Latest Posts

Topic: Frisian Balancing

WorldSavior
Avatar
Joined: 2016-10-15, 04:10
Posts: 2091
OS: Linux
Version: Recent tournament version
Ranking
One Elder of Players
Location: Germany
Posted at: 2018-06-01, 01:33

hessenfarmer wrote:

Depends on further arguments. Right now I'm only convinced that the frisian coal mines could need a speedup.

(For me the most important issues are normalizing wares per ressource to 1 for all tribes,

Well, resources have different costs, and the numbers of wares could also stay like they are.

What I meant is that one mining ressource in the map should only deliver one ware for each tribe which is currently not the case. I thought this was the starting point of your project to fix the mines. However I maybe wrong.

No, you are not wrong, I just mixed up your definitions.

have a look in the very fast coal production of the Atlanteans,

The Atlantean coal mine is slower than the Barbarian deepest, and the imperial deepest is not much slower. So I think that it's okay...

Ok I see your point. it is not that important. However as Atlanteans don't need to upgrade their mines they should be slightly slower than the fastest improved mines of the other tribes. Don't you think. But this is only a small change with possible neglegible effect.

Not necessarily. Atlantean mines need three workers, and deepest imperial mines need only two workers - and are probably cheaper.

the barbarians dep iron mine seems to be faulty,

Well, it is very inefficient, but one can manage to skip it always, so I don't think that a change is necessary

compared to the deep goldmine it just seems to be faulty. i would have expected it to have comnparable if not slightly better values than the deep goldmine. At least this would fit the other tribes and other improvement stages. And the possibility to skip the mine is not a solution as most players are not aware of this fact.

I think it's interesting that there is also a gold mine which is more efficient than its iron "twin". If someone uses a barbarian deep iron mine, it's his own fault. Everyone can just inform himself about the efficiencies of mines and conclude that the deep iron mine has to been skipped... Cleverness will be rewarded here.

and increase frisian coal production per second a bit.)

I didn't plan to change frisian mines, but when I change other mines, i can maybe also change the frisian ones. I could tell first what I would change, and maybe I could also change other buildings.

my idea was that we shouldn't do changes in the same area in parallel. So in my opinion the mining topic fits better in your project than in mine as it is not a purely frisians problem.

Okay

Theoretically there can be comma numbers, for example if two wares are transferred into one.

what I meant was we can't have it in the current implementation of the fur recycle loop.

Have you thought about what happens if barbarians and frisians play in the same team? The barbarians could build big numbers of game keepers, while the frisians build big numbers of hunters, earning much more fur than you expect?

No I didn't see this yet. However I deem this to be just a little advantage as the reindeer farm produces a fur every 65 seconds and the hunter currently does it in every 3rd cycle which results to 105 seconds plus walking times of the hunter.

But the reindeer farm needs also wells and barley farms.

So yes this could be used in a team match but only at the border and it isn't that effective in the current implementation, while it still does the trick to provide some fur in the early game to get the seamstress working.

If you think so...

Why shouldn't I?

Because the advantage for the Frisians could be big, if a bunch of reindeer farms, barley farms and wells (plus fishers, aqua farms, berry farms, berry collectors and even more wells) can be replaced by some hunters and gamekeepers.

Is there any other possibility? I am not that familiar with the options in online games.

I don't think so...

I think that the empire has more expensive buildings, because marble end especially columns are very expensive. Probably brick costs are not a big disadvantage...

brick is definitly a shortage over the whole game. I normally increase the target quanttity to have kind of a buffer. But the frisians really need a lot of bricks so I still believe the value of 3 is ok. All other tribes do not use that much refined log or other precious materials for building.

Barbarians use a very high number of refined logs for their military buildings... And is it the case that the empire needs significantly less marble than frisians need bricks? (Marble in marble columns do also count)

Have a look in the build cost excel. And in the decomposition into basic material I missed coal for the frisians, so it is worse than the current values. However you are right for the barbarian military buildings they use a lot of blackwood but the frisians needs lots of bricks for the normal economy as well.

the most compareable building material is the barbarian grout which is used in largely les numbers than bricks. Please look at the relevant excel to see it.

