Matrix Games Forums

Across the board, Distant Worlds 2 has pretty lacklustre customisation for empire automation. Diplomacy and Espionage is the focus of this thread.

I find managing diplomacy quite tedious, and would prefer to automate almost everything. Pirates and independents seem to be placeholders, and empires are rarely worth engaging with.

Ethically, I seldom want to invade independent colonies, and I would appreciate it if the automation did not set the policy to invade when able for every neutral independent, then suggest my invasion fleets attack.

The automation does not "understand" ambassadors are useful to support espionage missions. I have a great target empire for espionage missions, manually set the policy to undermine, with an ambassador assigned. The automation would move the diplomat to another empire. I have 3 other diplomats.

Espionage:
We need a customisable empire and mission type matrix where we can allow and disallow mission types, particularly extremely low value ones. I do not need territory maps for my nearest neighbour every few missions. Spies are too easily lost, too slow to replace, too expensive to recover and cost too much in lost opportunity for them to be used on something that can be gained trivially with a few exploration ships costing 2,000 credits each.

Maps and Steal Technology are really all that spies are currently good for. The choice of which technologies the automation prioritises stealing needs to be implemented, and the interface for finding technologies could be improved, particularly if targeting those with many techs..

I agree that the espionage part of the game need an overhaul.

StormingKiwi wrote: Tue Jan 31, 2023 12:17 pm Across the board, Distant Worlds 2 has pretty lacklustre customisation for empire automation.

I would argue it has more customization of automation than most, but I guess there's not much point in debating that.

I find managing diplomacy quite tedious, and would prefer to automate almost everything. Pirates and independents seem to be placeholders, and empires are rarely worth engaging with.

I find the latter part of this quite arguable. Pirates and Independents are not placeholders. Each has a significant role to play in the early game and to some degree going on into the mid-game. Independents are for either trade or peaceful or military integration. Pirates can be either enemies or allies, eventually integrating fully with your empire if you help them with the Hive.

Regarding Empires, what would make them worth engaging with for you? I find you can do a whole lot by engaging with other empires.

Ethically, I seldom want to invade independent colonies, and I would appreciate it if the automation did not set the policy to invade when able for every neutral independent, then suggest my invasion fleets attack.

You can change this of course and unless you've really cranked up the number of independents, there shouldn't be very many to deal with.

With that said, the policy will tend to default to invade if you are either playing an aggressive faction by default or have researched the means to invade, but don't yet have the means to colonize, or if the independent is so negatively disposed towards you that the automation decides it's not feasible to try to bring it on your side diplomatically.

The automation does not "understand" ambassadors are useful to support espionage missions. I have a great target empire for espionage missions, manually set the policy to undermine, with an ambassador assigned. The automation would move the diplomat to another empire. I have 3 other diplomats.

Would very much appreciate a save showing this situation. I completely agree the automation should understand how to do this right.

Espionage:
We need a customisable empire and mission type matrix where we can allow and disallow mission types, particularly extremely low value ones. I do not need territory maps for my nearest neighbour every few missions. Spies are too easily lost, too slow to replace, too expensive to recover and cost too much in lost opportunity for them to be used on something that can be gained trivially with a few exploration ships costing 2,000 credits each.
Maps and Steal Technology are really all that spies are currently good for. The choice of which technologies the automation prioritises stealing needs to be implemented, and the interface for finding technologies could be improved, particularly if targeting those with many techs..

I disagree that spies are only worth using to steal maps and technology, though those are certainly very good uses and great ways to train up your spies overall. Some factions are better at espionage than others, some are easier targets than others. Deciding to mount a campaign of espionage against another empire will entail risk and use up spies. It's not intended that espionage be a low risk activity or an easy button.

The automation certainly does choose what technologies to steal if left to its own devices, but I'm sure it doesn't always choose wisely.

Adding additional player guidance/policies on how to direct automated espionage is a good feature improvement idea.

Regards,

- Erik

Erik Rutins wrote: Thu Feb 02, 2023 8:33 pm

StormingKiwi wrote: Tue Jan 31, 2023 12:17 pm Across the board, Distant Worlds 2 has pretty lacklustre customisation for empire automation.

I would argue it has more customization of automation than most, but I guess there's not much point in debating that.

I guess this is a misleading/poor choice of words because I am referring to the "advisors" that steer, rather than the automation that propels.

Diplomacy

I find managing diplomacy quite tedious, and would prefer to automate almost everything. Pirates and independents seem to be placeholders, and empires are rarely worth engaging with.

I find the latter part of this quite arguable. Pirates and Independents are not placeholders. Each has a significant role to play in the early game and to some degree going on into the mid-game. Independents are for either trade or peaceful or military integration. Pirates can be either enemies or allies, eventually integrating fully with your empire if you help them with the Hive.

