Jump to content
Eternal Lands Official Forums

Llywar

Members
  • Content count

    157
  • Joined

  • Last visited

Everything posted by Llywar

  1. Multiplayer cheating

    I am under the impression that playing multiple characters at once is legal as long as they don't interact at all. If that isn't true perhaps it should be made clear in the rules (forgive me if it is covered somewhere). Anybody out to cheat would easily modify their client to remove any checks, or would just use vmware/etc to isolate individual clients. There really is no way to detect this - even with closed-source software it can be done undetectably (with more difficulty). Maybe with hardware TCPM it might be possible to detect, but let's not go there, please? Anybody who runs gentoo builds their own clients, so you can't even tuck away an easter-egg in the distributed binary. It should be fairly effective to look for signs of cheating in the logs after the fact, and not try to catch it instantly. If somebody cheats from level 5->7 and gets a warning nobody is going to be put out. And by the time chars are high-level they should understand the rules - nobody is going to jeopardize a high-level char for the sake of a little muling if they know they're going to get caught. And I'm guessing the current controls aren't all that manual. A simple log query would identify all same-IP trades. A slightly more complex one would catch any same-IP bag-trades. Most likely the admins have both in place already - and this is all a client-side solution would stop anyway. I don't think that cheating is that big of a problem - or at least it is under effective control. And I'm sure the mods go easy on newbies that have only been playing a week.
  2. Advanced Broker-Bot

    Do you perchance work for an airline, or a trade union? Competition isn't MEANT to be fair - it is meant to lead to more efficient prices. And I'm not too concerned with level-95 alchemists cornering the FE market - I'm sure they have better things to do than to click on mix all day for 10 exp each. And maybe if some raw materials were cheaper the final products might actually be profitable. Nobody should be dictating prices - people will figure out how low they are willing to go and a floor will form on prices - just like in real life. As others have pointed out - not listing the prices won't change anything, but it will result in everybody having to send 100 PMs to each bot to get the current prices. Nobody is going to just buy anything from the first bot on the list without comparison shopping - unless they just need 2 FEs or something silly like that. Unless you make a lot of gc/hour it is better to walk than to be lazy. And if time is THAT important to you then paying an extra few percent won't deter you from going to the nearest bot anyway.
  3. Support a good cause & win big prizes