Yes, grout is very similar. By the way, I don't have to look at the excel to know this stuff face-wink.png

I didn't mean to have a look there to recognize that grout is the most comparable ware but to see the whole picture why bricks are neede so uch by the frisians. Sorry to bother you.

No need to be sorry, that was not bothering.

I Know now and did so to achieve my best times although the cycle had been changed already. But in the later game the frisians still need more than 2 times more iron than gold so the cycle would be no problem. At least I did not discover any negative effects in the later game.

If you think so...

Any other opinion on this supported by some explanations would be highly appreciated.

But barbarians are slower than empire and atlanteans (in training their first perfect soldier). What about the fact that frisians are much slower than atlanteans? Maybe you should find out how fast you are at "Two frontiers" face-wink.png

Ok I will try this. For academical purposes I did all of my testing without opponent. Shall I do it with AI opponent? What would be a good result ( or you may call it a benchmark) there?

Good question. I forgot that my tournament match against king_of_nowhere - where I needed 33 minutes - is not the best benchmark anymore, because it was not a complete soldier (he lacked shields, and after build19 shields are important again). So actually my suggestion was not perfect. By the way: Yes, it's more interesting to know how fast one can play with opponents on the map.


Wanted to save the world, then I got widetracked

Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2018-06-01, 21:33

WorldSavior wrote:

No, you are not wrong, I just mixed up your definitions.

ok

Not necessarily. Atlantean mines need three workers, and deepest imperial mines need only two workers - and are probably cheaper.

Ok according to that logic the order would be barbarians depest mines fastest as they need upgrades as well as three miners atlanteans probably second than empire and frisians probably equal with only small differences between the differnt tribes. Is this right?

I think it's interesting that there is also a gold mine which is more efficient than its iron "twin". If someone uses a barbarian deep iron mine, it's his own fault. Everyone can just inform himself about the efficiencies of mines and conclude that the deep iron mine has to been skipped... Cleverness will be rewarded here.

I do not agree to this. Reason is that following this logic any bug would be just a challenge for the clever people. I think we should make widelands at least that consistent that it doesn't need a deeper study of the code to avoid such situations. This isn't documented anywhere alse than in the lua. So I would definitly vote to have the same values there as for the deep gold mine.

Have you thought about what happens if barbarians and frisians play in the same team? The barbarians could build big numbers of game keepers, while the frisians build big numbers of hunters, earning much more fur than you expect?

No I didn't see this yet. However I deem this to be just a little advantage as the reindeer farm produces a fur every 65 seconds and the hunter currently does it in every 3rd cycle which results to 105 seconds plus walking times of the hunter.

But the reindeer farm needs also wells and barley farms.

So yes this could be used in a team match but only at the border and it isn't that effective in the current implementation, while it still does the trick to provide some fur in the early game to get the seamstress working.

If you think so...

Why shouldn't I?

Because the advantage for the Frisians could be big, if a bunch of reindeer farms, barley farms and wells (plus fishers, aqua farms, berry farms, berry collectors and even more wells) can be replaced by some hunters and gamekeepers.

Ok I agree that the combination of barbarians and frisians in a team match could have some advantages. What I meant was to get this to take effect they would need a common border, where on one side of the border the barbarians produce the game and on the other side the frisians collect the fur. However the hunter produces a fur every 105 seconds (sleep time) plus x seconds hunting/walking time, whereas a reindeer farm produces a fur every 65 seconds. So it means you need at leat two hunters to produce at the same rate than a reindeer farm. If the buildings are not properly placed at the border of the two team mates there is a chance that some game might esacape the frisian hunters. So I really don't think the advantage would be that big. An additional argument is that in the late game fur is also produced by recycling centers. So fur isn't a bottleneck anymore in the late game.
So please reconsider whether you see still an unreasonable advantage for a Barbarian/Frisian team by including this change.
I f you provide me good reason I would consider to undo this in the next step.

