Horizons A very tangible reason for why storage is a bad idea



I just remembered my micro-economics class in college after I read the following links:

This one was a very intersting read: http://www.mine-control.com/zack/uoecon/uoecon.html

And this is similar, but is designed to go with a presentation and focuses more on the problems and solutions: http://www.mine-control.com/zack/uoecon/slides.html

This was posted by Agony-Aunt in response to a request for Player Economy. The coorelation with storage is the link between hoarding and the game's ability to keep running smoothly, i.e. server response time and game lags. The trade off is something one should consider before this is implemented. People are inherently pack-rats. I mean that when we think we are going to need something, i.e. socks for example, we don't buy just one pair, we buy several because, we can't wear the same pair every day. But some people when they need a blank cd buy 50 instead of 10, because they say in their minds that I might need them. This is a hoarding mentality.

The problem with this mentality and a video online game is that there is only so much bandwidth and so much server space and virtual memory (cloud storage) to handle the mass of stuff in the game. When this is compounded with several hundred thousand players all collecting "stuff" and storing it, well... you see instantly where this can lead. The last posted link in the quote from Agony-Aunt is an outline to the paper discussing the Ultima Online economy and the problems that crept in to the game's performance factors. Something to consider.
 
Last edited by a moderator:
Reasonable points however this is easily overcome with predefined limits as with materials now so not really an issue. Also possibly a big money maker for frontier if they were to sell storage space not that i,d recommend that approach.

Just a little space for each player would be fine. I,ll open with 200 tone commodities and 50 ship modules. much less server load than engineers uses now but would solve many problems and add a new dimension to the game instantly.

This is just an example of course.
 
What if FD did make storage space a store purchased item... Would you pay for it and if so what do you think the value to storage ratio should be?

I ask as you raise a very interesting point... :)
 
I did consider the predefined limit option, but as we have already seen, enough complaining brought our ship limit from 500 to 1000 on the MATs that our ships can hold. Of course, much of this space is taken up by useless carbon, sulfur, phosphorus and lead, which can be used if you need to constantly reload your missiles in space.
 
Last edited:
I just remembered my micro-economics class in college after I read the following links:



This was posted by Agony-Aunt in response to a request for Player Economy. The coorelation with storage is the link between hoarding and the game's ability to keep running smoothly, i.e. server response time and game lags. The trade off is something one should consider before this is implemented. People are inherently pack-rats. I mean that when we think we are going to need something, i.e. socks for example, we don't buy just one pair, we buy several because, we can't wear the same pair every day. But some people when they need a blank cd buy 50 instead of 10, because they say in their minds that I might need them. This is a hoarding mentality.

The problem with this mentality and a video online game is that there is only so much bandwidth and so much server space and virtual memory (cloud storage) to handle the mass of stuff in the game. When this is compounded with several hundred thousand players all collecting "stuff" and storing it, well... you see instantly where this can lead. The last posted link in the quote from Agony-Aunt is an outline to the paper discussing the Ultima Online economy and the problems that crept in to the game's performance factors. Something to consider.

Interesting read. Remember me the read about EVE Online player driven eco. I will finished the article but I have to underline the argument about storage and the server response time: It's not relevant today I think for several reasons:
1-- The article talk about Ultima Online in 1999. The storage problematic was real at this time. Today it's not. The server's storage capacities and the clients storage capacities are way more important. The data to store is still the same for objects, merely some bits of references. It literally the equivalent of a text file with a thousand lines. Not really bigger than in the 90's and very small to be handle by a server cluster or a client's computer.
2-- It's a guess but the server of ED host a billion stars galaxy. That's a billion references of star's system with in each several corpses of different natures: It's several billions of references. And yet the server handle that very well.
3-- ED has no player driven economics. There is no real interaction between star system beyond few jumps distance. Nothing like EVE Online or even Ultima Online. Beyond that there is no crafting in ED. The engineers modifications cannot be trade in any way with other players or even with the system (a modified module is sold at the same price of an unmodified one to the system. It's just a money/time sink from this point of view).


The paper is interesting but it doesn't apply to ED for me. The great difference between Ultima Online and ED is their different nature: One is a MMORPG by the classic definition of the term and ED is not (no crafting, no trading between players, no community tool gathering/researching a.k.a guild/corporation, no housing of any kind). I often read post in which ED is qualified as a MMORPG. For me it's not (that's not good or bad in anyway). The only feature it has now which is in a MMORPG is the grouping (a.k.a wings).
 
