does it feel like this game lacks a soul?

Users who are viewing this thread

The game's dialog mechanic uses a system "tags" and "weights" to determine which line an npc should speak, but currently when the game loads in the all of the dialog lines, it hardcodes the weights of all of the associated tags to a value of 1 instead of using the weights that are specified in the xml files.

oqUVt.png

That essentially blocks a certain portion of the dialog lines from ever being chosen, and it's also the reason that your spouse won't call you wife/husband, because the "PlayerIsSpouse" tag is supposed to have a weight of 10 for those greetings, but is instead assigned a weight 1 when they are loaded in. It also causes dialog lines with negative weights to be selected in the wrong situations.

V1E9-.png

It's anyone's guess whether hardcoding it like that is just temporary while they work on it, an oversight/bug, or they switched to a different system and haven't worked out all the kinks yet ?‍♂️.

Also, I think some of the conversation tags are either bugged or redundant, which has an effect too.

Edit: It's the LoadFromXML method in the GameTextManager class.
Your finding make me feel terrible frustrated, so the forum is complaining about this topic almost from day 1 and the stuff is there (better dialogs I mean) and no one took some time to check why is not working. No one on the team knew that your wife shouldn't call you 'my lord'?

Honestly the time that TW has invested in half backed stuff is incredible, if they were able to polish all these the game will be 5 times better but it seems that they don't know how to do it... again seems a poor leadership problem.
 
Your finding make me feel terrible frustrated, so the forum is complaining about this topic almost from day 1 and the stuff is there (better dialogs I mean) and no one took some time to check why is not working. No one on the team knew that your wife shouldn't call you 'my lord'?

Honestly the time that TW has invested in half backed stuff is incredible, if they were able to polish all these the game will be 5 times better but it seems that they don't know how to do it... again seems a poor leadership problem.
That actually makes me wonder how many people who made the bulk of to code work are still there...
 
Last edited:
I think it is intentional, the current prisoner recruitment system was in the code some time before it was ready. There is more stuff in the code that is not currently in use.
I've have to agree here, it seems to be intentional and Mexxico said once that these are small things that aren't a priority to them right now, it's more for the polishing phase later on.
 
I've have to agree here, it seems to be intentional and Mexxico said once that these are small things that aren't a priority to them right now, it's more for the polishing phase later on.
Intentional or not the feeling is more or less the same and it seems to be against modern software development paradigms, let me try to explain:

Now days what a software company hopes is deliver value to it clients as fast as possible, companies can not afford money investments for months (8 years?) without any return.

So the old water fall model has been progressively abandoned, it means... release a version, work in the new one for N time, then release the new version. From modern point of view if one team has finished X feature you want to deliver it asap to your customer to add value to your product rather than waiting Y more days to let all teams finish they planned work for version XXXX and then publish.

So new paradigms like agile management, continuous integration/deployment, micro service architecture, etc has been put in place to archive this objective in more or less degree in almost all modern companies.

TW seems completely anchor in the old water fall model, they have features half backed that can not be released due dependence of other stuff and until all task planned for version 1.5.X aren't finish nothing can be released, and if one of these dependence aren't planned until 10 versions ahead the content is not delivered.

it is a really sad situation that slow down the deliver process and generate a lot of frustration due the long time which seems to correct/enhance simple things like for example better dialogs in this case.

Sorry for the stupid post, but as I said I am very sad today.
 
Exactly, iterative development strives to have a working, integrated version ASAP. Not everyone waiting for someone else.
I've have to agree here, it seems to be intentional and Mexxico said once that these are small things that aren't a priority to them right now, it's more for the polishing phase later on.
That's not a very good argument. You don't leave every system unpolished and say "we'll do this later" and then it turns out you have a ton of work finishing the systems and you are stuck doing this for a long time, probably missing some of the needed changes because no one keeps a detailed list.
You complete a system as if it would be released as such. If for some reason you decide later it needs to be revamped, you do this later. Most of the time it turns out that the whole system will be released as such, even if you had better ideas, but ran out of time to implement them. But you didn't release half-assed crap.
 
That's not a very good argument. You don't leave every system unpolished and say "we'll do this later" and then it turns out you have a ton of work finishing the systems and you are stuck doing this for a long time, probably missing some of the needed changes because no one keeps a detailed list.
You complete a system as if it would be released as such. If for some reason you decide later it needs to be revamped, you do this later. Most of the time it turns out that the whole system will be released as such, even if you had better ideas, but ran out of time to implement them. But you didn't release half-assed crap.
Absolutely agreed, i was just saying what seems to be on their mind, this is only creating problems for later and probably it was this way of doing things that made them fall into dev hell and have to abandon lots of features and ambitions during development "let's fix this all later / oh **** this isn't working just axe it and rewrite a simpler code, we have lost too much time on this"
 
This is all well and good if you have a product owner who prioritizes what is the minimum viable product and what additional features would be good to have once you achieved the mvp, which is usually someone who is not part of the company that is developing the software, but the company that hired it to develop said software. None of us here hired them to develop Bannerlord, we are buying an end product.

TW has their own vision of what their mvp is and how much they will invest past that. The closest we can get to defining how features will look like when done is providing feedback and suggestions, and even then, they can simply ignore everything and keep their vision of what Bannerlord is supposed to look like.