Is there any other possibility? I am not that familiar with the options in online games.

I don't think so...

Just to be clear I tried the share HQ option but this would be only possible with same tribe and therefore the possibility to share game is only possible at hte border.

I didn't mean to have a look there to recognize that grout is the most comparable ware but to see the whole picture why bricks are neede so uch by the frisians. Sorry to bother you.

No need to be sorry, that was not bothering.

So do you see my point about the preciousness for bricks for the frisians now.

But barbarians are slower than empire and atlanteans (in training their first perfect soldier). What about the fact that frisians are much slower than atlanteans? Maybe you should find out how fast you are at "Two frontiers" face-wink.png

Ok I will try this. For academical purposes I did all of my testing without opponent. Shall I do it with AI opponent? What would be a good result ( or you may call it a benchmark) there?

Good question. I forgot that my tournament match against king_of_nowhere - where I needed 33 minutes - is not the best benchmark anymore, because it was not a complete soldier (he lacked shields, and after build19 shields are important again). So actually my suggestion was not perfect. By the way: Yes, it's more interesting to know how fast one can play with opponents on the map.

Ok I did not have time to do my run on this map. To busy with other bugfixing activity. But sure I will do so in the near future.


Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2018-06-01, 21:44

@ WorldSavior
By the way congrats to your astonishing success in Einsteins tournament. Hopefully as this is over you may find more time for our discussions. face-wink.png


Top Quote
WorldSavior
Avatar
Joined: 2016-10-15, 04:10
Posts: 2091
OS: Linux
Version: Recent tournament version
Ranking
One Elder of Players
Location: Germany
Posted at: 2018-06-09, 13:58

hessenfarmer wrote:

WorldSavior wrote:

No, you are not wrong, I just mixed up your definitions.

ok

Not necessarily. Atlantean mines need three workers, and deepest imperial mines need only two workers - and are probably cheaper.

Ok according to that logic the order would be barbarians depest mines fastest as they need upgrades as well as three miners atlanteans probably second than empire and frisians probably equal with only small differences between the differnt tribes. Is this right?

I don't know. I think the mine speeds of the old tribes could stay like they are...

I think it's interesting that there is also a gold mine which is more efficient than its iron "twin". If someone uses a barbarian deep iron mine, it's his own fault. Everyone can just inform himself about the efficiencies of mines and conclude that the deep iron mine has to been skipped... Cleverness will be rewarded here.

I do not agree to this. Reason is that following this logic any bug would be just a challenge for the clever people. I think we should make widelands at least that consistent that it doesn't need a deeper study of the code to avoid such situations. This isn't documented anywhere alse than in the lua.

Actually it is: In the tribe encyclopedia.

So I would definitly vote to have the same values there as for the deep gold mine.

I didn't mean to have a look there to recognize that grout is the most comparable ware but to see the whole picture why bricks are neede so uch by the frisians. Sorry to bother you.

No need to be sorry, that was not bothering.

So do you see my point about the preciousness for bricks for the frisians now.

Not yet, because your file does only include what happens if you build one building of each type or something like this. But so far playtesting didn't tell me that bricks are too expensive.

hessenfarmer wrote:

@ WorldSavior
By the way congrats to your astonishing success in Einsteins tournament.

Thank you

Hopefully as this is over you may find more time for our discussions. face-wink.png

Actually it is not really the case. I didn't play all the time for the tournament, I didn't play during some weeks before its end, I had no time for that...


Wanted to save the world, then I got widetracked

Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2018-06-09, 17:21

WorldSavior wrote:

I don't know. I think the mine speeds of the old tribes could stay like they are...

