We've already had this discussion. These are all extreme outliers, AND none of them even existed until their host game had sold millions of copies-- about 1.5 million for Quake if you're talking about the original TeamFortress, and around 6-7 million copies for Half-Life in the case of DoD and CS. I have zero information on WC3/DotA, but it's a safe bet that WC3 had already sold a few million copies by the time DotA exploded. With all of these, any potential sales driven by the mod itself (which is something more than slightly difficult to determine) were significant, but still dwarfed massively by the pre-mod sales of the host game - even if you make the ill-informed assumption that every sale of the game after the mod has hit 'mainstream' is a result of that mod.

The standalone Counterstrike box is the only case of a mod that moved serious volume (~4 million copies to Half-Life's 10 million), and even then only because it was ultimately a retail product. No version of the CS mod had anywhere near one million plus downloads, and you of course have to account for repeat downloads (which are quite common) in those kinds of statistics.



FA sold well over a million copies as of 2008. There's no exact number posted online, but 1.3-1.5mil is a safe bet. Even if you add up every download for all of the most popular FA mods on every site they've ever been hosted on, including across versions (which would reduce the number of unique downloads very significantly) it will not even come close to 5% of 1 million copies.

If this is true, then THQ is a true ******* publisher by even not releasing the last patch.

You cant say that without being certain they made a lot of money off FA. AFAIK, FA wasnt exactly a gold mine either.

It is really difficult to say without knowing the costs and everything when a game is successful.

For example, 1 million copies for a RTS sounds good to me... but there were around 125-150? people working on it at GPG?

Sins of a Solar Empire is also known as a big success, but they only had a few people working on it and a very small budget.

I guess Supreme Commander 2 was a bigger commercial success cause they had less people working on it and a very brief development cycle. Even if they may have sold less copies than FA or SC1...

I really would like to know what sales number THQ would consider for SC1/FA for calling it a successful RTS?

Most big-budget games, even PC-only games, sell over a million copies now. Even SC2 has probably exceeded a million by now. That doesn't mean they're wildly profitable, nor that pouring more money into patches once sales have died off will move more copies.

Please note that I'm not arguing against a publisher supporting their game nor against adding mod support. There are a few examples where a publisher/developer has gone all-out with continued patching and mod support and had fantastic results, but it's naive to assume that 'one more patch' or 'just add a mod manager' will somehow have that effect on a game like FA or SC2.

Obviously patching it in now will have little effect. Considering all the work had been done for supcom, they were fools for not including it.

I still think you're pulling figures outta your arse.

All of these numbers are easily available online. But I also remember the rough sales numbers for Quake and Half-Life when their mods appeared (I'm oooold), because both games were setting sales records for PC games at the time. Which is why the mods were ever a thing in the first place.

Yes i'm old too however I had the courtesy of linking you to my limited source of information.

Look you could be correct, but i'm not going to take your word for it.

Unless you have a better source than google for old news articles that listed copies sold at date of print, it's impossible to get exact numbers for any point in the past. I can't find a single article for any game that I mentioned older than about 2004, and unsurprisingly their sales numbers haven't changed much if at all between 2004 and the present. Nothing on the internet sticks around that long unless someone purposefully archives it.

Of course, mod download statistics are even more fleeting, and highly inaccurate to boot. The only situation in which a game's mod downloads could be properly counted is if the game required all mods to be distributed via a GPGnet-like service. I'd say Vault download statistics that account for unique users would probably be a lot more accurate than trying to conglomerate download counts for popular mods from various file hosting sites, but we're not likely to get those.

It's of course worth noting that CT said that SupCom & FA were $11m to develop. Crackdown for 360 sold 1.5million copies, but they still only broke even on the game.

Mithy is completely right- large mods only ever appeared for games which were in as of themselves huge releases. Name a single modding project for a game where the game was not absolutely huge beforehand- that is, no Valve games, no Blizzard games, no Epic games, no iD games, at the very least. The "era" of CS/TF/DoTA? That never actually happened for anyone that was not a ridiculously huge force to begin with. And, of course, those mods were written in C/++. Not Lua. And there's good reason for that, which is that Lua, quite frankly, absolutely sucks, and it's a fairly huge waste of time.

I'm also pretty sure that the game sold over a million. If you want to check yourself, you can take the fact that CT said that less than 10% of all SupCom and FA buyers ever even tried to register on GPGNet, and compare that to the registered accounts- which total about 150k, from memory, but they publish the number on the Welcome page. That puts the game at about 1.5million sales.

Also, I don't think FA ever retailed at full price (aka $50). So 1.5 million sales might seem like a lot, but 1.5 million at $10 a game is not. I don't know the details, but for such a large project like SupCom1 I'd like to have a hefty profit margin, like 40%. And most of a game's revenue is generated within the first 6 or so months. So THQ is not wrong or bad in not releasing a patch, despite what jetsnguns might think. FA apparently did not sell that well in its "sales period" and thus does not justify a lot of support. And most games don't get support after about a year after release, so the window for FA patches has long since closed.

And unless some DLC is released for SupCom2, there might only be 1 or at most 2 more patches for SupCom2. March 2, 2011 is closing in.

Ok so we agree then that none of us know the answer. All your conjecture is useless to me.

Moving on.

The fact that games see the vast majority of their sales within the first 6 months to a year is not conjecture, it's fact for all but a tiny minority of high-profile releases - and even those still see a massive drop in sales after the first year or two, mods or no.

