Those enticements are there to get more donors, and i agree with them, and yes i do agree that holding non-donors as if their lesser is a problem, we're on the same page now, the argument is over.
Why? Keep in mind that this income isn't to the personal benefit to the owners, its to the community as a whole. Maximising income means better infrastructure to run more packs for more people faster and with better connections. How is this a bad thing? Edit: Or hiring good coders and staff, etc.
At the expense of an enjoyable experience for non paying players. As a donor, my opinion is that donators should get nothing. Yep, zilch. Absolutely nothing.
Exactly what is at the expense of non-donors? Or let me rephrase this, what have I advocated for that would give donors preferential treatment? This is off topic, I was advocating for an overhaul to the reset system so that it maximises retention by making players less upset. The longer you keep a non-donor on a server, the more time you have to convert them to donors. So what's the problem?
The mere fact that they get anything is already preferential treatment. Expanding upon what they get is wrong imo.
he didn't suggest that.... and given me and @bryan84 's past history it's quite surprising we can agree on anything, trust me
So how do you intend for the server to pay for itself? Also, how does this impede non-paying players from having fun? Nothing is being taken from non-donors to give to donors....
A true fact to think about. The survey we have running: 50% of the submissions see the performance as a shadow site and quit packs because of it 0, not a single one said anything about server resets (+95% of the submissions were from players that have been around with the network for +3 months) Going by this, it is not a big deal for the general player base. While the goal is coming across I do not see a solution. Yes there have been suggestions, but they are either not possible (chat trigger for staff), not feasible (from the 40 moderators and senior moderators there are maybe 3-4 that could deal ok - well with those raging players, if those players are not completely blocking the outside, the senior+ staff is better focused on improving tps -> see survey results) or are likely to have negative impact (letting players and staff vote about a server reset, empties the server, even though the vote might not be successful - empty servers are a high server resource expense). The shop (next to friendly staff) has been the main reason for staying with MyM. Another thing most are probably not aware of (let me ensure that has never been a reason for a reset) is that 30-60% of a servers income is during the first week, although the server being up for +2 months at least.
Im guilty of this. I tried playing Infinity 5 when it came out but I could not endure permanent sub 10 tps in the first week. ps: when is regrowth coming out?
we ran into some worldgen issues (95% ocean), we know how to fix it as booker did it before, so hopefully tomorrow.
What's a shadow site? I wan't suggesting that everything I mentioned be implemented, I was just tossing out ideas on how you could maybe implement them. Frankly I just think that having an automated message come up that says "Players have asked for a reset due to crowding etc., please click here to vote", let them vote for the restart and while this is going on have messages that say "There is a restart vote in progress, if you're a new player and don't want to chance it, try "Ferret Invasion" or some other pack and " or please try back again in a few days, reset or not it'll be out of limbo in (countdown from when vote started)!" Make the voting short, like 2 days and the 3rd day is the reset. All I'm saying is that the Surprise your base is gone with no warning!" party is the worst kind of surprise party. This should keep that from happening. ^^^Forgot to mention, if you do implement voting, I'd weight the votes of players who've been on consistently in the past 3 weeks heavily while making the newest and oldest votes count least. That way you don't get old players who've done everything and rarely play now voting no to keep their awesome base alive. Same with people who just logged in and want to get a cleared playing field. I wouldn't give all players an equal vote on this, it should be decided by those who have the most to loss or gain, not a lost soul wondering in. Another alternative; I remember there being a claim system (maybe Towny) that would revert claims when a claim is abandoned. Instead of resets, a lot of the time you could just regen the chunks that non-current players have claimed. So if a player is off for a month, replace his chunks with the natural area that preceded it with a copy of the chunks from the original world gen. *That* would clean out the vast majority of crowded servers; I stumble across peoples bases constantly and never know the person. A lot of times, out of curiousity, if I can get the name of the player I'll type /seen playername and it's always months. I think if you routinely cleansed the server of bases abandoned more than say 3 wks, the need for frequent resets would diminish; also maybe periodically regen areas of the world that have been overharvested if that's possible to do. A third option is that maybe there could be a subscription based model where you have a server that runs better hardware, more cycles for a certain mod that is spun off popular packs, and you charge a flat fee for access to Premium mirrors of the free packs with travel in between like the old monster nodes or something. I know that on the surface this sounds like I am advocating for what i said a few posts above I wasn't in favor of but.... If dedicated players, who tend to be the most resource consuming from tps standpoint can pay for a better server they reduce lag on the free servers by moving there. I think a reasonably priced subscription model like that could help the whole server because it moves the non casual players to a dedicated area and it provides income to help subsidize servers with low proportions of donors. I mean donating to a decent tier can be expensive for someone without a lot of disposable income but even a broke person can probably afford $5-10 a month, so 15-30 cents a day to play on a fast server. As long as people on the free servers can do the same stuff and get the same gameplay w/o negative incentives to coerce them into subscribing, I don't see this being a problem. Its just like being willing to spend money on getting a good internet connection if you already have an OK one for free. Think of like municipal wifi vs fios. Did the survey ask how people felt about resets or just about gripes in general? When restarts happen it always suddenly seems to come to the forefront on the chat of any server I've been in where it was brought up.