Regarding Empires, what would make them worth engaging with for you? I find you can do a whole lot by engaging with other empires.

Independents are more expensive, but already established, colonies that you have the opportunity to decide how to integrate - that choice is great. However, trade volumes are frequently too low to notice, appearing to generate much less traffic than in DW:U. Diplomatically, they would be more interesting if they did something other than wait around for an empire to integrate them or ask the player to colonise them when that value is above ~ +20. They are not supposed to be actors, but at the moment, they are completely passive.

Pirates are more interesting by comparison, however, I believe they were more developed in DW:U. The only one worth allying with or making a conscious effort to fight in the current save is the one that every pirate ship event gifted ships to. The others all have known military strength zero - the one nearest to me probably suffered too many casualties fighting other pirates/empires, and so legitimately has a very small number of ships present in the galaxy. Creatures harass my civilian shipping more than the pirates do, and that's because the freighters jump to the same star as the creatures en route to somewhere else.

Empires: well it's two parts:
1) "consider during gameplay". In the galaxy, there is one AI empire (Kygnos Consciousness) in the galaxy I actively monitor.
2) Engaging - by this I mean actively participating in diplomacy to achieve goals to frustrate the ambitions of my rivals. Unfortunately, it is a similar problem experienced across the strategy genre - my empire could engage in trade agreements and military agreements with 'allies', but cooperation does not seem to be as rewarding as conquering.

Ethically, I seldom want to invade independent colonies, and I would appreciate it if the automation did not set the policy to invade when able for every neutral independent, then suggest my invasion fleets attack.

You can change this of course and unless you've really cranked up the number of independents, there shouldn't be very many to deal with.

With that said, the policy will tend to default to invade if you are either playing an aggressive faction by default or have researched the means to invade, but don't yet have the means to colonize, or if the independent is so negatively disposed towards you that the automation decides it's not feasible to try to bring it on your side diplomatically.

The policy can not be changed on a global level. Yes, I can change the diplomatic relationship per independent to Manual - but I'd prefer just to set a policy in military or diplomacy saying "Do not Invade Independents", and then set my military attacks to automated.

The issue is that invading an independent decreases your reputation. Reputation affects colony approval across the empire and other empires' perceptions of you.
Internally, approval affects taxation, therefore revenue, therefore growth spending, research spending and available surplus.
Externally, reputation affects relationships with other empires.

The AI does not understand the cascade of effects that invading an independent can have on the empire's economy.

There are times when it costs more to conquer an independent than the long-term gain of one more colony.

Espionage

The automation does not "understand" ambassadors are useful to support espionage missions. I have a great target empire for espionage missions, manually set the policy to undermine, with an ambassador assigned. The automation would move the diplomat to another empire. I have 3 other diplomats.

Would very much appreciate a save showing this situation. I completely agree the automation should understand how to do this right.

I'm not sure that ambassadors should be used to support espionage.

There should be a spymaster character type you either keep at your capital or send to the capitals of empires and pirates. That character type should be rare. Spies who do missions should be common.

Espionage:
We need a customisable empire and mission type matrix where we can allow and disallow mission types, particularly extremely low value ones. I do not need territory maps for my nearest neighbour every few missions. Spies are too easily lost, too slow to replace, too expensive to recover and cost too much in lost opportunity for them to be used on something that can be gained trivially with a few exploration ships costing 2,000 credits each.
Maps and Steal Technology are really all that spies are currently good for. The choice of which technologies the automation prioritises stealing needs to be implemented, and the interface for finding technologies could be improved, particularly if targeting those with many techs..

The automation certainly does choose what technologies to steal if left to its own devices, but I'm sure it doesn't always choose wisely.

Adding additional player guidance/policies on how to direct automated espionage is a good feature improvement idea.

Regards,

- Erik

I am fully aware that automation does steal techs. The issue is wisdom.

For example, I frequently do not want it to steal techs that fulfil any or some combination of the following criteria:

• I have invested cash and resources in initialising.
• I will not use.
• I consider low priority to research.

I disagree that spies are only worth using to steal maps and technology, though those are certainly very good uses and great ways to train up your spies overall. Some factions are better at espionage than others, some are easier targets than others. Deciding to mount a campaign of espionage against another empire will entail risk and use up spies. It's not intended that espionage be a low-risk activity or an easy button.

The issue is risk vs reward. Espionage is a high-risk activity with extremely low rewards.

To add to my previous list, Spies earn experience far too slowly, are too easily lost, too slow to replace, too expensive to recover and cost too much in lost opportunities.