Probably yes, cause the difference is not that big. But still they do not obey to one common logic. For this I found your explanation regarding the atlantean mine quite useful to apply such logic to all mines. I believe this si definitly nothing to worry about for b20. Probably after the release we will have more time to analyze this somewhat deeper.

I do not agree to this. Reason is that following this logic any bug would be just a challenge for the clever people. I think we should make widelands at least that consistent that it doesn't need a deeper study of the code to avoid such situations. This isn't documented anywhere alse than in the lua.

Actually it is: In the tribe encyclopedia.

Could you please provide me a link or a procedure how to find this information in the encyclopedia. Until now I didn't discover it yet. Maybe I am to blind to see it.

Not yet, because your file does only include what happens if you build one building of each type or something like this. But so far playtesting didn't tell me that bricks are too expensive.

Hm, while trying to create a fully trained soldiers in the beginning my biggest shortages were bricks and reet. After the change in the brick kiln, it was reed and clay (clay reaulting in a shortage of bricks). After reintroducing the claydigger it was not that easy to tell anymore, cause other shortages of material came to the front . I already increase the target for bricks to make the brick kiln start earlier and provide some more buffer but after all I still had brick shortages. I never reached a stock amount for bricks more than 15 in the first 2 hours of the game. Per haps you would consider doing a test of the frisians in their current state and provide a replay (as upload to the bug) so we can discuss the issues resulting from playtest. I will do so as well when I have time to do my run on the map you proposed. I'll have to fix my bugs first though.


Top Quote
WorldSavior
Avatar
Joined: 2016-10-15, 04:10
Posts: 2091
OS: Linux
Version: Recent tournament version
Ranking
One Elder of Players
Location: Germany
Posted at: 2018-06-24, 13:35

hessenfarmer wrote:

WorldSavior wrote:

hessenfarmer wrote:

I do not agree to this. Reason is that following this logic any bug would be just a challenge for the clever people. I think we should make widelands at least that consistent that it doesn't need a deeper study of the code to avoid such situations. This isn't documented anywhere alse than in the lua.

Actually it is: In the tribe encyclopedia.

Could you please provide me a link or a procedure how to find this information in the encyclopedia. Until now I didn't discover it yet. Maybe I am to blind to see it.

When you play a match, you can simply select "tribal encyclopedia" (by clicking at the symbol or pressing F1). Then you can select "wares" and select an ore, there you will see how many ores are produced per consumed wares.

Not yet, because your file does only include what happens if you build one building of each type or something like this. But so far playtesting didn't tell me that bricks are too expensive.

Hm, while trying to create a fully trained soldiers in the beginning my biggest shortages were bricks and reet. After the change in the brick kiln, it was reed and clay (clay reaulting in a shortage of bricks). After reintroducing the claydigger it was not that easy to tell anymore, cause other shortages of material came to the front . I already increase the target for bricks to make the brick kiln start earlier and provide some more buffer but after all I still had brick shortages. I never reached a stock amount for bricks more than 15 in the first 2 hours of the game. Per haps you would consider doing a test of the frisians in their current state and provide a replay (as upload to the bug) so we can discuss the issues resulting from playtest. I will do so as well when I have time to do my run on the map you proposed. I'll have to fix my bugs first though.

I think that I've got no time for that...


Wanted to save the world, then I got widetracked

Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2018-06-29, 22:00

WorldSavior wrote:

When you play a match, you can simply select "tribal encyclopedia" (by clicking at the symbol or pressing F1). Then you can select "wares" and select an ore, there you will see how many ores are produced per consumed wares.

Ok, got it now. However, although it is documented this information is not obvious to find. Furthermore this behaviour is not consistent and illogical, especially in comparison to the goldmine. However as we seem to be the only people discussing this topic we won't find an agreement probably. So I would prefer to open a bug report on this to get a more widespread attention and hopefully a solution where everybody could live with.
By the way how is your work with fixing the other issues in the mines going on? I couldn't remember seeing any branch dealing with this.

I think that I've got no time for that...