probably lost in translation. Schattenseite - Englisch-Übersetzung â Linguee Wörterbuch So kinda the things that MyM is bad at, things that you don't like..
It might not feel like that so much, but we made the experience more than once. A single post on the forums from a player asking if there could be a reset or starting a vote = server on 1/5th of the player base. I do not agree. My experience has been, old players care less, new players care less. Medium progressed players care the most. Meaning it will most likely result in a no vote all the time. This will result in the server dying slowly and dragging down the quality of the network. The lag is the worst for new players that need to go mining by hand. Once the mining is automated the lag isn't so bad anymore. This sounds a lot more simple than it really is. It is something on the todo list for quite some time now and we investigated it thoroughly. The issues at hand are: evaluation of inactive claims (owner offline for 4 weeks, trusted players still active, but on this claim or their own?) players going on vacation (requires some sort of vacation mode, but how is it handled with players that are off minecraft un-planned without being able to put their stuff in vacation mode) removing chunks during runtime often crashes the server (in modded minecraft) removing them outside of runtime requires to write an app that can read the region files and clear out the required chunks. It also needs some sort of api, as the inactive claim information requires data which is only available on runtime. => it is possible, but it comes with many issues before the translation and implementation already started => BUT, do we know for sure that it will solve or at least better the issues? While I believe in it bettering the situation I don't believe it will solve it for good. It might give 30% boost in server performance. Is that enough with a server that is super old. Another note on this, is the performance the only reason for a reset? If we only do new packs, we run into the issue of having the broad band of players going to the one new server, the one that isn't battle tested. No one really likes to go to an old server, but when there is a refresh/reset many think, well I always wanted to play that pack, now is the time. So there needs to be a new incentive to motivate players for action. MyM Survey - Sep 2015 Regarding server hardware, there is not much to upgrade to. The bottleneck is the single threaded cpu power. The CPU's we use are 1270v3 rated at 2,195. The worlds best enterprise cpu regarding single thread performance is at 2,343 at around 4-5 times the cost, that's not even 1 tps more. We can setup "premium" servers for the "major" packs, but seeing as infinity and dw20 are not much better than monster regarding performance, we will probably dip into the same issues which we already had. Maybe there is a way to set it up better, make it work better. From my side I don't see much that could really make a difference. Again, drifting amounts of players, non update compatible updates are a thing, too. Resets due to performance issues are maybe around 60%. There is one server setup that I thought about in the past and where I think it could solve all the issues and the only thing holding it back being limitations regarding cross world. Kinda the thing we had with monster where the farmworlds only were on one server. So this would be a design where every player/group of players would have their own world. The reason for this is that it gives a ground level of determining the resource consumption per player/group. We can use a sort of stop watch that measures the time the world requires to tick (process one cycle). There are a few things like conduits and applied energistics that are not being dealt with during the world tick, but those are limited and could probably be tweaked around. Once we know how long a world takes to update, we know how much resources the world/the player/the group requires. This allows us to set a limit, once the limit is exceeded updates get skipped. So if the limit is 4ms and your base is at 8ms, it will only update every second tick, meaning it is running at half speed. Resulting in players that do not optimize their base and like to build big, will cut into their own flesh instead of the one of others. On top this could be commercialized. So you wouldn't pay for a perk like fly, you would pay for more server resources being dedicated to your world. The other big advantage of this system is scale-ability. So there is a new pack which everyone was waiting for, they all get on the one server and it is exploding by the amount of players/slots are being used up. (we don't announce new servers in a big fashion as the new servers can't handle the load and for multiple they are too unstable at the beginning) On the other side there are the old packs like b-team with 3 