Ranger Archive
Thread: Curing, Tanning, Preparing Hides/ Meat/ Bones for selling.
It think something along the lines of this past is worth persuing. I would like to ask the devs about the technical aspects of this before we go over the philosophy. We can be 100% perfect in our clear and concise reasoning behind this, but it is all for naught if they can't/won't change things in the database. What technical questions would you like me to ask? I see the following:
- How would refining a resourceaffect the DB?
- What kind of effort would be needed to make a craftable item be able to be used in a resource slot ina schematic?
- Others?
Hakai wrote:
777erc wrote:
Perhaps we could require this process to use camps instead of factories
oh man that.....is just.......wow...i like!!
Idea comming.....
FB's and HTFB's have an option to leave a "trade post", or just a new and seperate entity where we can sell organics on a vendor... or some such silliness 
It would be very cool to process all that junk organics into flat 900s though, or take an ALMOST perfect resource that you be grind quality due to a bad OQ or SR and make it sellable
What you are asking is impossible in terms of what can realisitically be done with real-life code in a working application. If you had unlimited money and labor you could rewrite the game to accomodate it -- that's about what it would take.
Changing assumptions about fundamental datatypes is like trying to change the foundation of a house -- after the house is built.
Should we shut down the conversation for technical reasons? Absolutely. We have much, much bigger problems than this to deal with.
DFH wrote:
They might not be able to do everything but there is nothing here that is entirely impossible. Difficult to do, yes.
What you are asking is impossible in terms of what can realisitically be done with real-life code in a working application. If you had unlimited money and labor you could rewrite the game to accomodate it -- that's about what it would take.
Changing assumptions about fundamental datatypes is like trying to change the foundation of a house -- after the house is built.
Should we shut down the conversation for technical reasons? Absolutely. We have much, much bigger problems than this to deal with.
Remember that the Devs add a new resource every week for most types. There is nothing stopping them from adding a new resource whenever someone decides to make one. That code is already there. The problem is that it is unlikely to be effecient. If you had a hundred new resource types per shift then you will run into problems. Actually making a new resource is done every week. And making a stack of that resource is done every time you harvest. They can do it in a brute force method like this. Without needing to change much of the code. But it will cause a huge increase in DB requirements.
Lythender Nirou
Master Ranger/Master Rifleman
Eclipse Server
I have seen enough of system design to know what is and isn't feasible at each stage of development. There are decisions that, once made and acted upon, are essentially impossible to revisit. Since software is not a tangible thing it is easy to believe it can be changed at will, but that is just not the way it works. It would be one thing if the game was at the design stage and including resource refinement meant redrawing some Booch diagrams or rewriting some UML. Including it now very likely means changing assumptions at the core of the code, and you just don't do that lightly and preferably not at all. It is possible that I am wrong and the system was implemented in a more flexible fashion but I very much doubt it. Frankly, the DB part of SWG seems to be held together by little more than faith and I don't think it would be wise to try to tamper with it in this fashion.
Finally, the devs don't have enough hours in the day to respond to every crackpot proposal that comes down the pike on these forums. If you have been paying attention you'll notice that most threads of this nature draw no offical response at all. If you're expecting a dev to come along and give you an opinion on this, I suggest you have a seat and find a good book. The The Mythical Man Month by Fred Brooks would be a good start.
I like the idea of refining our harvested resources. Some notes:
(1) I share DFH's concerns about the limitations in the code. Multiply the number of base resources by the number of rangers running refinements by the number of refinements each ranger will make and I think the servers will craft themselves some legs using our advanced resources and come to our houses and kick us in the butt. Is it equally difficult to allow for craftable items to be used in a resource slot in a schmetic? This is the part that will likely kill the dream. (But since I don't have to rewrite the code, I say get cracking!)
(2) This will cut down on dabbling. One simply does not dabble in a crafting class where quality counts in the final product. Who dabbles in weaponsmith? An armorsmith will only want high quality hides from a MR. Tailors and Architects can buy from scouts and ranger dabblers,whileArmorsmiths, Chefs, BEs and Docs will buy from Master Rangers.
(3) High quality products will almost always be available.