If you expect links with verified, authoritative info on game sales figures, you'll probably continue to be disappointed, because there is no solid central source for this kind of information. It trickles in as offhand mentions in interviews or press releases, and occasionally a company will issue a statement with overall sales for a particular series or franchise 2-5 years after it's ended. If you don't believe me, try to find solid sales info on a ridiculously popular game like Half-Life other than the big 10-year anniversary press release that valve sent out three years ago.

Oh no I believe you. I have tried to find this info many times in the past. However you talk as if you know what you're on about, whilst admitting the information is impossible to find.

Conjecture doesn't stand up in court, in research or even in university. Likewise it proves nothing here.

There is no way to determine the increased sales from mod support.

This isn't court, research, or university, so I don't really care about any of those. I've seen what I've seen and if you don't believe it, then I really don't care, so you can go ahead and retreat if you like.

I agree completely, even if we had perfectly accurate graphs of sales data for these games and accurate graphs of total unique mod downloads. This is exactly why only a very financially-secure company like Valve or Blizzard can take on the huge risk of pouring money into mod support: because it links any potential return on your investment to a fickle community (whose size is dependent upon the original sales of the game) that may not ever produce anything of note to theoretically drive further sales.

You can hopefully understand why it's that much safer to invest in creating your own DLC or expansions and relying on the limited but more assured profits from those instead of trying to nebulously re-invigorate the sales of the base game with free or user-created content.

Even in the unlikely event that the mod community does take off in a huge way (something that has only happened for what, two games to date?), the potential profits from the then-reduced price of the game driven by mod interest are going to be dwarfed by the game's initial full-price sales. Valve was extremely fortunate and clever and took its popular mod properties to retail (effectively returning the original game to full price for anyone without the game who was interested in the mod), and probably made a ridiculous amount of money that it wouldn't have had it passively used those mods as a marketing tool for the original game.

Hmm.. I'm not going to agree on that, DMG. Why LUA absolutely sucks and it's a huge waste of time?. C++ mods are for pros only. Friendly modding must go with higher coding levels, IMO. At least as a basis and after that, if you allow C/++ or whatever alike to create more deeper mods then the better.

For instance, and I'm not breaking any NDA here on what I'm going to tell ya, the .... never mind.

Ok, I tell ya. Barcelona, Spain Olympic games 1992 project under OS/2 platform. We were mostly using Model Data Language (MDL), which was a pre-processed code in between C and Cobol. We were coding in pure C as well, but MDL was created in order to make things easier to the old Cobol crowd who was working in the project. Just to mention one on here. Just my 2 cents.

Firstly, because interoperation between Lua and C++ is an incredibly disgusting process. It's difficult and exceedingly repetitive to code, and managing the memory and making sure it all cleans up is an incredible bitch. And after all that's done, the language itself is no great shakes either- extremely limited control.

On the other hand, exposing a couple of native entry points and providing a couple headers is relatively trivial- you just post some internal headers and don't need to worry.

More importantly, native mods have The Power™ to do what they like. If you allow for a native mod, then you also allow for Lua mods if someone who knows a tiny bit of C++ cares to put in the ridiculous effort, and you allow managed code (.NET) and such mods too, and those mods have normal application power. You want to add automatic mod transfer? You can. You want to stream live to a server? You can. You want to replace GPGNet? You can. FA mods can do none of those- and trust me, I've been there. Whereas in Lua, your options are limited (to say the least) by the language and pre-provided libraries.

Finally, Lua itself as a language just sucks. It's slow, it's object-orientation is primitive at best, and all the dynamicness of it decreases the programmer ability, not increases it. The quantity of work that the C++ compiler does to support the developer is extremely large, but Lua offers none such. There's no actual advantage to it over C++ except that it is interpreted at run-time. That's fine for rapid prototyping, but mods can be compiled by the author in advance and don't need that.

Actually it's much more than that. 360,428 as of this moment. And it increases about 10-20 per day it seems lately.

Yes and no. I can't think of a game offhand that allowed this level of freedom with its extension language. Compiled scripting languages like UnrealScript and QuakeC are definitely far more powerful than moho's Lua, but both still put limits on what you can do within the framework of the game. Allowing no-holds-barred compiled C++ libraries in mods would be cool, but would also be pretty risky, since there'd be few limits on what that code could do.

Plot is about to thicken? http://twitter.com/#!/merobbins/status/ ... 7017857024

"Big announcement later today. Can't wait."

"Big", hmm. SupCom 3 after all?

Pretty sure that the Source engine's mods are in C++. ICBWT. And, there are no big mods for Unreal- it's widely used professionally, but I don't know of any mods for any Unreal games that have been big. And Half-Life the original too- "The most highly changed sections are the game logic; ours being written in C++ and Quake's being in written interpreted "Quake C"." from Wiki.

Oh no, Sorian's Skynet is about to be released... :3

Age of empires online. KNK may actually be on hold, they cant make TWO and compete with themselves can they?

Most work on AoE Online was already done, like MS say on their blog it's about making (paid) content and race packs, they've already questioned beta users about how much they would be willing to pay for content packs and what extra content they would like to see.

AoE Online isn't bad (it really isn't) but I don't see how it's gonna succeed when CoH Online had to close. It also uses GfWLive which is a bit of a bummer.

Have to say I am disappointed as to me this very much sounds like a less ambitious project. I wonder how long the free-to-play bubble will exist before it explodes.