Unit values
1, 2
posted on March 1st, 2011, 8:09 pm
Dominus_Noctis wrote:Same DPS... but when...
Unit A is being attacked by Unit B - Unit A regenerates faster than it can be killed.
Unit A is being attacked by Unit C - Unit A is now killed on the first volley.
However, if there are 3 of Unit C, they will all fire on Unit B and waste 2/3 of their firepower. Ships with high burst damage can break enemy regeneration but they pay the price of wasted damage.
Dominus_Noctis wrote:values themselves, even if designed to be totally linear, will betray themselves as having non-linear relationships on the battlefield.
Oh sorry, I didn't mean that the relationships are linear. The relationships between these numbers are incredibly complex and fun to work with.
I'm just saying I think the units of measurement should be kept as simple as possible, so they are easier to modify. Even if it means the equations are more complex, this way the complexity is all in one place instead of spread around...? If that makes sense.
---
By the way, I just want to mention that I love learning through debate, but people say I look angry sometimes when I'm doing it. I'm really enjoying this conversation.
posted on March 1st, 2011, 8:21 pm
Tryptic wrote:However, if there are 3 of Unit C, they will all fire on Unit B and waste 2/3 of their firepower. Ships with high burst damage can break enemy regeneration but they pay the price of wasted damage.
That's assuming of course that somebody does not micromanage those ships, and that there are more than 2 ships involved
. If we talk about 20 Sovies versus 20 LuSpet's, at a certain point balance breaks down 
Tryptic wrote:Oh sorry, I didn't mean that the relationships are linear. The relationships between these numbers are incredibly complex and fun to work with.
Ah, I see - but then isn't that also lying to the consumer, since the values themselves have different effects as you talk about a higher-level ship (as in that regeneration example)?
Tryptic wrote:I'm just saying I think the units of measurement should be kept as simple as possible, so they are easier to modify. Even if it means the equations are more complex, this way the complexity is all in one place instead of spread around...? If that makes sense.
Not sure what you mean by 'easier to modify' however - the formulae and changes seem to work out quite nicely in my opinion.
Glad you enjoy it
posted on March 1st, 2011, 8:51 pm
Well, it's becoming more and more apparent that I won't know what I'm talking about unless I'm allowed to see your graphs and equations. Maybe this entire thread is just me pulling a stunt to try to get the devs to show them to me. 
I didn't even know it myself, I swear!
Seriously, please go and try Warzone even if it's just once. I have never seen another RTS with a 15 second-delay-weapon (the tank killer) and it's fascinating to play with. You can devastate the enemy if you micro them all, but the faster chassis have crappy hitpoints and the slower ones can't maneuver well. Anyway the point I want to make is, 4-5 seconds may be longer than other weapons in Fleet Ops but I don't think it's long enough to cause serious problems.
I guess "lying to the customer" was too harsh. It's true that estimations will never be exact unless the recipient knows how you estimated it: for example system value could be linked to either percentage of the total hitpoints regenerated/second or actual hitpoints regenerated/second. Either one would technically be an accurate representation.
I do still have a problem with hitpoints being non-linearly tied to defensive value. This particular number is simple even if Offense and System can be shown to have complex relationships with the factors around them. If a larger ship has double the hitpoints of a smaller ship, it should have double the listed defensive value. Why would you want to make it different?
I didn't even know it myself, I swear! Dominus_Noctis wrote:That's assuming of course that somebody does not micromanage those ships, and that there are more than 2 ships involved
Seriously, please go and try Warzone even if it's just once. I have never seen another RTS with a 15 second-delay-weapon (the tank killer) and it's fascinating to play with. You can devastate the enemy if you micro them all, but the faster chassis have crappy hitpoints and the slower ones can't maneuver well. Anyway the point I want to make is, 4-5 seconds may be longer than other weapons in Fleet Ops but I don't think it's long enough to cause serious problems.
I guess "lying to the customer" was too harsh. It's true that estimations will never be exact unless the recipient knows how you estimated it: for example system value could be linked to either percentage of the total hitpoints regenerated/second or actual hitpoints regenerated/second. Either one would technically be an accurate representation.
I do still have a problem with hitpoints being non-linearly tied to defensive value. This particular number is simple even if Offense and System can be shown to have complex relationships with the factors around them. If a larger ship has double the hitpoints of a smaller ship, it should have double the listed defensive value. Why would you want to make it different?
posted on March 1st, 2011, 10:06 pm
I've now bookmarked it, but not certain if I'll have time/desire to play it in the near future
Yes, seeing the actual scripts would be very nice, but since they are scripts... means they have to be written to be releasable (which will happen).
I'm still not sure why you want to change Defensive Value, yet you are fine leaving System and Offensive value. Defensive Value has a large number of variables that are used to calculate it as well, and it is involved in a whole host of other descriptors . I guess I should just ask - what is the problem with the current system? Do you have issues comparing ships/stations?