    In reality the cream probably cost $500M to bring to market, and $1.80 per unit to put on the shelves. The Romanians enact price controls at $2 and the manufacturer sells it at that price since 20 cents profit is better than no profit. However, you'll never recoup the $500M 20 cents at a time when you only have ten years of patent life. In the US the prices are much higher which is where the profits actually get made above and beyond the initial outlay. If prices were frozen at $2 everywhere it would be cheap in the US as well, but nobody would bother coming up with an improved version of the cream. Maybe there are some people out there now who the cream doesn't work for - do we want them to have a cream of their own? The issue has nothing to do with labor costs in Romania or the US - nobody in Romania worked on R&D for the product in the first place. It probably isn't even manufactured domestically - if it were then the product bought in the US would be just as likely to have been made in Romania. You are actually in a somewhat similar situation with the EL store (but on a much smaller scale). When you sell somebody a thermal serp sword does it really cost $90 to change a record in their storage data? You could give everyone 5000 thermal serps tomorrow and it would probably only take 20 minutes of your time in total. But that neglects all the effort that you put into having a game in the first place - if you hadn't worked hard for many years we wouldn't be in a position to be buying those swords at all. Now, since EL is the result of a small development team it is possible that you might have undertaken the work even without the possibility of making much money from it. However, you can't really scale that up to an R&D team of thousands of people (which is the number of people in R&D at any of the major Pharma outfits). The bottom line is that somebody has to pay the $500M for R&D. Now, there are a lot of other ways to do that than charging the people who need the pills. However, you can't just make the costs go away by enacting price controls - this European tactic only works because the US tolerates bearing most of the costs of drug R&D. If the price on your cream drops to $2 everywhere then you won't see any new drugs coming out of private R&D. Personally I think there are better solutions - such as leaving the private system alone and starting up a competing public system that funds R&D and outsources later development to the established pharma companies - the resulting drugs would be widely licensed and would be dirt-cheap - taxpayers would bear the heaviest costs. The competition would keep prices down on less innovative drugs. However, if private industry developed a treatment not available from the public R&D effort they could still make lots of money from it - and as the treatment wouldn't exist at all otherwise it is still a net-benefit to the public. If the public R&D effort is highly successful it would eventually replace the private efforts via normal market forces. If the public R&D effort becomes a bureaucratic boondoggle then we'll still have the status quo (as opposed to no new medicines). Proponents of both private and public solutions can sleep soundly at night knowing that their side will win out in the end, and either way the consumer is better-off. No matter what the public effort will help benefit those who have rare disorders that will never be profitable for private R&D to address. Keep in mind that R&D most benefits people who are currently healthy - new treatments for disease will almost certainly come too late for people who are sick today. R&D benefits most people who are healthy now, but who unknowling will contract some horrible condition that is currently untreatable. In the US the cream developed two years ago will cost $10, but the one that was used 15 years ago is probably $1.50 - and the same will be true 15 years from now. If you freeze R&D the cost might seem to disappear, but it is only because you take away the choice from the consumer about whether they'd rather spend more for a newer product. The main thing that I'm trying to communicate is that the drug cost problem is a bit more complex than it might seem on the surface. That doesn't mean that the status quo is acceptable. It does mean that simplistic solutions aren't necessarily the right ones, and that solutions that destroy drug companies aren't necessarily in the best interests of those who are ill (and even more so those who are currently healthy). Just look at how hard it is to balance the EL economy and think about what it would be like if there were 30,000 highly-skilled workers involved in it full-time! And I really don't like discussing this topic in this particular forum. I think we can all agree that supporting this cause is a great thing and it really is above debates over the best way to solve the worlds public health problems/costs. Again I'd like to thank those who have donated both prizes and money to this effort!
  4. Support a good cause & win big prizes

    I would think that if a cure gets found a foundation would probably license it pretty openly and it would be cheap. However, more likely than not a cure won't be found, but some really good ideas on how to go about developing a cure might. If that happens and a pharma company is the first to actually develop a cure then it will be expensive - at least until somebody else does develop a cure and makes it cheap. If a patent on the "really good ideas" is obtained then the pharma companies probably wouldn't bother looking into it at all unless it were exclusively licensed. However, most foundations probably wouldn't bother pushing for additional treatments/cures once drugs are on the market that work. It just wouldn't be cost effective most likely. For as expensive as drugs are they do reflect about what it actually costs to R&D a treatment for a disease - and for a non-drug-company to do it there might be even more expense since they aren't optimized for drug R&D. And yes, I know that drug costs go to pay all kinds of expenses other than R&D - but that doesn't necessarily mean that others could do it more cheaply even without the expenses of marketing/profit unless they were doing this sort of activity full-time for many years at a large scale. About the best a foundation could do is subcontract the drug R&D part of the project to a major pharma outfit - but that would be very expensive as it would have to compete with the normal profitable drug R&D pipeline resources. I'm sure if you gave a big pharma $500M or so they would develop your drug, optimize it to some degree, and obtain data necessary for a marketing application. However, that is a lot of cash for a foundation to come up with, and there is a half-decent chance that nothing will come of it. Don't get me wrong - I think that there are certainly problems with the accessibility of medical care in the US. However, while it has become fashionable to blame this on pharma companies there are a LOT of people making a profit from saving lives. And a good chunk of pharma R&D expenses go to doctors involved in clinical trials. Ultimately if you want a lot of people to work seriously at something you need to pay them, and that costs a lot of money. While the world would certainly benefit from open-source drugs just as they benefit from FOSS there are some problems in applying that model to drug R&D (high capital costs, need for human beta-testing, large number of skillsets required, etc). There are certainly ways to redistribute how drug R&D gets paid for, but you can't really change the fact that somebody has to pay for it. If a foundation wants to take on all of this expense so that drugs can be dirt-cheap for the first ten years and not only the years after this then I'm all for it - but most often those crying for cheap drugs aren't willing to actually pay to make them happen. Regardless - I think this idea is a great one - thanks to all that chipped in to make it happen!
  5. NMT breakable poll