I would add the following about server bandwith and player storage interaction:

In EVE Online the Devs were constantly, and still are, facing bandwith problem in multiplayer battles. Why ? Was it because of player storage references ? No. It was because of the real time nature of a battle. Let me explain:
In a battle, thousands of players reunite and fire at the very same time. This is several thousands transmissions towards EVE Online server at the same time which requires thousands of responses from the server in the same time frame. This is when the bandwith role become apparent. The dialog between player's computer and the server require precise timed responses.

When you talk about player storage the bandwith is not really a problem. It could be with a huge number of players at the same time but if you reach that point it will merely be one of many other problems ;-)
Interaction with your inventory doesn't require precise timed responses from the server or the client. Of course there is a limit to ensure playability. Nobody wants to wait several seconds each time he move one object from his inventory in his cargo for example (or 10 seconds for the billboard to appear ;-) ). So you can have a huge inventory with a thousand references in it, the bandwith won't be a problem because not all the players will engage in heavy inventory interactions at the very same time. If it happen, the server will just scale it's responses to the clients: You will see the results of your interactions several milliseconds later than usual in order for the server to respnd to every player. But for an inventory interaction who cares ? You won't even notice.

In short: Bandwith is important for real time (a.k.a combat). Storage doesn't require high bandwith.
 
What if FD did make storage space a store purchased item... Would you pay for it and if so what do you think the value to storage ratio should be?

I ask as you raise a very interesting point... :)

Well Mr Mallory i,m almost loathed to answer but since you pose the question. If Fdev offered me tomorrow 200 commodities and 50 modules of easy accessible storage Id throw 5 of your English pound at them now, although as i said i don't think many would agree with that approach. But lets face the reality where desperate for such facility right now. my main point is the OP,s post although well considered is not applicable to this scenario be it paid or free.
 
I would not want storage for all commodities. Only for those not available at the common market and related to the engineers.
And there could be a limitation for them. They could be handled like materials and data.

Module storage is definitely a requirement now with the engineers.
Because of the RNG we can not easily reproduce a successful module and creating them demands a lot of time investment.
We desperately need a way to store our favorite engineered modules for later use.
In this case I am ok with a reasonable limit.

If, for technical reasons, FD is unable to provide module storage (which is a big priority for me) then they should change the way we engineer our modules so they can be recreated easily.
I am not against the RNG, but without module storage, the RNG should be removed.
 
I don't really know what they mean by "storage" (what, how, how many).

But if it's just having a place (rent or buy with in game credits) where we could put some commodities,
50 or 100 for example, it would be nice. A good limit, IMHO, could be : one possible rented storage in
each engineer base as I can't see anything else really requiring storage. Mats are already account
virtual goods and, as it's been already said, I also think the "keep it, in case !" way is wrong.
 
Storage can actually be bought today; buy another account and you're basically all set. My storage Imperial Cutter is now a very valuable storage containing weeks of hard to get mission-only commodities. I hope it doesn't blow up...
 
This was posted by Agony-Aunt in response to a request for Player Economy. The coorelation with storage is the link between hoarding and the game's ability to keep running smoothly, i.e. server response time and game lags. The trade off is something one should consider before this is implemented. People are inherently pack-rats. I mean that when we think we are going to need something, i.e. socks for example, we don't buy just one pair, we buy several because, we can't wear the same pair every day. But some people when they need a blank cd buy 50 instead of 10, because they say in their minds that I might need them. This is a hoarding mentality.

The problem with this mentality and a video online game is that there is only so much bandwidth and so much server space and virtual memory (cloud storage) to handle the mass of stuff in the game. When this is compounded with several hundred thousand players all collecting "stuff" and storing it, well... you see instantly where this can lead. The last posted link in the quote from Agony-Aunt is an outline to the paper discussing the Ultima Online economy and the problems that crept in to the game's performance factors. Something to consider.
I think that was posted there to point how an market works. As somebody else said, in our days with so many data storage possibilities this will hardly or never be a problem.

Reasonable points however this is easily overcome with predefined limits as with materials now so not really an issue. Also possibly a big money maker for frontier if they were to sell storage space not that i,d recommend that approach.