servers and an amount of players that would justify 1 server. The player inventory would be stored on the database, the world would be stored on a storage server. The player logs on and goes to the lobby. In the lobby he clicks join. This will inform the server with the lowest load to load his world from the storage server. If all online servers reach a defined level of load another one is being started and added to the fleet waiting for world load requests from players. => only stuff of online players is loaded, servers can be scaled up and down automatically in the matter of a few minutes In my opinion this would solve all the issues around resets and especially lag. Unfortunately there is a big but as there are many limitations, to count a few: you can only live in small pre-generated worlds you can not visit offline players unless you are trusted.. delay until you can play (world needs to be moved to the proper server and loaded) you can not move items between worlds (farmworld might be on another server) mods like blood magic and computer craft store data outside the player file and world (would get lost if the world is loaded on another server next time) thaumcraft only works if you connect directly to the server server crashes - in cases where we don't know which world lag spikes will still be a big thing
I've heard the term shadowsite before it just didn't make sense in context. But thanks for the super nice snarky ajax scrip [DOUBLEPOST=1442702526,1442701115][/DOUBLEPOST] Before reading the below, since MS bought MC, have they converted it into C++? Or is the a way to convert it into another language that has natural multithreading built in? What you said is interesting and potentially feasible. I'd use sharding to eliminate the problem of scalability though. How about a setup like the following. I like the idea of a centralized player data tracker but I'm thinking more along the lines of there being "words" or dimensions of a few chunks that are seamlessly integrated into the overall map; just make it tileable for expansion. Essentially any X amount of chunks are a private server and you simulate on their own thread and what happens when you reveal a new chunk when you're exploring. When boundaries are crossed, you essentially transport from a private dimension (thread) that looks the same as the reg ow to the actual ow. I had half finished typing the below when i really got into the details of your idea so there's a lot of overlap: Is it impossible for a server to be setup using hyperthreading where ea premium player has a large "island" like crash landing that integrates into an overworld where the player data is synced across? That way you could have a zone where you had awesome tps that would integrate into the regular map sorta seamlessly, or maybe a dimension running on its own thread. Synchronization would occur at the borders when you move back and forth between them. So basically the island is a private dimension that perfectly maps the ow, with only certain chunks represented in the island so that everyone has the experience of 1 map. It should also be possible to have a party of friends that have perms to enter your dimension. It's been awhile since I've looked at the actual architecture of how smp works nowadays but by almost by definition, chunks allow for sharding since they're discrete; this is how massive website datasets (ie mysql - see facebook clones) winnow down massive datasets to discrete elements that are manageable. I'd think you parallelize portions of the map onto different hardware and only synchronize at the borders. This could be made almost trivial using portals to control boundary transfers. I admit it would require a good degree of coding to make this happen but I don't think its insurmountable. As long as you can sync player data through every boundary using a repository you shouldn't have problems with inventory or thaum or bm for that matter. Admittedly its been awhile since I've had to project manage something as complicated as this but it seems possible to me but I will admit that with my limited understanding of the current architecture this may be naive.
I would definitely be in favor of some form of transferring a base, even if no items could be kept. As for claim blocks, I can see why people would dislike losing those, even if it doesn't particularly affect myself, because of donating. Actually to be quite honest, I think the claim block system could use some Quality of Life changes. I'd prefer if we got more from voting, yet have the count of blocks be network-wide (consistent across all MyM servers). It's always frustrating trying to decide between some mod-packs and then wasting your votes on one only to have all your friends play another, or other similar situations.
unlikely and especially not any time soon. They made a win 10 version with net, but that is like the mobile one.