There was a (4) but I can't remember what it was. I guess there is NO! ... number (4).
Holy moly! A unique ability that would actually make rangers useful for a change!!!
It's a great idea. And I'm sure it will just get lost in the shuffle like all the other great ideas we've presented over the past year.
DFH wrote:
Not sure where the static variable point comes from.
All the fields for a resource except the quantity -- type, name, attributes -- are almost certainly static. (If they're not, they should be.) That means you can't just edit the (say) OQ of a resource -- you have to generate an entirely new resource type. Creating a new data type is a bigger deal than creating an instance of an already-extant data type, even one that has dynamic fields.
To do it right you wouldn't have a static setup, you would have to have a record for every permutation of all the stats... one for OQ 1, PE 1000, one for OQ 2, PE 1000, etc... I would make a table full of all the spawned resources, and the stats of that table are kept unchanging from day to day, but all it would take is adding to that table another row with the new stats.. a new record isn't that hard to add to a database... but I could see it becoming oversized from what they want for a given server resource list, but that can be worked with as well.
But in short, its not a new data type, merely a new record.. something they do all day long as stuff gets made and changed...
Yes database bloat... but we have that already.... witness the lag issues, and rollback problems on server crash...
The output of the refining process can't be a crafted item, because the crafting process is looking to have objects of type Resource inputted. Even the UI for crafting is looking to have Resource objects put into the boxes for required resources. Depending on the particulars it might be possible for Demon Hackers From The Bowels of Heck to kludge in a routine that intercepts the Refined Resource objects before the crafting routines see them and makes them look like Resource objects, but a) I wouldn't bet on it, b) that's not the kind of thing you really want in big commercial software that has to be maintained, even though the game is probably full of that kind of code already, and c) it would still be a ton of work. Either way, it means database bloat, a lot of new code, too little benefit to the game and a good chance of screwing with game balance.
This one's a non-starter.
Herbivore meat (e.g.) has four attributes: DR, FL, PE and OQ. Each one can range from 1 to 1000. That means there are 1000^4 or 10^12 possible types of Herbivore Meat. That's 1,000,000,000,000 new records you want to add to the database. Since some schematics (notably ours) care about planet of origin, that's per planet as well, currently 10. Right now, the database only has to keep a record for every type of Herbivore Meat that has ever been generated. (It could, in principle, drop records for types that have no instances. I sure hope it does, but it's possible their garbage collection isn't up to snuff.) The game has been live since...June? July? of last year...call it ten months, 45 weeks. Resources shift about once/week (less often in many cases, but we'll use that for an average), and there are ten worlds, each with its own type of Herbi, so a ballpark estimate is each server only has to care about 450 or so types of Herbi, total, for all worlds. I feel safe in saying the increase being discussed qualifies as "oversized".
And yes, I know you wouldn't have to generate all 10^12 possible types at once. You could add them as players create them. But one Ranger who tries ten times to cure his Naboo Herbi just created ten more records. The community of Rangers, small as it is, could easily generate more records in one day than the server has had to generate in its commercial lifetime. Again, you can prune ones that have no instances, but you're going to be multiplying the number of instances as well. This isn't doing pleasant things to your garbage collection times, either.
And that's just one type of meat. There are ten types of meat, plus milk and eggs. It's hard to imagine a process for refining eggs, but milk could be pasteurized, homogenized, fortified.... And then there's hides and bones, which have five attributes, meaning 10^15 possible records times ten worlds times number of hides (4) and bones (2). And then the Artisans are going to want to refine their resources, so make it every resource in the game. And all of this for what benefit? So you can improve a resource by maybe 10%? The devs aren't going to make a system that lets you turn OQ 300 meat into OQ 900 meat. The high-OQ stuff is rare because it is intended to be rare. You're not supposed to have products made with quality 900+ resources all of the time.
Had enough? I certainly have.
But in short, its not a new data type, merely a new record.. something they do all day long as stuff gets made and changed...