Just a little space for each player would be fine. I,ll open with 200 tone commodities and 50 ship modules. much less server load than engineers uses now but would solve many problems and add a new dimension to the game instantly.

This is just an example of course.
+ especially for ship modules cargo :)

What if FD did make storage space a store purchased item... Would you pay for it and if so what do you think the value to storage ratio should be?

I ask as you raise a very interesting point... :)
Even if I think it wouldn't be right to pay for it because this should be a basic feature in the game, I will finally buy it :p

I would add the following about server bandwith and player storage interaction:

In EVE Online the Devs were constantly, and still are, facing bandwith problem in multiplayer battles. Why ? Was it because of player storage references ? No. It was because of the real time nature of a battle. Let me explain:
In a battle, thousands of players reunite and fire at the very same time. This is several thousands transmissions towards EVE Online server at the same time which requires thousands of responses from the server in the same time frame. This is when the bandwith role become apparent. The dialog between player's computer and the server require precise timed responses.

When you talk about player storage the bandwith is not really a problem. It could be with a huge number of players at the same time but if you reach that point it will merely be one of many other problems ;-)
Interaction with your inventory doesn't require precise timed responses from the server or the client. Of course there is a limit to ensure playability. Nobody wants to wait several seconds each time he move one object from his inventory in his cargo for example (or 10 seconds for the billboard to appear ;-) ). So you can have a huge inventory with a thousand references in it, the bandwith won't be a problem because not all the players will engage in heavy inventory interactions at the very same time. If it happen, the server will just scale it's responses to the clients: You will see the results of your interactions several milliseconds later than usual in order for the server to respnd to every player. But for an inventory interaction who cares ? You won't even notice.

In short: Bandwith is important for real time (a.k.a combat). Storage doesn't require high bandwith.
Exactly what are you saying. In EvE some time ago was a pain in the ass to go into an 200 vs 200 players fight because of the lag and that was fixed as time passed and some good technical upgrades. But is still is a problem now when so many players can fight each other (and I am talking thousands vs thousands or something like that) but storage was never a problem. I still have millions of stuff in a station there and many many other stuff in different ones and I am just too little, others have impressive amounts of ore and modules and this never ever been a problem. From what I know and remember a bit but I might as well be very wrong, item storage is on a different server from the main one but don't get this for real :)
 
I don't think database storage space or bandwidth are quite the problems you think they are. If there are 100 types of commodity, you'd probably store that as 100 x short integers. That's 200 bytes of storage. Transferring that would result in slightly more bandwidth because it's transferred in JSON which is effectively ASCII but it's still not large. Consider how much information the game is already transferring in terms of ship storage (with all the outfitting, fire groups, module settings). Consider the fact that all the materials are already stored this way. Consider the state of the game, where you are in space, how much health you have, what missions. Cargo storage is a drop in the ocean compared to that lot..
 
Last edited:
I just remembered my micro-economics class in college after I read the following links:



This was posted by Agony-Aunt in response to a request for Player Economy. The coorelation with storage is the link between hoarding and the game's ability to keep running smoothly, i.e. server response time and game lags. The trade off is something one should consider before this is implemented. People are inherently pack-rats. I mean that when we think we are going to need something, i.e. socks for example, we don't buy just one pair, we buy several because, we can't wear the same pair every day. But some people when they need a blank cd buy 50 instead of 10, because they say in their minds that I might need them. This is a hoarding mentality.

The problem with this mentality and a video online game is that there is only so much bandwidth and so much server space and virtual memory (cloud storage) to handle the mass of stuff in the game. When this is compounded with several hundred thousand players all collecting "stuff" and storing it, well... you see instantly where this can lead. The last posted link in the quote from Agony-Aunt is an outline to the paper discussing the Ultima Online economy and the problems that crept in to the game's performance factors. Something to consider.

As a developer myself, it breaks my heart whenever I see loyal fans/players trying to justify flaws of games (such as the no-storage issue in ED) bringing up issues/trade-offs that are sadly not the case.

The costs of implementing a storage system for cargo and/or modules will be the SAME as an existing feature: Ship storage.