    I'd argue it should be more like 50% stronger. Right now though it is probably 5,000% stronger or more. I think that if you want to attract new players you can't start them off 7 years behind everyone else. Otherwise leveling becomes a goal for its own sake. I don't think that this is accurate - but it depends on the relative break rates and costs of each item. So, suppose all your armor elements cost 50k each, and you have 5 of them. Suppose the NMT costs 100k. Suppose all normally have a break rate of 1:10,000 hits. On the surface it might look better to use the 100k NMT to protect 250k worth of armor. However: Without the NMT on each hit you have a 5:10000 chance of losing a 50k item. After 1000 hits you are 41% likely to lose at least one 50k item. That is 20gc/hit loss on average. With the NMT on each hit you have a 5:40000 chance of losing a 50k item. After 1000 hits you are 12% likely to lose at least one 50k item. You are also 10% likely to lose the 100k NMT. That is 16gc/hit loss on average. The difference isn't that large - and if you factor in the fact that degraded items have value it could be a net loss. If the NMT break rate is much higher or lower that obviously impacts the math quite a bit, and that is going to be the ultimate determiner of the value of the NMT...
  6. NMT breakable poll

    I think we have different ideas regarding typical play time. Maybe if somebody works 12 hour days for 7 days they might be able to hit 50/50, but if you only play a few hours per week (this is a game, right?), and do something other than just fight, you're not going to be at 50/50 anytime soon. I've been playing 7 months and am only at 40/40 a/d. Ok, I think we need to think about what motivates people to play games. If the goal of EL is to encourage people to spend their first 2 weeks on the blue lup bush to raise 360k I really question what the game has to offer. I mean - really - what exactly is entertaining about walking back and forth to a flower shop? I guess people have different motivations to play games. And if the NMT becomes breakable it won't sell for 200k. It would probably sell for 50-100k tops I'd think. If those who find them price them higher nobody will bother to buy them anyway - unless they just have money to burn. If they are bought by an NPC for more they'll just all leave the game. Kind of like titanium bars - nobody bothers selling those on the market as nobody is willing to pay more than the NPC... In any case, as long as people know whether an item is breakable or not everybody will adapt. It is just a game - I'll have fun regardless. But I think fundamentally it is better for the economy to make items depart. It might also make sense to make the NMT manufacturable as an engineering item, or something like that - it could still be fairly expensive. But if it is breakable I'd probably not make it TOO expensive depending on the break rate - why protect a 60kCoL with a 100k cape that breaks just as easily?
  7. NMT breakable poll

