14 replies to this topic

[Anddos]

I like the idea of auto downloading patchs but how about a choice to play the newer version or stick with the old, that way you can patch it after you know other's have tested it for bugs etc, also you would be able to play all the older versions to , so there would be like an archieve of game versions to play, what do you think?

[jbadams]

This might be good if you've managed to make a change that the majority of your player-base doesn't like -- but in general, I think it would more likely just lead to a fractured player-base all playing different versions of the game and therefore having less potential people to play against in a match-making environment.


Consider for example that you have a total player-base of 100 players.
If you apply patches for all users, then any individual player who connects to your match-making server looking for a game to play has 99 potential opponents.
Now, if you have instead allowed players to choose which version to play with, and have released 10 patches that are equally popular, there are suddenly only 9 potential opponents for your player.

Obviously this is a pretty contrived example, but I think it is a genuine potential problem with the idea.

[jefferytitan]

Yeah, if it's just a graphics patch maybe. If it's a DLC that offers new areas, that can be dealt with (although it partially segregates users as mentioned above). But if gameplay/physics/network/ai are involved, it's generally not plausible for a player with version x.0 to play with a player who has version x.1.

[Stormynature]

I think in a single player game, establishing an archive for at least the previous version patch would be sensible. As an example Skyrim under Steam ran into a number of issues with at least two different patch implementations requiring a large amount of players to rollback their game to previous versions in order to be able to continue game play while waiting for the patches to be fixed.

[jbadams]

Stormynature, on 23 February 2012 - 08:38 PM, said:

I think in a single player game, establishing an archive for at least the previous version patch would be sensible.

That's an excellent point -- the problems I mentioned above do not apply to a single-player experience, and the option to easily roll back game updates might be welcome in some situations.

[Anddos]

well the game would be able to match make people against diferent versions and not just the latest, what if you want to see how far a game has come?, you would first load the very first version play it and then patch , so you can see the diferences of development as the game progressed if you was a late comer to the game.

[Stormynature]

This almost sounds like phasing patches (or expanded DLC's) i.e. You play the game in it's original format complete whatever goals etc the game is set to then download the next patch which then alters the world at which point you play the new version to the goals set and so on. If you are talking literal content change like that then how do you co-op with someone for example when you have reached a point in the game where the city you are in has been devastated and the other player is still in the early stages of an alien invasion. If on the otherhand you did matchmaking based on version installed - How would you deal with diminishing player levels at the front end of the game i.e. new gamers would struggle to find others for co-ops. If you are talking more on the scale of the introduction of advanced weaponry or similar gear aspects then why wouldn't I want the best immediately.

From a games development point of view i.e. learning how to make games it might carry a certain amount of cachet but I struggle to understand your concept fully I think. More specifics on what is in a patch implementation would be appreciated.


I must admit though your last post compared to your original post has kind of thrown me for a loop. Now I am not sure whether you were initally addressing a technical issue or a content issue.

[Acharis]

A patch should be a better than the previous one, there should be no reason to not download it. If it introduces a new mechanic (like a new set of units) then add an option to disable it (treat it like a mod). There should be no players playing different versions if possible (especially if they play multiplayer, in such case they should be forced to upgrade to the newest version before they can play with others).

Yes, there will be a small portion of the player base that will be unhappy because the patch changed some minor thing that they liked previously better, but they have to deal with it. We can't make separate games adjusted for each player...

[jbadams]

Acharis, on 24 February 2012 - 05:08 AM, said:

We can't make separate games adjusted for each player...

Can't we?

[Acharis]

jbadams, on 24 February 2012 - 05:28 AM, said:

Acharis, on 24 February 2012 - 05:08 AM, said:

We can't make separate games adjusted for each player...

Can't we?

We can't We have no budget for this And even if we had the budget for this it would me much better to spend it on something else instead. Well, if our pricing strategy is like 1,000$ per unit then it might look different then

[jbadams]

Acharis, on 24 February 2012 - 05:52 AM, said:

We can't We have no budget for this

I know that's the generally accepted answer, and that budget and developer time are both very good reasons for it... but sometimes good ideas can arise out of challenging commonly accepted ideas like "we can't make separate games adjusted for each player". I don't personally have any good solutions for doing so, but can you imagine how successful a game might be if they found a way to do so that was cost effective? I imagine the solution would lie with some combination of a very powerful procedural content generation system, dynamic difficulty adjustment, and a very effective analysis of how the player interacts with the game.