Yes, seeing the actual scripts would be very nice, but since they are scripts... means they have to be written to be releasable
(which will happen).I'm still not sure why you want to change Defensive Value, yet you are fine leaving System and Offensive value. Defensive Value has a large number of variables that are used to calculate it as well, and it is involved in a whole host of other descriptors
. I guess I should just ask - what is the problem with the current system? Do you have issues comparing ships/stations?posted on March 2nd, 2011, 12:50 am
posted on March 2nd, 2011, 7:33 am
Last edited by Optec on March 2nd, 2011, 7:35 am, edited 1 time in total.
The attributes were introduced to messure usefullness, not to messure hitpoints. there are numerous aspects in a game that dont follow a simple linear equotation. actually, almost everything.
defensive value, for example, also has influence on shield regeneration. that means that the higher the defensive value is, the higher is the percentage of vessels, which dps gets completely compensated by the ships regeneration.
or another example. consider that the majority of weapons in the game would deal 20 damage. 20 additional hitpoints are much more important to your usefullness then, if you do just haved 10 hitpoints to start with, in contrast to a ship which already has 200 hitpoints.
there are countless other functions maping similar situations, which work in the background. we tried to move most of them out of the system layers which are directly in contact with the user - the attributes - but something can't be avoided. At the end of the day, we would not have introduced the attributes if they were just "divide hitpoints by 17". Directly showing the hitpoints would be more then enough.
In FO you can do both. you do have your attributes for beginners and fast reference (or if you just dont want to tinker with the math during the game - like me), if you want to do the theorycrafting behind for your favourite (or most hated) ship, you can take a look at the healthbar tooltip or the weapon tooltips.
Some values in the formulas seem odd, like the modificator mentoined above, but that's because they are not set by hand, but calculated from heuristical analysis of the rest of the game. just like real life, games are a strongly interconnected system. you can't balance it if you just look at a single vessel one by one. Or at least I cant
defensive value, for example, also has influence on shield regeneration. that means that the higher the defensive value is, the higher is the percentage of vessels, which dps gets completely compensated by the ships regeneration.
or another example. consider that the majority of weapons in the game would deal 20 damage. 20 additional hitpoints are much more important to your usefullness then, if you do just haved 10 hitpoints to start with, in contrast to a ship which already has 200 hitpoints.
there are countless other functions maping similar situations, which work in the background. we tried to move most of them out of the system layers which are directly in contact with the user - the attributes - but something can't be avoided. At the end of the day, we would not have introduced the attributes if they were just "divide hitpoints by 17". Directly showing the hitpoints would be more then enough.
In FO you can do both. you do have your attributes for beginners and fast reference (or if you just dont want to tinker with the math during the game - like me), if you want to do the theorycrafting behind for your favourite (or most hated) ship, you can take a look at the healthbar tooltip or the weapon tooltips.
Some values in the formulas seem odd, like the modificator mentoined above, but that's because they are not set by hand, but calculated from heuristical analysis of the rest of the game. just like real life, games are a strongly interconnected system. you can't balance it if you just look at a single vessel one by one. Or at least I cant
posted on March 2nd, 2011, 4:32 pm
Last edited by Tryptic on March 2nd, 2011, 5:23 pm, edited 1 time in total.
Hmmm.
So the real problem is that I'm trying to build the game using attributes without knowing the equations behind them, which is impossible. When trying to build a database for the ships I have to ignore the off/def/sys values and instead plug in hitpoints and weapon information. Those are the smallest building blocks required for me to recreate the odf folder, not the values.
So for the sake of my own work, I'll need to calculate intermediate DPS, EHP, and REGEN values to use the way I thought I could use off/def/sys. I'm sure your program uses these as well, they're just buried in the program and these days you don't touch them anymore, you let the program handle them. If I go ahead and develop my own program that far, I'll probably reach the same situation.
Of course, I may not go to all that trouble. I'm learning a lot with this project and it would be fun, but today I'm meeting with a business partner to discuss creating our first Android/IPhone game, and that project will take a lot of my time. It's gonna be so sweet though...
. . . 
. 
. 
. 
. 
. 
. ... 
I've waited so long to work on this stuff, the first 2 years of college teach you very little about computer programming. If my kid wants to do this someday, I'll seriously consider sending him to a tech school.
EDIT: future kid. I don't actually have any kids right now.
So the real problem is that I'm trying to build the game using attributes without knowing the equations behind them, which is impossible. When trying to build a database for the ships I have to ignore the off/def/sys values and instead plug in hitpoints and weapon information. Those are the smallest building blocks required for me to recreate the odf folder, not the values.
So for the sake of my own work, I'll need to calculate intermediate DPS, EHP, and REGEN values to use the way I thought I could use off/def/sys. I'm sure your program uses these as well, they're just buried in the program and these days you don't touch them anymore, you let the program handle them. If I go ahead and develop my own program that far, I'll probably reach the same situation.
Of course, I may not go to all that trouble. I'm learning a lot with this project and it would be fun, but today I'm meeting with a business partner to discuss creating our first Android/IPhone game, and that project will take a lot of my time. It's gonna be so sweet though...