    Uh, people who can't kill Ogres without a CoL probably won't be PKing in any case - they would be killed the second they walked into KF. One issue with EL is that there is a HUGE span in power as you gain levels. If you took every level-40 a/d fighter and matched them up against a single 100 a/d fighter they would probably clobber all of them. So, anybody who hasn't been playing for more than a year (or who doesn't have 100 hours/week to play) is going to focus on levelling as fast as possible to try to catch up. If the level span weren't quite so large you might see more PKing - luck and tactics would matter a lot more. Right now if a 50/50 and a 100/100 fighter step into the ring the result is a foregone conclusion. The higher-level fighter should certainly have an advantage, but perhaps it shouldn't be so large. But all of this is only somewhat-related to the NMT. I think that making it breakable would increase the value of the perk (which right now isn't a good investment - 7PPs=3.5Mgc at nexus/hydro rates). As far as upsetting current players goes - they have a chance right now to sell theirs for only a moderate loss before any changes get announced. Or maybe make an NPC buy them for the first week after the change to give people a chance to dump them (make them breakable first, then make them more common after the NPC is gone). Then people can cash out if they have them already. Any time you have unbreakable items it is a problem for the economy. And the NMT effectively throws off the economy for all weapons/armor. Long-term the break rates on the armor will just increase so that they break often enough, and those without the NMT will get really hammered. But eventually everybody will have one anyway if they never leave the game. By making people use PPs for NMT you will make it more of a trade-off and not everybody will just take it...
  8. I'll order 2k sulfur - that's 3800.
  9. Disclosure of the rate changes

    Not sure that is such a good reason to implement... The question is WHY would this make people unhappy? I think that we'd need more resources and more things to do that don't depend on harvesting for this to work. When levelling any skill requires tons of harvesting then eliminating the harvesting gets rid of everything else. Of course, it doesn't have to work this way. Unfortunately, I'm not sure i can think of a better way. I have some ideas, but I don't think they'll work - they'd all involve fundamental changes in how harvesting works (get rid of emu, or have harvesting go direct to store, or have claims so that you can stake ownership apart from the muling - lots of other options). Prices would probably change quite a bit - harvestables would become much more expensive. Unless the xp system changes that means levelling will be much harder as well. If we just scale up the xp to go along with the decreased number of harvestables then we won't actually solve the problem - we'll just multiple everything by a factor of 10 or something (stuff costs 10X more, but gives 10X more xp, so just as much supply/demand/etc)...
  10. Combinational Special Days

    How about Sun Tzu and Peace day?
  11. Disclosure of the rate changes

    Ok, I realize that EL isn't RL – but I think that the scarcity/hoarding issue is probably related. Why do people hoard things? They get hoarded when it is perceived that they will be worth more in the future. Now, some people have argued that they don't hard to sell later for cash – but this doesn't actually matter at all. If they hoard to use it is because they don't want to have to buy the item later – and that is because they expect the item to be expensive to buy at a later date. (For those who argue that it won't be expensive – just inconvenient – I argue that they are in fact the same thing. Expense and convenience are trade-offs – post on channel 3/33 that you'll buy up to 100 EFEs for 25k each and you'll have people running to the store from all over C1/C2.) The root cause of this is often inflation – the fear that a gc today is worth a lot more than a gc tomorrow. If there were no fear of inflation then (rational) people would sell their excess stuff now, and then just buy it back later if it were needed. Now, I'm not sure if the issue is with inflation in general, as people aren't hoarding everything – just a few particular items whose quantity is carefully controlled. You don't see the value of FEs changing much. I don't think that the total number of an item is a good indicator of how rare it should be – at least not by itself. I think you need to look at the trade in an item holistically, where hoarding is just one factor. Normally I'd just say not to worry about hoarding and just be more liberal with rares with the goal of maintaining a given price range. However, I do realize the problem of dumping. One possible solution to hoarding is decay. If you give rosts and EFEs a half-life of maybe 2 months or so then people will definitely not hoard them – and it wouldn't matter if they tried to. Items would quickly find their way to people who could make use of them. Rosts are a bit of a special case – many people carry them as an insurance policy even if they don't intend to risk death – so a rost might stick around for a long time with lots of people having 1-2 each. Another solution would be a banking system – if people had something to do with their cash then they'd be less likely to hoard items. Instead they'd sell them and put the cash into some kind of investment. That is essentially how most governments regulate the monetary supply. Of course, in EL anything that adds cash to the economy is going to be an issue in general. I think that keeping actual rare generation ratios secret isn't a bad idea – people will still get the general feel for what the ratios are but it won't have an instant effect and will allow for more gradual adjustments to take place.
  12. A field for low lvlpkers

