From some older post I've read, Xbox Live is not supported by XNA. Is it still blocked? I wanted to add leaderboard support to my game if possible over Xbox Live. I have my own web server to do this, but is networking and server access still blocked through the latest version of XNA?

SedativeChunk:

From some older post I've read

Why not read your current documentation? The entire section about using networking over Xbox Live?

Xbox Live is not supported by XNA. Is it still blocked? I wanted to add leaderboard support to my game if possible over Xbox Live. I have my own web server to do this, but is networking and server access still blocked through the latest version of XNA?

Networking and server access are not 'blocked'. You are programming for an XBox, you access XBox Live for the purposed of finding other players to play with. There is no reason to access an outside server.

You can implement a peer to peer leaderboard over XBox Live by connecting to other people playing your game and synchronizing scores. Do some searches. I believe that someone has done most of the work for you, and released a little framework to get that done. Maybe jwatte.

No game can access XBox Live without being paid for, and being a gold member.

Daaark:

SedativeChunk:

From some older post I've read

Why not read your current documentation? The entire section about using networking over Xbox Live?

Xbox Live is not supported by XNA. Is it still blocked? I wanted to add leaderboard support to my game if possible over Xbox Live. I have my own web server to do this, but is networking and server access still blocked through the latest version of XNA?

Networking and server access are not 'blocked'. You are programming for an XBox, you access XBox Live for the purposed of finding other players to play with. There is no reason to access an outside server.

You can implement a peer to peer leaderboard over XBox Live by connecting to other people playing your game and synchronizing scores. Do some searches. I believe that someone has done most of the work for you, and released a little framework to get that done. Maybe jwatte.

No game can access XBox Live without being paid for, and being a gold member.

You can't connect to anything over the 360. You can only do player to player matchmaking over XBL.

Peer to peer leaderboards for Gold Members running the paid for game is what you get.[/nothing more to say]

SedativeChunk:

As for the leaderboards, peer to peer won't cut it! I want everyone who buy's the games score to be saved on an international leaderboard.

I've read some people have used the built in framework in C# to connect to a database. Really, this is all I'm looking into doing. Connect to either a MySQL or MSQL database and save the highest scores in the game. Would this be possible at all? I'm going to try it now!

On windows you can do whatever you like

On Xbox there is my System.Data, SQL Server and mYSQL won't run on the platform and there is non System.Net or System.Web to connect to anything and the XBL network doesn't give you access to the outside world anyway.

I hate the Microsoft documentation to be honest. That's usually the last place I go to ever find any programming information because it's poorly written and usually has bad examples.

Personally, I find it to be of high quality in most of the cases. There are some cases where there will be documentation like "property bool Fribozz: this enables and disables Fribozz" without telling you why you'd want Fribozz enabled, nor when it's legal to enable it. However, most of the time, the documentation is correct, clear, and complete. If you can't get what you need from the MSDN reference documentation, then chances are the problem is in the other end of that connection.

When it comes to XNA, I find that the additional "how to" documentation also helps a lot.

However, when all else fail, I suppose you can use teh internetz!1!

And, as it turns out, my component is the very first hit.

To start off, I hate the Microsoft documentation to be honest. That's usually the last place I go to ever find any programming information because it's poorly written and usually has bad examples.

Thank you!!!   I totally agree.

Have you actually looked at the XNA documentation? Or are you just going off past experience with Microsoft documentation? Because I have to say that the Help files installed with the XNA framework are pretty dang good.

My idea, I haven't tested it yet, is to have a 2nd xbox set up and in your game have a 'leaderboard server mode' that is only accessible to an xbox live profile that you create specifically for the program.  There would be maybe a menu option for it that only appears when that specific profile is signed in (and signed into xbox live).  When other games try to view the leaderboards, they would look for an online session with a sessionProperty just for the leaderboard server and then download the leaderboards from the xbox server and send in their scores to it as well.  If your leaderboard server was down for some reason, it can just return an error saying that your leaderboard service is down at the moment.

I haven't tried or tested this, but I have toyed with the idea here and there.

Cost vs Reward.

Are you going to be able to truly justify doing all that for your Indie game? Are you confident that you will attract so many extra buyers because of that functionality that it will pay off in the end? If so, that sounds like a neat idea.

Yami M:

My idea, I haven't tested it yet, is to have a 2nd xbox set up

At least one game - Biology Battle - has a 'server' they leave running all the time.

and in your game have a 'leaderboard server mode' that is only accessible to an xbox live profile that you create specifically for the program.  There would be maybe a menu option for it that only appears when that specific profile is signed in (and signed into xbox live). 

