By Administrator, I mean *nix Volunteer ;)
Then I guess the more trusted users in the BYOND Linux Guild would be asked to setup/maintain the thing.
Rushnut wrote:
Then I guess the more trusted users in the BYOND Linux Guild would be asked to setup/maintain the thing. | |
Linux? I thought we were talking about Unix! *laughs*
| |
Rushnut wrote:
Linux? I thought we were talking about Unix! *laughs* I think I missed up my abbreviation there for a while. <.< But yeah, Linux is most probably the OS of choice. | |
Tiberath wrote:
Rushnut wrote: I was joking ;) | |
Sigh.. Again... A lot... That is twice in a row that Chrome decided to hit me with a 'Oops! Something went wrong so we have to reload your page' error that I've never seen before. Apparently it is caused by me using the spell check... :(
Moving on: A summary of the summary of my first post. This concept brings to mind: To what point and purpose? There are plenty of good game concepts that would not need hundreds of people in a server. To that end, there are servers all over the hub that are up 24/7 but rarely have people playing on them, if ever. BYOND is in no way in a shortage of hosts, so adding more hosts when that isn't the source of any of BYOND's problems seems pointless. People are looking at the X players per server limitation the wrong way. It isn't some limitation that stops any good games from prospering. It is a challenge that can easily be circumvented to allow more unique features to be implemented into large scale games like that. This is an opportunity for ingenuity, in areas where people have not really bothered before, but aren't that difficult. Imagine a large scale RPG meant to handle hundreds of players, spanning across multiple servers. Each server can hold its own map, and when you travel from one server (continent, if so chosen) you will simply transfer servers. A way to seamlessly change from one server to another would be very helpful here, but not necessary in the grand scheme of things. *EDIT: Was just playing around with link()... I seemed to remember there being a confirmation window before, asking if you wanted to change worlds. But it worked instantly in the test I just did... so yea. Ignore that. I sincerely doubt that Falacy's games are losing any playability because they cannot have 100+ people in the same place at the same time. I also sincerely doubt that anyone's games would feel any effect from that. Appropriate game design would not only eliminate the problems at hand, but add opportunity to expound on the previous conceptions of what BYOND is limited to. In conclusion... If people really wanted to do something to truly help BYOND progress, then they would get together a large group of people and actually do some research behind these network anomalies so the BYOND dev team will actually have something to work with. Solving simple things like the graphical tick skipping would benefit BYOND far greater than even more servers than we already have floating around. | |
AJX wrote:
This concept brings to mind: To what point and purpose? BYOND suffers from the shortage on resources that it's developers seem to be stuck with. Money is a valuable good that they can not spare most of the time, as is shown when visiting the Classified Ads forum. A good deal of original games is still lacking reasonable (24/7) hosting. AJX wrote: There are plenty of good game concepts that would not need hundreds of people in a server. I agree. I would never claim anything different. But what about the plenty of good game concepts that benefit from a stable 24/7 server? If you can only get one more creative/innovative game featured and played on BYOND, is it not worth the effort? AJX wrote: To that end, there are servers all over the hub that are up 24/7 but rarely have people playing on them, if ever. That is a problem I can not 'fight'. I have my own ideas on why this is the case and tried to cause a difference there (e.g. eye candy), without success. AJX wrote: People are looking at the X players per server limitation the wrong way. Yes, the concept of multiple servers for a game world is not a new one and part of the functionality change on link is based on Alathon's work on the subject. However, even with multiple little distributed servers, it is still something a serious developer has to pay (or trust somebody else to host). AJX wrote: I sincerely doubt that Falacy's games are losing any playability because they cannot have 100+ people in the same place at the same time. That might be because Falacy is investing massive amounts of money into a good stable server and thus his games can indeed, handle the hundreds of players you are talking about. AJX wrote: In conclusion... If people really wanted to do something to truly help BYOND progress, then they would get together a large group of people and actually do some research behind these network anomalies so the BYOND dev team will actually have something to work with. I'll gladly leave the bug fixing and tracking in the (capable) hands of Lummox JR and Tom, who know the C++ back-end code and focus on things I can change. Should they ever really need my assistance on tracking down bugs, they know how to reach me (for that very unlikely case ;)). | |
AJX wrote:
In conclusion... If people really wanted to do something to truly help BYOND progress then they would get together a large group of people and actually do some research behind these network anomalies so the BYOND dev team will actually have something to work with. They'd get together and do something to help the developers of the games, and the people who generate interest in the product to begin with so more memberships are bought and a larger income (not to mention new people with fresh ideas) for BYOND is made, potentially giving the opportunity for an additional programmer to join the team and work on bugs such as your network issue. You have to remember, fixing bugs is only useful if there are enough people out there to care that the bug is fixed. I would rather see more developers creating more games so another C++ programmer could be hired. Not only will more bugs be fixed quicker, but more features will be added quicker, which will generate more interest in the product and the cycle continues. | |
Find me seven people willing to pay $10 a month or ten people willing to pay me $7 a month, or any number of people willing to, in total, pay me $70 a month, and I will provide a server with 4GB of memory with a bad-ass internet connection, Quad-Core Opteron processor, and a nice little web panel for game control. It's really not a bad idea, I mean, surely we could find fifteen people on BYOND that would be willing to toss up $4.75 a month to help out developers?
| |
Airjoe wrote:
(...)in total, pay me $70 a month, I'd say that's a bit of a... steep start. I was considering ATP Development, but more like a 12.5$ server for first month, given the likeliness of having to pay most the first expanses out of my own pocket ;) Airjoe wrote: It's really not a bad idea, I mean, surely we could find fifteen people on BYOND that would be willing to toss up $4.75 a month to help out developers? Are we talking about the same community here? ;) Given the age distribution I do not find it astonishing that most every Classified Ads posting names 'GM/Owner/Credit' as compensation. I wouldn't recommend starting out too big. If things go well, one can always upgrade later on, but relying on donations in an environment such as BYOND is a tricky point. | |
The concept is certainly an interesting one (although I think it would be better if it was more of a group of BYOND devs decided to go in on a server together and split expenses).
Clearly, the hardest part will be limiting system resources. As far as maintaining CPU goes, I would say that the use of priorities should handle this well. There should be some limit (determined by the number of games hosted) that is set, lets say 10%. Any game using more than 10% will be given a lower priority. If the CPU isn't maxed out, it won't matter that they have a lower priority and they will run unhindered. If the CPU is maxed out, the games running over 10% will suffer much more greatly than those running under 10%. This is assuming a *nix server, which is the more logical choice. As far as RAM goes, I don't have a great idea for that one yet, but since RAM is relatively cheap, it really shouldn't be problem (unless some program has a memory leak). And it would be fairly easy to install apache2 and host a simple little interface (login necessary) to allow devs access to what they need easily. Anyway, good luck with this idea. | |
Tiberath wrote:
They'd get together and do something to help the developers of the games, and the people who generate interest in the product to begin with so more memberships are bought and a larger income (not to mention new people with fresh ideas) for BYOND is made, potentially giving the opportunity for an additional programmer to join the team and work on bugs such as your network issue. This statement I strongly agree with. Hopefully soon we will see a strong influx of new good games for people to enjoy. :) | |
Stupot wrote:
The concept is certainly an interesting one (although I think it would be better if it was more of a group of BYOND devs decided to go in on a server together and split expenses). Though that is kind of contradicting the basic idea behind the concept. The thesis is that a lot of developers can not (or do not want to) afford paying (shares) on a server and thus lack the opportunity to host their innovative games. Which are somewhat missed out by the community. Stupot wrote: Clearly, the hardest part will be limiting system resources. As far as maintaining CPU goes, I would say that the use of priorities should handle this well. Actually, I was considering a rather different attempt. Limiting the number of players allowed per game (on a 'by experience' base) and after that was reached, only allow BYOND members to join. That way, there might be come incentive to register as BYOND member (hopefully using the referral link of the server ;)). Stupot wrote: Anyway, good luck with this idea. Thank you ;) | |
Schnitzelnagler wrote:
The thesis is that a lot of developers can not (or do not want to) afford paying (shares) on a server and thus lack the opportunity to host their innovative games. Find us an innovative game that is worth playing that doesn't have a host and we will find it a host. | |
AJX wrote:
Find us an innovative game that is worth playing that doesn't have a host Freeze Tag(2) Hazordhu Murder Mansion Proelium(2) Space Castle Tanks Wurms(2) I fear that 'worth playing' is rather subjective though. And I'm not going to continue the list in the search for games, since those are just the ones off-line on my pager right now ;) | |
Schnitzelnagler wrote:
Freeze Tag(2) Haha, I was working on a Freeze Tag game back in 2006. I wonder if I should look into doing it again. | |
I'm going to friggin kill chrome...