    Might want to modify the mapwalking code then to make it an option to walk or not walk through PK-able areas. Otherwise people will have to micromanage every little trip so as to not wander into PK-land...
  13. Invasion

    One issue here is allowing players to be able to be able to match up with appropriately-leveled enemies. I'm not sure you can really compare real life to EL. In real life there was certainly a difference between a serf issued a sword and shield and a knight in full plate. However, that knight wouldn't have been able to single-handedly mow down an entire battlefield full of serfs. In EL the situation is very different - one cyclops could probably kill every low-level character who shows up to fight. In the same way, as a 30/30 I could mow down every rabbit on IP for hours without taking any damage at all. So, if you have deer next to yeti the low-level fighters will just run, as they'd be killed in one hit if they got anywhere near the yeti. So the deer would end up just getting cleaned up by the high-level fighers. A/D levels in EL have a MUCH larger impact than in real life - where the best warrior is only moderately superior to an untrained one. As a result it is hard to have people with 10/10 and 80/80 fighting on the same map...
  14. Exping in teams

    Here is an idea - how about a team inventory? You get a new inventory box (maybe controlled by the team lead with delegated access, or maybe just common access). The team inventory is the sum of all players unused emu. So, if 100 players want to mule iron they could harvest into the common inventory, which might have 30k emu. Then it could be split up in any way back at store, or even all dumped into a single char's storage or all given to a bot or something. For mixing projects you could put a pile of stuff in the common inventory and people could all dip into it. This way individuals don't have to try to balance their loads of food, ingreds, harvestables, etc. It would encourage team play, but fundamentally it wouldn't let you do anything you couldn't do by having people carry stuff individually or create hyper bags. There would be some details - like proximity to each other (you can't put one half the team next to hydro ore, and the other half next to store, with instant feeding of S2Es and ore back and forth). But you would have to allow for map changes (don't lose inventory just because the whole group doesn't click flag at same instant). If players leave the group or are killed perhaps their emu's worth of inventory gets dropped at their last position (so your team inventory isn't immune to being dropped if you're hit by PKers/chims/etc). There might also be practical control against mass-scamming, like the one new person in the group storage-trading the whole haul into their personal store - there are a few ways to do this.
  15. Invasion

    I think many would debate whether the current invasions are balanced, and right now you don't have any chance at all unless you're probably at least 50/50. What is the average active player a/d? I'd guess it is about 30/30 or so. Most invasion monsters as WAY out of league for these sorts of chars. I'm sure many will question the 30/30 assertion. With the number of newbies starting out every day it is probably about accurate - there are a lot more new people than established players. An invasion of goblins would be challenging for many players, and would outright slaughter many more. If you want to keep new people around you need to cater some of the content to them. (The same applies to introducing new items/monsters at lower levels as well as on the high end - the frying pan is a good example of this, as are vial/mercury manufacture.) Also - keep in mind the log scale of xp - a level 30 character is not halfway between 0 and 60 - it is WAY closer to 0. If you want to just take a simple arithmetic mean then figure the average xp and then convert that to a level. By having high-level and low-level invasions in two different places at the same time you give a lot more people a chance to participate. High-levels wouldn't be bored killing orcs, and low-levels wouldn't be bored just getting out of the way of orcs.
  16. Invasion

    A good idea. When I think of invasion the only thing that comes to mind is not to leave yourself AFK in votd. By the time I'm up to fighting cyclopses I'm sure they won't be spawning anything less than dragons... I do appreciate that a goblin invasion wouldn't be very interesting to the die-hards on the top-200 A/D list. Having different levels in different places might be a good answer. Maybe high-level invasions in places like C2 or KF/GP, and low-level invasions in more common places like votd/gold-mine.
  17. Channel #33 - market place for PLAYERS