Don't kid yourselves thinking Early Access is meant to include us in the decision making, it is just an embellishment of "we want to monetize while developing and maybe even listen to you if it fits what we want". It works best for small companies with no other titles, or for companies that are venturing into unexplored IPs and could use the help to shape a successful IP that will attract potential buyers. TW fits neither, we can complain all we want but Bannerlord is already a commercial success and by the time they release their next title our vision of what Bannerlord was will be warped by mods.
 
Last edited:
I wished that instead of dropping the "hot stuff" aka rebellions and pillaging they spent an entire month doing things that you know are "minimal" maybe take care of fifthy of those minimal stuff and you have yourself a good patch that makes the game fun and entretaining
 
I wished that instead of dropping the "hot stuff" aka rebellions and pillaging they spent an entire month doing things that you know are "minimal" maybe take care of fifthy of those minimal stuff and you have yourself a good patch that makes the game fun and entretaining
I actually agree - but back when a few patch notes were full of lots of small tweaks, added dialogues and new quests, this forum was full of people screaming for big new features.
 
Voice acting would go a long way in making the characters alive. And no, not generic garbage either - we have enough generic companions. Every family head should be voice acted with a unique voice, and unique personality. I barely recall any lords names in all my time playing this. They're all bland and generic NPCs with expensive armor that take longer to die than the other NPCs.
 
I've have to agree here, it seems to be intentional and Mexxico said once that these are small things that aren't a priority to them right now, it's more for the polishing phase later on.
On the other hand, they should probably put at least some focus on the things that players are getting frustrated with even if they are small things that don't have priority. You have to remember that the game is released to the public and that "EA" tag will only go so far. If people who are playing a frustrated by say the issue where troops don't use siege towers effectively or sometimes at all, they may very well stop playing and may very well tell other interested parties that the game is such a buggy mess and the devs so inept that after an entire year, they can't even get troops to climb the damn siege towers and to not bother with the game. This sort of stuff is one of the huge issues a developer has to deal with when they go the "EA" route to get early money for selling an unfinished product.

My point is BECAUSE these are small things, they tend to have big impact on player perceptions and because so many of the small things haven't been dealt with, many in the playerbase, including myself, don't really feel the development of the game has made any real progress in the last year.
 
The game's dialog mechanic uses a system "tags" and "weights" to determine which line an npc should speak, but currently when the game loads in the all of the dialog lines, it hardcodes the weights of all of the associated tags to a value of 1 instead of using the weights that are specified in the xml files.

oqUVt.png

That essentially blocks a certain portion of the dialog lines from ever being chosen, and it's also the reason that your spouse won't call you wife/husband, because the "PlayerIsSpouse" tag is supposed to have a weight of 10 for those greetings, but is instead assigned a weight 1 when they are loaded in. It also causes dialog lines with negative weights to be selected in the wrong situations.

V1E9-.png

It's anyone's guess whether hardcoding it like that is just temporary while they work on it, an oversight/bug, or they switched to a different system and haven't worked out all the kinks yet ?‍♂️.

Also, I think some of the conversation tags are either bugged or redundant, which has an effect too.

Edit: It's the LoadFromXML method in the GameTextManager class.
Coincidence? I think not!
Fixed an issue that caused conversation tag weights (that affect dialogue variation) to be set improperly.
 
There's a purpose for hand-crafted characters in games, and a purpose for generic characters, and those purposes are generally not the same. In original M&B we had hand-crafted characters who fit the game, and it worked surprisingly well, at least until you got tired of recruiting the same list of characters in campaign after campaign. Warband merely took the same characters and reused them, without even making the effort to adjust their dialog to the new map, and again it worked reasonably well until they once again got "stale" after a while.

The solution as I see it is to have a MIX of generic and hand-crafted characters, and only a small randomized sub-set of the hand-crafted characters would appear in any given game. If you only see a character every 4th or 5th game, they don't "wear out", yet you still have some occasional unique dialog that fits the setting, rather than only generic lines from generic NPCs with generic faces. If the initial stats for a few of the generic characters are chosen from a list to provide at least one character for each "skill set" (tank, archer, scout, healer, merchant, engineer), possibly with slightly less-generic skill-related dialog, then you at least have a possible chance of filling out a roster with specialists, or else pick those with generic skill sets and train them up from scratch.

Most importantly, you don't get the same combinations in every game, but a few of them may still become "familiar faces" you hope to find, after a few games.

As for dialog, if you have at least 2 versions for the same basic topic, that allows one generic character to use one set of dialog, while another uses the alternate set, so you have at least the illusion of a personality difference. When 3 or more different characters say exactly the same thing in the same words, it's painfully obvious that they're mindless NPCs, and it mentally kicks me out of the game world for a moment. When NPCs in Oblivion said the same words in the exact same voice, sometimes regardless of race, it shattered the illusion completely....and then they switched to a SECOND voice for the next line, which was humorously awful as an obvious "bug" the first time, but painfully annoying when it kept happening with different NPCs. Bannerlord has to do better than that.
 
Check hotfix notes :razz: The bug was slightly different from what Bannerman showed (the decompiled code isn't quite right). We had a chat in PM (sharing stories of marriage) and basically the cause of the issue is that the assignment of the weights was nested within a debug code, which meant that everything looks great to the people running the game in debug, but not so much for those of you playing it regularly.

I won't go into details, but we haven't said that we would drop voice overs.
 
Back
Top Bottom