That's the whole point. Given that the current data type for resources is static, in order to implement refined resources you have to create new records, and (as shown above...again) that is prohibitively expensive. Correctly implementing resource refinement means changing the data type for resources to allow the attributes to be dynamic, or creating a new data type for refined resources -- both of which would also have costs in terms of memory and performance. And in the latter case, the UI for crafting and probably every schematic in the game would have to be modified to accept the new data type. This is the part where the programmers come to your house with pitchforks and torches. (Oh, what am I saying. SWG is made by SOE-Austin. They'll bring guns.)
when someone takes the time to come up with a unique and original idea I think we should be supportive.
This idea is not unique or original. It isn't even good. It was, as others have pointed out, rejected back in the early stages of beta largely for the reasons given above. Like the idea of mechanical farms/harvesters for creature resources (also rejected in beta) it keeps coming back, though, like the ontological proofs for the existence of god that every generation has to see refuted for itself.
Rancorrider4 wrote:
"This idea is not unique or original. It isn't even good. It was, as others have pointed out, rejected back in the early stages of beta largely for the reasons given above. Like the idea of mechanical farms/harvesters for creature resources (also rejected in beta) it keeps coming back, though, like the ontological proofs for the existence of god that every generation has to see refuted for itself."
(Best Simpson's Comic Book Guy voice) "Worst..post..ever."
I give up. I have finally met someone in the Ranger forums I just don't like. Continuing to argue this is pointless. I like the idea, Calc likes it, Nemo likes it, and so do several others. You are the lone (I could be wrong, but certainly the loudest and most negative) dissenting opinion. I have not seen the idea and Ive been playing since Beta 3. If the Dev's say no then so be it.
By the way there is a God...PM me some time and I'll tell you about him. It's easier than understanding this programming you keep talking about.
Just to make it clear, I completely understand where DFH is coming from in his posts. I am just saying that there are ways that could make it possible. For example, it was recommended that there only be an increase based exactly on the level of the Ranger doing the upgrade. If you limit that to 3 points (novice, half way up a tree, and Master) then the database bloat is not nearly as severe. There are intelligent ways to do something like this. Whether or not it is a good idea, I'm not sure. What a lot of people fail to take into account is the fact that the Devs want good resources to be rare. That's the whole reason they shift in the first place. I just think that the idea is worth looking into a bit.
As it is, the database bloat for a single piece of armor is similar to the bloat for a new resource (the armor has serial number, effectiveness, special resists, encumberance, and durability [current and max]--that is a similar number of variables stored as a stack of meat would have). But those variables are stored with the armor item. A new resource is likely to be stored in a different location. The resource type needs a global storage location because it can exist on any world at the same time (splitting stacks and such). The armor can be stored on a more local server. Having too many different resource types might cause too much strain on a global server (assuming they have such a distinction to begin with). The actual resource stack almost certainly just stores a number of resources and points to the global resource. It is also unlikely that they have a decent cleanup method for the resource types (although they almost certainly do for items like armor or weapons). Making a new resource is not much different than making a newpiece of armor or a new weapon. But the difference could be enough to cause problems. This is a point I am willing to leave to the Devs. They know more about the system than I do.
As for good resources needing to stay rare, look atthe crafting changes the Devs were looking to add. They were trying to make it so that maxing item stats was hard. Adding better resources is the reverse of that policy. While almost any crafter you ask would love to be able to get better stats for no reason, the Devs have their reasons for keeping good resources rare. I agree with the Devs on this matter. I don't like the attitude where everyone wants to be uber. I do understand it, though. Such a change is leaning towards the uber end of things. It is also something that would give Rangers a distinction from Scouts.
Technically, I am willing to argue that the idea could potentially work. Practically, the idea has some merits. Overall, I think the idea is worth discussing. I can't claim to be entirely for the idea as I'm one of the few people out there who actually dislikes all the really nice items out there--they are just too common now. But I do think that some good ideas could come from this. I know that the crafters out there would love this idea (anything that gives them the ability to make better stuff will keep them happy). I also think it could help to improve the economy (in its current state--I think there are much better ways to go about that but this really isn't a thread to go into that). Most players would love to see this sort of ability added into the game (as they want the best stuff they can get). But I can't say that it would actually end up improving the game (yet).
Lythender Nirou
Master Ranger/Master Rifleman
Eclipse Server