This isn't something I think would necessarily be wise to pursue, but the possibility is certainly an interesting one.

[Stormynature]

Okay have spent some time thinking about this (no I haven't).

One way in which you might make this idea work would be through the use of achievements and rewards specific to each version. As a loose example from World of Warcraft MMORPG the vanilla (first) version had a questline (the opening of AQ) which was achieved by very few people in game. In turn those who completed the quest chain at that time earned themselves a title and a mount. As this was basically a one time event per server and disappeared under later expansion releases, it became somewhat of an exclusive item ingame. The other aspect of WoW with subsequent expansions was the development of Vanilla guilds i.e. people who did not want to play at the next expansion level as well as developing culture that talked about the good old days in vanilla or Burning Crusade etc.

The ability to do rollbacks could conceivably work in such an example and in part has been effectively shown in part by the refusal of some to upgrade to later expansions. Given that existing resources were used to service both the vanilla community as well subsequent expansions, the idea might actually be less severe to a budget than immediately anticipated in previous posts. The larger issue would be the rollback itself without having to do a full re-install to access earlier versions. Given that we can patch forward though on original content -- It should not be that difficult to implement a rollback patch version. I am of course referring to new product I wouldn't want to untangle some of the spaghetti of existing products. I am skipping over a lot of the grittier technical issues that would make this annoying though.

[Stroppy Katamari]

It's good to give the player the choice to play an earlier version. Your multiplayer (if any) should check version anyway.

Starcraft II simultaneously contains all the previous versions; they are automatically used to play back replay files recorded in that version.

[Stroppy Katamari]

Acharis, on 24 February 2012 - 05:08 AM, said:

A patch should be a better than the previous one, there should be no reason to not download it. If it introduces a new mechanic (like a new set of units) then add an option to disable it (treat it like a mod).

And if the patch adjusts 100 different build times, resource costs and invisible things behind the scenes which don't have a name but change how the units' pathfinding feels... checkbox for each?
No, that's terrible. If it's technically feasible, just allow older versions to be played. This is likely already necessary if the game has recording like Starcraft does.

Quote

Yes, there will be a small portion of the player base that will be unhappy because the patch changed some minor thing that they liked previously better, but they have to deal with it. We can't make separate games adjusted for each player...

No one's talking about adjusting the game for each player, just letting them play versions which already exist.

[SimonForsman]

jbadams, on 24 February 2012 - 08:38 PM, said:

Acharis, on 24 February 2012 - 05:52 AM, said:

We can't We have no budget for this

I know that's the generally accepted answer, and that budget and developer time are both very good reasons for it... but sometimes good ideas can arise out of challenging commonly accepted ideas like "we can't make separate games adjusted for each player". I don't personally have any good solutions for doing so, but can you imagine how successful a game might be if they found a way to do so that was cost effective? I imagine the solution would lie with some combination of a very powerful procedural content generation system, dynamic difficulty adjustment, and a very effective analysis of how the player interacts with the game.

This isn't something I think would necessarily be wise to pursue, but the possibility is certainly an interesting one.

I think the best way to do this is to make the game easy to mod, (This goes for both singleplayer and multiplayer games), Look at games like quake, starcraft, skyrim, etc, For multiplayer games things become a bit more limited ofcourse(unless you allow proper mods) but a few things can easily be controlled by the server in a multiplayer game, (for an FPS you can let the server change things such as weapon stats, gravity, movement speeds, etc quite easily using for example scripts)

If you provide the tools then the players themselves will take care of the rest.

In the OPs case the easiest way would probably be to separate the engine from the game cleanly, make engine updates mandatory but stick with a fixed interface towards the game and then simply treat the game and mods in exactly the same way, older versions of the game can then be loaded as mods if players prefer a previous versions gameplay. (You could also release the source for the game portion under a restrictive license to allow the community to tweek it as they see fit)