. . . . . . . . . ... I've waited so long to work on this stuff, the first 2 years of college teach you very little about computer programming. If my kid wants to do this someday, I'll seriously consider sending him to a tech school.
EDIT: future kid. I don't actually have any kids right now.
posted on March 3rd, 2011, 5:45 pm
I just had a very revealing conversation with my new business partner (we're planning to create some IPhone/Android games together).
He explained to me how Open Source works, and that developers will often release the library itself for public use, but they don't release their tools on purpose. This way people can look at their code and use it in a limited fashion, but they can't modify it effectively. This ensures that while the code gets freely distributed, only the creators can take full advantage of their work and anyone wanting to seriously use the code will have to come ask for the tools as well.
In effect, the Flops files we download are deliberately difficult to work with, to prevent people like me from doing what I'm trying to do now: recreate the program using my own back-end software. I've been trying to break down a wall that you guys want to stay there. Otherwise it gets too easy for everybody to take your hard work and use it without staying in contact with you as part of the community.
It makes perfect sense and I don't mind, I was just getting frustrated because I didn't understand the situation.
In any case, it looks like this new project of mine will be enough to keep me busy. I'll put my Fleet Ops modding plans on hold until such a time that I reach an agreement with the devs to access their tools in addition to the core Flops files. If that never happens, it's cool.
For reference, my plan was to create a more fast-paced version of the game and call it Strategic Ops or something. In it, all small ships would get cheaper and slightly weaker and all large ships would get more expensive and stronger. Construction times would increase and yard costs would get severely reduced, to create a more Starcraft-esque style where 10 intrepids go against 6 rhienns and 1 D'deridex in an even fight, and only the larger units warrant paying enough attention to micro them around while the smaller ones are controlled in groups. It was just an idea of mine to play around with.
He explained to me how Open Source works, and that developers will often release the library itself for public use, but they don't release their tools on purpose. This way people can look at their code and use it in a limited fashion, but they can't modify it effectively. This ensures that while the code gets freely distributed, only the creators can take full advantage of their work and anyone wanting to seriously use the code will have to come ask for the tools as well.
In effect, the Flops files we download are deliberately difficult to work with, to prevent people like me from doing what I'm trying to do now: recreate the program using my own back-end software. I've been trying to break down a wall that you guys want to stay there. Otherwise it gets too easy for everybody to take your hard work and use it without staying in contact with you as part of the community.
It makes perfect sense and I don't mind, I was just getting frustrated because I didn't understand the situation.
In any case, it looks like this new project of mine will be enough to keep me busy. I'll put my Fleet Ops modding plans on hold until such a time that I reach an agreement with the devs to access their tools in addition to the core Flops files. If that never happens, it's cool.
For reference, my plan was to create a more fast-paced version of the game and call it Strategic Ops or something. In it, all small ships would get cheaper and slightly weaker and all large ships would get more expensive and stronger. Construction times would increase and yard costs would get severely reduced, to create a more Starcraft-esque style where 10 intrepids go against 6 rhienns and 1 D'deridex in an even fight, and only the larger units warrant paying enough attention to micro them around while the smaller ones are controlled in groups. It was just an idea of mine to play around with.
posted on March 3rd, 2011, 6:18 pm
Last edited by Dominus_Noctis on March 3rd, 2011, 6:41 pm, edited 1 time in total.
Like I said before, the tools are undergoing a rebuild to be released, it will just take a lot of time 
.
This is why there were so many bugs for 3.1.5 in relation to passives, disabling, etc.
The Fleet Ops tools are heuristic and have an "AI" which assess variables globally as I understand it, and it is extremely difficult to separate the functions without exporting the entire, very hefty, system.
This in fact follows pretty much all development of Fleet Ops - in that it is very difficult to separate out each of the systems used (i.e. for tooltip creation, ODF parsing)
If Optec had not wanted the tools to be released in the first place, I would never have been able to build the unit values system in the guide.
EDIT: typos
.This is why there were so many bugs for 3.1.5 in relation to passives, disabling, etc.
The Fleet Ops tools are heuristic and have an "AI" which assess variables globally as I understand it, and it is extremely difficult to separate the functions without exporting the entire, very hefty, system.
This in fact follows pretty much all development of Fleet Ops - in that it is very difficult to separate out each of the systems used (i.e. for tooltip creation, ODF parsing)
If Optec had not wanted the tools to be released in the first place, I would never have been able to build the unit values system in the guide.
EDIT: typos
posted on March 3rd, 2011, 7:41 pm
aye, evil Optec doesn't like somebody else to mess with his files. I wonder why we still did not package all ODFs into an encrypted container.
Hm, now that i think of it, that would also solve Dominus poking me about every single flaw in any odf!
Hm hm, ...
it's on todo!
I wonder why we still did not package all ODFs into an encrypted container.Hm, now that i think of it, that would also solve Dominus poking me about every single flaw in any odf!
Hm hm, ...
it's on todo!
1, 2
Who is online
Users browsing this forum: No registered users and 1 guest