worth it though. Middleware, if it meets the needs of your product can shave literally half of the time off your development. If it doesn''t come close, you''re stuck with paying someone for something that you don''t actually end up using because it''s too far from what you need and you''ve probably burned a month or so evaluating it''s capabilities, so now you''re a month further behind.

If it fits what you need, go for it, if it doesn''t... well, I think you can figure out what to do.

Actually, considering the costs of customer service reps, you might be better off spending some money on a CRM product.

I agree with you, most MMPRPG are spending about a third of the subscription fees to maintain the datacenter. If you look closely at some MMP middleware, you see that they:

a) increase user count/server (Zona)
b) increase the efficiency of the backend database (RebelArts)
c) decentralize the game state (Quazal)
d) provide an integrated solution (BigWorld)

In my own evaluation, I realistically mesured how much performance my system would achieve compared to off-the-shelf components. I concluded I prefer to give some money to a middleware companies, money that I would have spend anyway on bandwidth and/or on building my server-side infrastructure.

The decision to choose middleware or not is very dependant on your own convinction. Myself, I don''t mind using the same engines as other studios, it just motivates me to differenciate myself on content rather than technical details. With the recent success of Tony Hawk, GTA and Dark Age of Camelot, middleware vendors seems to get it right. If you take the time to talk to them, you''ll see that you get a partner, not a piece of code... This is good for 3D, physic, AI, network, sound, etc. Of course they all want royalty, but they want our games badly. At the end of the day, it is a matter of negociating and you can get the best balance between tested technology and royalties. And instead of spending a month to figure out their offering, I was able to get a phone meeting with a developer followed by an on-site bootstraping seance, free of charge...

Also don''t forget other successful games such as Halflife, which effectively used the Q2(?) engine license as middleware. But it is very obviously not Quake 2!!!

Even more amazing is the fact that Half-Life was actually Quake 1. Quite the impovement indeed.

My bad, didn''t quake 2 use the quake 1 engine or something anyway?

quote:

---

With the recent success of Tony Hawk, GTA and Dark Age of Camelot, middleware vendors seems to get it right

---

I think that there is something that all of these have in common though... a large part of what they get is the display interfaces.

DAoC used NetImmerse (or whatever it''s called) and for a UI in an MMOG, it''s perfect to use middleware. Considering the fact that the graphics are one of the highest cost portions of any game (what, you''ve got like 4-8 programmers and 10-14 graphic artists of different types?) and if you can cut that down so that they''re being more productive, it''s a major savings in your overall cost of the game.

quote:

---

a) increase user count/server (Zona)

---

Not necessarily the most important thing, then again, I don''t know how they''re doing it. If you can increase customer/server load by 100% or more, you''re saving major cash, but there are actually better ways to deal with this. I''ll cover this more near the end.

quote:

---

b) increase the efficiency of the backend database (RebelArts)

---

Not sure how important this is. In most situations, each zone/area server is going to have a copy of the majority of the local character (PC and NPC) information, so hitting it all on the main db server is usually not a necessity to be blindingly fast.

quote:

---

c) decentralize the game state (Quazal)

---

This can be very important, though I''m not really sure what they truly mean by ''decentralizing'' the game state. EQ is technically decentralized, each area is on a different server.

quote:

---

d) provide an integrated solution (BigWorld)

---

From what I''ve seen, they may actually have one of the best solutions out there. I am really intrigued by what they are doing. I''m suspecting that they are doing something similar to what our team is doing, though I''m not really privy to the inner workings of their engine.

I have a strange feeling that all but ''b'' is actually the same thing. There are certain methods for handling things that actually allow you to do all of those at the same time and oddly enough, they are completely dependent on doing all at once, at least in the case of what our team is doing.

Either way it''s all about your priorities. If getting your product to market sooner is the most important thing for you, then middleware is the only way to go. If making sure that the product is perfectly suited to your purposes is most important and maximizing return on investment, building your own product is the best way to go.

There are benefits to each and bad points to each. If your product averages 150k users/month for 2 years, then you''ll have an income of $75Million over the course of those two years. You can expect to earn about 20-25% profit ratio, so around $20 Million of that will be profit. If you used a middleware product you have to ask yourself whether it was worth $6-8 Million. If you doubt the skill of your programmers to build the product well from the ground up (you don''t have to state that you don''t... we all know that it''s a VERY big project for the uninitiated in server programming), then it probably is.

The problem comes in when you decide to go with 2 different middleware products, one for client and another for the server. You could easily end up with giving away 15% of your income (not profit, income). This means that your company is getting 5% profit for having done 70% of the work. All of a sudden, your $4 Million investment is returning only $4 Million. You''re barely breaking even while the middleware producers are making money hand over fist.

Now go back and think that your game is the next EQ... you''re getting something on the order of an average income of $135 Million over 3 years. You''re also giving away around $13 Million of that to the middleware developer. Yes, they may be a ''partner'', but is that partnership, even if they devote 5 programmers to working with you every day, worth $13 Million? Did they really do so much that they deserve that much money? Oh yeah, you bought a middleware client too and are giving them 7%, so there goes another $9 Million. They didn''t make your graphics, game logic, none of that stuff, but they''ve probably made more than 3 times as much off your game than it cost to build the middleware in the first place and now they can advertise that your product is built on their engine, thus selling more of their engine and probably helping your competition. Do they care if you go down if it''s done by a company that is paying them a higher royalty? Heck, they want that company to knock you down if it means that they''ll get more money.

My best advice is to make sure that you put a cap on the royalties/quarter. Otherwise you''ll find that the company that you got your middleware is making more money off your product and continued investment than you are.

Wow... I need to try and keep my posts shorter and not go into so much detail. I probably just killed another thread by becoming the demon of detail.