This would be a fail in peer review. Having code which reviewers are unable to test would be seen as possibly hiding content. Reviewers must be able to review your entire game. What if you can use the secret menu to update the games content to something over XBL? So don't do this. Have a peer to peer system that works like several of the games, its ok to even have a 'server' so there is always one peer out there but don't code special server options.

What would be the process for this peer review failure?  When I peer review I'm not given any tools that show me how much of the instructions or data have or haven't been accessed.  I personally would have no way of knowing if some sort of "cheat code" hidden functionality was present, so I would likely not see anything that would cause me to fail the game.  Do other people have tools like this?

Would this be a violation of the principle of peer review, with no actual enforcability, or is there some actual, observable clue that would cause people to fail games in peer review for this reason?

Yes you are technically correct, we cannot fail for what we don't know. In this case it would only be possible to fail if the developer declared a mode which would only work for a certain gamertag which would mean a fail because the other reviewers could not get there.

There's been a couple of developer who have alluded to cheat modes while in review, and also I know one person accidentally found some cheats because he has a text pad attached and the developer left in some keyboard cheat modes. So press on your text pad while playing a game.

Really the warning is more there for the developers.... if it comes out later that you hid content or game modes then you broke the agreement with Microsoft when you uploaded it. That means game pulling, possibly money withholding and account bans depending on the offense. In other words don't do it!

Now I'm just brainstorming ways to exploit the rules to get around something I don't want.  I appreciate your patience with me, and here goes:

It seems like it'd be theoretically possible for a game to detect if it's being played as a live, released game, not being debugged or peer reviewed:
http://msdn.microsoft.com/en-us/library/microsoft.xna.framework.gamerservices.signedingamer.presence.aspx

It's also possible to include a sequence-of-button-presses code that cannot be learned by disassembly alone:  use a background thread (need System.Threading for that) to compute MD5 or RIPEMD160 hashes of a long sequence of inputs, and compare with a known hash.  The hash can certainly be read by disassembly;  cryptographically secure hashes are difficult to reverse so it's non-trivial to learn the original sequence-of-button-presses that yields that valid hash.

If the game is still in peer review, a "test" server-mode code can be distributed to testers.
If the game is out of peer review, the "test" code will no-longer work, and only a "production" code will be accepted.

All code submitted for peer review can be tested by peer reviewers.  The function checking the recently-entered code against the test code can execute, as can the one checking for the production code.  The actual server-mode code can also be tested by peer reviewers.  All of the statements in the code can be executed -- the only thing unavailable to peer reviewers is one of the two paths that lead to server-mode code.

That's my argument -- and I already see it's full of holes.  I'm not trying to debate, so much as beg "can this be allowed?"  One possible reading of the rules says this behavior could be accepted.  Maybe not, though.

What do you think?

Michael Spencer Jr:

What do you think?

I think you should just get to work on a game. What your doing here is just like finding loop holes to being annoying to someone else. Like when someone just follows you around and points at you from 2 inches away all day because technically, they aren't poking you anymore.

You aren't being clever. You are just wasting your own time. Your game will be pulled, and the wording of the EULA will be updated.

This is a game programming community, not the Underhanded C contest.

If you want to impress, be creative.

Thanks Nick.  The way you're discussing is how I would implement it, and your advice makes a lot of sense.  I started pursuing this because the statement about peer review highlighted something that seemed strange to me, and it seemed the original poster may benefit from the discussion because it speaks to a statement in one of their posts.

It feels like there's a kind of corner case, a weird area where the stated rules seem to accomplish something that makes sense -- but they also accomplish something beyond that, a bit more restrictive than that -- and I'm trying to figure out if the extra was intended or not.  It makes good sense to mandate that peer review be meaningful:  reviewers should be able to experience everything that end users might experience.  They may not see everything, but they should be able to.

If you want people to follow your rules, even in weird complex situations that aren't explicitly defined by the rules, sometimes you have to share your reasoning behind your rules.  I can understand a requirement that peer review be meaningful.  My question above served to ask:  in this weird corner case that I was shining a light on, did you mean for the rule to include or to exclude that?

So this question isn't really meant for anybody in this thread specifically, more for the Powers That Be, but:  did you mean to exclude that situation as well?

Nick, The_ZMan -- maybe you can help run this up the chain?  I know answers to policy questions like this can't really come from us lowly forum users -- I'm hoping for something more official.  Nobody is required to answer the question -- you're certainly free to say "here's the legal document;  read it and do what you think you can get away with."  I'm hoping for something a bit more friendly though.  :-)

Thanks again!

There is no way for you to make your game behave differently on marketplace than it does in peer rview. Thats by design becuase peer review is the checkpoint. So your idea isn't going to work unless Microsoft add something to the framework http://connect.microsoft.com/ is the place for that. The requirment isn't the peer review is simply meaningful its that all parts of your game are availalbe to be tested and checked so that problems don't occur in marketplace. Having a special mode totally precludes that. Its almost impossible to get the MS guys to comment on anythng so our opinion is the best you are going to get.. 