    That is why I like the channel. I can just leave it open all the time, and it is low traffic. My only complaint is when people start a dialog on the channel - it really isn't intended for price checks or debate over prices. It works best if people just say "I'm selling x (optionally for y)" or "I'm buying x (optionally for y)." and that's it. By keeping it quiet lots of people stay tuned in, increasing the impact of your ads without having to broadcast them repeatedly. I love the convenience of bots and the fact that they'll buy just one of something. But if I have 1k SRs I doubt I'll ever get a good price for a bot. So each has its place, and they don't really compete much (which is why a bot can can find people willing to pay 4.5gc for an FE). And the market forum is great for high-value orders where getting the best price is most critical or cases where it can take days to find a buyer/seller. I'm glad that most bot owners have respected the coal of #33 - they've probably realized that anything that improves the economy just benefits everyone. Being able to sell your greaves to a human for a decent price just makes you more willing to buy iron bars for 50gc from a bot...
  18. Specific Bot market channel?

    Perfect example - until the player sends a perfectly formed command there is no way for the player to know what a perfectly formed command actually is. Hmm - as long as responses are sent via PM I don't think help messages will be a problem. However, responses should be limited to 1 line. Otherwise the user will get spammed with 5000 lines of "help" - with responses from various bots intermingled. Maybe the addition of a probability multiplier should be encouraged - ie a bot has a 10% chance of sending a help message (keep in mind that that virtually guarantees a response with 100 bots on-channel). Another issue is that because of implementation differences one bot might understand a command, and another might not. So, there are two ways to handle this and pros and cons to each. If non-understanding bots stay silent the user gets uncluttered reponses from the bots that do understand. If non-understanding bots reply then the user is aware that he didn't get all possible responses. Oh, if optional commands are to be supported then bots should avoid responding with help to anything other than "help" or something similar. Otherwise the use of optional commands will always generate errors from bots that don't understand them. Another option might be to have bots announce help on the channel VERY INFREQUENTLY - since the help messages will be similar no matter who is advertising. Maybe one message per 24 hours - which becomes a message every 15 minutes if there are 100 bots online.
  19. Specific Bot market channel?

    I really like this idea. If I want a good price on a common item I don't mind having to scroll through 200 lines of PMs. And if you don't really want to know about EVERY bot selling something just look at the last 5-10 lines. How much bandwidth could that text REALLY use? We're talking about maybe a few kilobytes, and how often are people going to run these queries if you hit every bot at once? I often buy or sell from bots in small quantities, and half the pain is finding one that has a decent price. I don't mind walking a little, but if I have 100 steel bars to unload for quick cash I'm not going to try to hunt down a human when some bots offer 45gc for them. It would be nice to find out where the good deals are quickly. Hopefully entropy will weigh in as to whether this idea is acceptable, and then bots can go ahead and implement this. If a particular bot owner doesn't want to no big deal - they just won't get as much business...
  20. Channel #33 - market place for PLAYERS

    A very good point. I've actually tuned into channel 3 while AFK for an hour or two just to catch a log of the latest tradebots. It is hard to know what is available otherwise. Actually, I'd really like it if somebody posted a list of tradebots and what they're selling. I'd prefer having pricing as well, but even if people balk at that it would at least be nice to know where to go to buy/sell something quickly in small quantities without having to PM every bot in the game and hunt through chat logs. And 33 has worked out very well - very low noise and I can just leave it open all the time...
  21. Garage SALE!

    1 Book of Steel Axe Construction 1 Book of Crafting Potion 1 Book of Magic Potion 1 Iron Sword Construction 1 Book of Orc Fighting 1 WhiteStone Ring Building 1 Portland Ring Building 1 Titanium Long Construction 1 Book of Puma Summoning 1 Book of Titanium Molding I'm potentially interested in these, but I don't have a lot of cash. I don't know the prices on each of these offhand, so feel free to drop any uber-expensive books on that list and sell me the rest in bulk. What would you consider a fair price? I already dropped the obviously-expensive books from the list. I'm looking to spend a total of 5-10k or so.
  22. Infinate Slots in Storage?