That is a pity. So as my changes regarding balancing are in trunk now, perhaps we get some other feedback regarding the balancing of the tribe. Although I really would have appreciated you doing a short 1 hour run and provide your opinion on the current state. Without any further feedback I am lost now in considering improvements. Anyway thanks for your contribution.


Top Quote
GunChleoc
Avatar
Topic Opener
Joined: 2013-10-07, 15:56
Posts: 3324
Ranking
One Elder of Players
Location: RenderedRect
Posted at: 2018-07-06, 13:32

The building documentation is less detailed on specific wares than the wares and worker documentation - I never expanded on that, because the pages are already quite long. If only we had tree views and/or tables of content capability...


Busy indexing nil values

Top Quote
WorldSavior
Avatar
Joined: 2016-10-15, 04:10
Posts: 2091
OS: Linux
Version: Recent tournament version
Ranking
One Elder of Players
Location: Germany
Posted at: 2018-07-08, 14:35

hessenfarmer wrote:

WorldSavior wrote:

When you play a match, you can simply select "tribal encyclopedia" (by clicking at the symbol or pressing F1). Then you can select "wares" and select an ore, there you will see how many ores are produced per consumed wares.

Ok, got it now. However, although it is documented this information is not obvious to find.

I think it is. There is a button next to other important buttons for the encyclopedia, and F1 is a common help button.

Furthermore this behaviour is not consistent and illogical, especially in comparison to the goldmine.

This doesn't have to mean a lack of logic, because mine values are just different.

However as we seem to be the only people discussing this topic we won't find an agreement probably. So I would prefer to open a bug report on this to get a more widespread attention and hopefully a solution where everybody could live with.

By the way how is your work with fixing the other issues in the mines going on? I couldn't remember seeing any branch dealing with this.

As I don't have much time now, I haven't started yet.

I think that I've got no time for that...

That is a pity. So as my changes regarding balancing are in trunk now, perhaps we get some other feedback regarding the balancing of the tribe. Although I really would have appreciated you doing a short 1 hour run and provide your opinion on the current state. Without any further feedback I am lost now in considering improvements. Anyway thanks for your contribution.

You're welcome.

Even without playtesting I can say that for example the output of 4 wares in deep frisian mines (for each meal) is far too low to be enough. And probably barley farms are still too bad.


Wanted to save the world, then I got widetracked

Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2018-07-08, 15:14

WorldSavior wrote:

Furthermore this behaviour is not consistent and illogical, especially in comparison to the goldmine.

This doesn't have to mean a lack of logic, because mine values are just different.

Normally gold is harder to find and mine in reality. Therefore it seems to be very illogic for me that a mine of the same upgrade level and same worker experience delivers more gold than iron. So from my point of view this is something illogic that doesen't need to exist / is easy to fix.

As I don't have much time now, I haven't started yet.

Shall I try to provide a branch including the changes from our dicussion in the other thread. I could try to improve frisian balancing and I could fix the barbarian deep iron mine to have the same values as deep gold mine.

I think that I've got no time for that...

That is a pity. So as my changes regarding balancing are in trunk now, perhaps we get some other feedback regarding the balancing of the tribe. Although I really would have appreciated you doing a short 1 hour run and provide your opinion on the current state. Without any further feedback I am lost now in considering improvements. Anyway thanks for your contribution.

You're welcome.

Even without playtesting I can say that for example the output of 4 wares in deep frisian mines (for each meal) is far too low to be enough. And probably barley farms are still too bad.

regarding the mining you might be right probably it should be 5 per meal to have an adequate cost per ware. On the other hand there are some values even higher than the frisian cost per ware. Probably atlanteans gold mines do not deliver much for the food and drinks.
Barley farms are sufficient in my experience when they are finally running at 100%. However it lasts somewhat long until this is the case. But this is a weakness of the tribe that we could call intentional. If at all we could lower the barley growth time by a minor value to speed this up a little.


Top Quote