I don't understand where this thread is going but anymore discussion off topic and about how to work around peer review will result in this thread being locked

I almost forgot about this thread.  :-)

First, some background to go along with my apparent troublemaking:  I'm a Comp Sci grad student, finishing up my Masters next year.  I also TA an Intro to Game Programming II class that teaches XNA.  I talked to a couple of the XNA folks at PAX this year, and hope to see you guys at PAX East in Boston.  (I may have to address this in person there, if I continue to utterly fail to get my question understood here.)  XNA seems well designed to let indie game authors do strange and weird things that mainstream game publishers haven't done yet.  That interests me.  Some of the strange things a game may want to implement may use interesting sorts of peer-to-peer networking.

(I'm not tooting my own horn too much here:  the sort of fault-tolerant distributed system I'm talking about is probably beyond my own ability to create.  I'm not the point of this thread though.  When I'm inevitably asked "what would be the point of a system like this -- what could you use it for," my own limited imagination is also not the point of this thread.)

So now that I've totally hijacked this thread :-) my policy question for an XNA team member: 

It's clear that every behavior a game can exhibit must be available to be tested by a peer reviewer.  It's plausible to imagine a situation where a developer has given themselves some sort of "superpowers," which could be harmful or beneficial depending on context.

(Harmful context:  the developer uses their powers to kick people out of multiplayer sessions they should have no authority over.  Beneficial context:  the developer publishes a public-key authenticated message which is spread between peers, indicating a piece of user-created content violates terms of service and should be purged from peer-shared storage.  Again, let's not restrict the discussion to the realm of my limited imagination.)

It's possible to imagine an implementation method where testers are able to test-drive all of these "superpowers".  For example, test the process of creating profile-specific public key pairs, publishing user-created content, rating and sharing that content, and then banning it from the network due to a simulated violation.  Then, when the game is released, have these "superpowers" be limited to only certain trusted users -- the developer, or the XNA team or LIVE team, or whatever restriction could plausibly make this scary peer-to-peer system less scary and more controllable.

Finally the question: Ignoring for now the plausibility of a peer-to-peer system for sharing user-created in-game content:  can it ever be allowed, even through case-by-case approval from Microsoft, for a developer to create a game that allows all functionality to be checked in peer review, but to restrict which functionality can be used by which users once a game is released?  Or is it already decided that this is always prohibited, and every end-user must have the same abilities and permissions as every peer reviewer?

This question is on-topic because, while there are better ways to implement Xbox Live leaderboards, the implementation method discussed by the original poster would seem to be allowed or prohibited based on the answer to this question -- and this question has not yet been answered in this thread.  Variations of the question have been answered.

Ignoring for now the plausibility of a peer-to-peer system for sharing user-created in-game content

That has happened for a long time. Kodu Game Labs does it. Easy Golf does it. A number of other games do it, too.

can it ever be allowed, even through case-by-case approval from Microsoft, for a developer to create a game that allows all functionality to be checked in peer review, but to restrict which functionality can be used by which users once a game is released?

That happens all the time. Games feature "unlockables" or "secret levels" that you only get to if you're good enough, for instance -- that's a bar not everyone can pass.

However, consider the case of you running the server all the time -- perhaps your machine goes down now and then. Wouldn't it be great if there was some back-up? It turns out that with the distributed highscores component, which has been in use for over a year, in a number of released XNA games, you get that for free. You can let your box sit online and idle for a long time, and it will largely serve as a server. However, other boxes will also serve as a server in whatever fashion is needed, either if your machine is down, or if your machine is overloaded. It should scale fairly nicely under load (the main limitation being Xbox Live match-making).

And, because highscores are exchanged between all peers at some point, if another box happens to serve as back-up server while your box is overloaded, those scores will likely make it to you after a while anyway. Note that a single Xbox can only serve 31 other xboxes at the same time. A single node would significantly limit the number of players who could use your system, if you have a hit on your hands. 31 players is not a whole lot. I believe the peer-to-peer approach (as taken by the distributed xna highscores component) is the best option available, as long as we don't have real, Live!-backed leaderboards.

Michael Spencer Jr:

It seems like it'd be theoretically possible for a game to detect if it's being played as a live, released game, not being debugged or peer reviewed:
http://msdn.microsoft.com/en-us/library/microsoft.xna.framework.gamerservices.signedingamer.presence.aspx

The documentation talks about what you will see on the dashboard. There is no way to read that text from within an XNA title.

jwatte:

I believe the peer-to-peer approach (as taken by the distributed xna highscores component) is the best option available, as long as we don't have real, Live!-backed leaderboards.

Amen.