    Is there some huge objection to databases among the EL developers? In just about ANY project of the scale of EL that I've seen the data storage would be implemented in a database of some kind. It takes care of all the implementation issues related to storage for you. And you don't have to be worried that your pile of flat-files will end up with corruptions if you make a code change. It seems like very little time gets spent on databases in computer science programs, but in the real world there is a reason that Larry Ellison gets to play with his Gulfstream. In theory you can maybe beat the DB with careful optimization of design by doing things on your own, but in practice it will generally be better to spend those man-hours on something else. You'll never beat the DB in the general case - your only hope is to customize your storage design around the very specific needs of your project, and then be stuck when your needs change. A single storage-slots table would only consume space for slots that are actually occupied, so increasing the quota wouldn't cause problems with having tons of inactive accounts around. And it will be very fast if you index it correctly (which isn't really all that hard to do - just look at your queries). Even if you don't want to completely rewrite the backend to use a database throughout, why not just create a DB and start putting new stuff in it, and as issues come up migrate one table at a time to the DB. I'd think that it making the storage more dynamic in a flat-file-based system wouldn't be any easier than just modifying the code that reads this data to instead run an SQL query. I think that storage limitations are a good idea from a gameplay standpoint - it probably benefits newbies because it leads to stuff like med-level books being sold at firesale prices rather than just hoarded until somebody is willing to pay 500gc less than the NPC for them. It also causes players to have to make tradeoffs. When I think of EL being classless I think of it being open for players to change directions at any time. That doesn't mean that any EL player should be able to switch from being a fighter to a crafter instantly and end up being just as proficient in the one as they were in the other. What you focus on is going to be where you will be strong. So, I don't think storage limitations get rid of the classless aspect of EL. You can just have a storage sale and change directions without any real penalty. If EL were a classed game then your only option would really be to delete your character. So, while I think that storage limits should be set at some level for the sake of gameplay, I'm not big on the idea of limiting them due to software design issues. Sure, the practical side of things is that this simply has to be done until the design can be changed, but long-term the game should aim to eliminate technical barriers to ideal gameplay.
  23. Infinate Slots in Storage?

    Well, I can't speak for how it is actually implemented, but here is an estimate. Assume 10k accounts (identified by a 32-bit number). Each account has 1k slots in store. Each slot can hold a 32-bit number of items (thats 4 billion). Each item has a 32-bit item ID. You need an array containing: 32-bit userid, 32-bit item ID, 32-bit quantity. That's 96 bits per row, or 12 bytes per row. You have 10k accounts * 1k slots per account. That's 10M records. So, 10M * 12 = 120MB of storage for that array. I'm not a database guru, but figure another 120MB for indexing. Other than indexing by userID I'm not sure you really need anything else. Sure, 0.25GB isn't a trivial amount of space, but it isn't a huge amount of space either. Since most accounts are infrequently used you don't need most of it in RAM at any given time. Now, there might be something I'm missing, and it might be that the storage system was implemented differently and wouldn't be easy to change at this point. But at least in theory it isn't a major problem.
  24. Infinate Slots in Storage?

    How is storage implemented? Isn't it just a database table? If so how big could it possibly be? What is a few hundred thousand records between friends? My mythtv tables have 1.5M records and I'm hardly running on server-class hardware...
  25. If you're focusing on harvesting it isn't easy to level - due to the hourly limit. You could burn coins on harvesting medallions I suppose to get more xp in this area. I don't play nearly as much and for only about as many months, and I'm at OA 59. None of my individual levels are in the 50s, although I'm in the 40s on alchemy. Still, having levels in the 50s is nothing to laugh at - the higher levels take a LOT of time to advance. At those kinds of levels I'd avoid focusing on quick advancement and find other goals within the game - like using your levels to help others out who are newer...
×