We don't have storage now because the developers are holding it back due to whatever reasons, one of them being they don't think it is game breaking without storage. Which it isn't IMO, just a very bad user experience compounded with other grindy features *coughengineerscough* designed to draw out game time.
 
Reasonable points however this is easily overcome with predefined limits as with materials now so not really an issue. Also possibly a big money maker for frontier if they were to sell storage space not that i,d recommend that approach.

Just a little space for each player would be fine. I,ll open with 200 tone commodities and 50 ship modules. much less server load than engineers uses now but would solve many problems and add a new dimension to the game instantly.

This is just an example of course.

To be honest I would go less then that. Maybe 50t of commodities storage and 10 modules. You can still store modules on ships as well. The reason for only 50t of commodities storage is the same reason why FD are not keen on it, so we don't stockpile a load of gold for a CG or flood the markets etc. 50t should be enough as most of them can be bought anyway, so buy them when you need them and store the few that can't be bought.

Materials storage should be around 2000, but reduce the in ship amount back to 600, so there is some management still needed.

Also you need to pay for this storage as well. I also think we should be paying for our ship storage too, the bigger, the more expensive it costs.
 
Commodity storage could be handled by the Engineer giving you 32T space at their base when you unlock Grade 3.
As in limited, but you can specifically store the commodities that you want to use with that engineer at a later date.

Module storage is a bit more complicated - possibly a better option is to add an option in outfitting that shows you what modules are fitted on the other ships you have parked there with a swap function.

Materials storage needs to come in as well - the magic backpack is a bit naff.
Mats should be destroyed with your ship but only if there is somewhere else to keep them.
 
Last edited:
lots of ingredients + crafting = huge need of storage.

It will always follow this road no matter what, now regarding technical issues i leave it to you guys.
 
Without better knowledge of how the program works, any issues with cargo storage is an unknown. We can only guess what obstacles surround the storage of such data and the code that manages it.

My suggestion to the engineer upgrade commodities problem of cargo-blocking on ships, was that we be able to deliver cargo to Engineers in advance of using it.

Unless pre-delivery of cargo to Engineers is possible, upgrading ships with cargo-dependant mods is impossible on ships without cargo racks.

Module storage is another issue.

As long as we cannot store modules, module rebuy cannot have depreciation in value. Also, if we can't deliver to Engineers in advance, then the only alternative is takes modules as cargo and swapping from cargo to outfitting and back.

A solution to both could be two changes.
Store ships with cargo (this defines a maximum on what can be stored).
Transfer modules from outfitting to cargo and back.

Why?

You could buy a Class 5A Frame Shift Drive in a Cobra Mk III and "equip as cargo" (20T), take it along with the materials needed and the 1 x Magnetic Emitter Coil as cargo to an Engineer for the Level 5 FSD upgrade.
Return to your home station and "Transfer cargo to outfitting"
Swap to another ship, such as a Federal Assault Ship (which may have no cargo racks) and equip from outfitting as you do when you sell and rebuy currently.

Any of these options would resolve some of the current road-blocks and don't differ wildly from what we already have in the game.

Can it be done or not? Only the devs can say.
 
To be honest I would go less then that. Maybe 50t of commodities storage and 10 modules. You can still store modules on ships as well. The reason for only 50t of commodities storage is the same reason why FD are not keen on it, so we don't stockpile a load of gold for a CG or flood the markets etc. 50t should be enough as most of them can be bought anyway, so buy them when you need them and store the few that can't be bought.

Materials storage should be around 2000, but reduce the in ship amount back to 600, so there is some management still needed.

Also you need to pay for this storage as well. I also think we should be paying for our ship storage too, the bigger, the more expensive it costs.

Well yes i would take whatever fdev were willing to give us, but i can't agree that 200tones of gold or any common commodity is a stockpile i can pick that up in 2 seconds in a python or bigger with very few exceptions 200ton would be irreverent to the BGS and the server, fdev are holding back for a reason we as yet don't know, if it,s because of they see it as unimportant to the player base then there very very wrong. Technical reasons (NO) Negative effect on game (NO) Think it,s unimportant compared to other things (Probably) we need to help them understand there wrong and get it done.
 
Last edited:
I'm sure that, when FD starts making online game, they knew that they will have to deal with cost of servers and bandwidth. FD isn't a one person start up with a server powered by potato.
spudsclose.jpg
 
Back
Top Bottom