Anyway~ Basically all I said in the post I just lost was the majority of those games if put under a 24/7 server situation will suffer from 'empty server syndrome', where players will not join and wait for people to join, and so they will stay empty. Never played Hazordhu, Proelium 2, or Wurms. I'd LIKE to play Proelium 2, but I have never ever seen a server hosted for this game or players in it. Tanks: I loved this game. This game was awesome. However, Tanks died because of stagnant issues, not because it wasn't an awesome game. If Tanks was still being updated then I'm sure it would still being played. | |
AJX wrote:
I'm going to friggin kill chrome... For short games, where it's like a 10 minute battle, or really where the game ends for everybody after something occurs, the BYOND pager system fails. It might help if the pager had some system that could inform players that a game will be occurring in 30 minutes or whatever. There are a lot of different ways to attempt to implement this, but it could be a worthwhile endeavor. Anyway, that was fairly off-topic. The thesis is that a lot of developers can not (or do not want to) afford paying (shares) on a server and thus lack the opportunity to host their innovative games. BYOND staff shouldn't have to (and most likely won't) pay for servers for devs to host their games. Unless they will eventually make profit from their initial investment, they shouldn't. And if that is the case, I don't see why it needs to be BYOND staff at all. It could easily be a group of devs who go out, by a server, and determine what games get hosted and for how much a month. It's just a matter of who makes the initial investment and who decides whom is a competent developer worthy of including. Limiting the game to a certain number of players isn't something that can easily be handled by the server itself (if at all). From my experience, it's best to have a self-maintained system rather than rely on users to do something correctly. Plus, just because a game only has 5 people on it doesn't mean it might not use 25% of the CPU (assuming quad-core). Once again, you'd be relying on the users (in this case the devs) to make sure their game is efficient right off the bat. | |
Stupot wrote:
BYOND staff shouldn't have to (and most likely won't) pay for servers for devs to host their games. If you did read the blog posting carefully, you might have come across this passage, which basically says just the very same: "I doubt (and don't think) that BYOND would (should) take the venture, because of the financial risk and the immense workload involved (imagine the spam by 'programmers' wanting to get a free slot), but a private person/group should have less trouble, if the concept is well planned." I never claimed in any way that BYOND should be responsible for this. Stupot wrote: Limiting the game to a certain number of players isn't something that can easily be handled by the server itself (if at all). If you only have hand-picked developers, you can either trust them to do something (with your help), or you took the wrong decision and should pick another developer instead. The idea behind limiting the amount of players is not meant to keep resources down, but to encourage people to buy a membership (as only BYOND members can join past the limit, if you want to play 24/7, it's only fair if you support the system). | |
Stupot wrote: Limiting the game to a certain number of players isn't something that can easily be handled by the server itself (if at all). I don't think it's impossible to limit the number of active connections to a single port. | |
By Administrator, I mean *nix Volunteer ;)