Anyway, while the conversation about the hypothetical and theoretical design is pleasant, the game does not work as intended.

Here's the evidence to back up that claim.

Automation uses mission difficulty caution to consider what missions to do - the lowest possible chance of success the automation will consider is 70%. Keep that in mind while reading.

Method:
I have used the empire victory progress ranking provided by the game to order the list of empires.
Then, chose a few empires from that list to look at mission success chances for my best spy to perform each type of mission.
Where multiple targets for the mission exist, I have standardised the targets across the empires for the validity of comparison.
I have estimated the credit benefit of an eventual successful mission for some of the mission types.
(The benefit of research is an estimate of the cost to initiate plus the cost to immediately crash, the benefit for the others is an estimate of how much it would impact the enemy)
I have estimated the credit cost of the successful mission as (35,000 credits / Chance of Mission Success), and estimated the economy-wide disruptions of Incite Revolution and Assassinate Leader as 30% and 10% of private revenue, respectively.

Credit benefits:
I did not determine a method of determining the credit benefit of Steal Operations Map, Sabotage Construction and Deep Cover, and could only trade Territory and Galaxy maps via diplomacy with two of the empires selected.

Results:
Enter Zenox Superspy, Arramy Kodansky. A citizen of the Memban Ascendancy from 2760.06 until 2853.05, before repatriating to a better Empire. This is the most phenomenally good spy I have seen in Distant Worlds 2 - and so this is a best-case scenario at the extreme end of the spectrum. The average spy is much worse than Arramy Kodansky.
I'll let those speak for themselves.

In practice, the majority of spies across every empire are performing counter-espionage. Most missions are well below the 70% success chance threshold. For these target empires, automation only considers a limited selection of mission types.
The only mission type worth the risk of a lost spy is stealing research information.

Finally, the Teekan Corporation would be my least desired target to send a spy on espionage missions to sabotage construction. I'd like to use Espionage to undermine enemies stronger than me, not punch down.

In general I don't think that DW 2 have bad diplomacy from a genre standard but I certainly hope it will get some development in this area in the future.

I do think that we need more than bi-lateral diplomacy and more diplomacy within influence spheres and trilateral agreements between empires. This would make diplomacy so much more interesting, dynamic and fun. It should be quite rare that diplomacy between intergalactic empires and factions would not influence each other.

In general I also think that most games make espionage into some too detailed system rather than more of a dynamic system. DW2 certainly is way more detailed rather than dynamic in my opinion... I basically only find steal research to be useful and worth my time and that also in many ways is too powerful. I would rather want espionage to be more event driven and you would give your spy organisation guidance on what to focus on. Focus all your effort on stealing research would not be 100% more efficient then only focusing half on it against a specific target. As in all things, dumping twice the resources into something rarely give you twice the outcome.

This is also why I don't like character driven systems in games of this nature, because no individual is that important on a civilization basis. Instead you should build up organisations in empires or even individual places... their bonuses and stats can then changed based on how you focus their support. This would be a much more interesting mini game without having to focus much of the player energy on it, also more easy to automate the AI.

This would go for both diplomacy and espionage organisations. How much you like to spend on this should be more of a financial thing than anything else, something you need to nurture over time. It should always be quicker to lose organisation influence than gaining it if you remove resources to support it.

I know that many people like characters, but I find it just gamey and uninteresting. It would be much better if you would set both diplomacy and espionage in the diplomacy screen at each empire as a measure of influence you want to have based on the money you dump into your overall diplomacy and espionage corps. Then you could also set focus for each toward each faction what they should focus on or if you want a balanced approach etc...

While Storming isn't taking into account a few important factors (and I would question the validity of the credit estimation), he is right that spies are effectively practical only for stealing tech - and map/territory data. I believe this- espionage- was said to be slated for another pass at balancing. I had believed this was the case because of that above all else.

Yep. #11 on the roadmap:
"Espionage needs a balance pass. It’s no longer too easy as it was at initial release, but the range of success or failure chances is currently too wide in some cases. The spy academy facility also needs to do a bit more."

Though we already understand it, I think its worth repeating... As the system stands, the races with bonuses have a tendency to end up with less spies because they will send them on missions, which inevitably ends up with them getting captured eventually. While having the espionage bonus grants those races a higher likelihood of getting more spies, to doesn't keep up with the losses. Then... In my last play, I shamelessly took advantage of a 1 colony rebel empire to pump my spies with free experience, as the Teekans. I didn't even need ambassadors to have a 100% chance of stealing tech or territory data with 1 month missions. And then promptly lost them in 80%+ missions and was left with but a single, albeit pretty good, spy for a long time.

I think Jorgen is onto something. Perhaps divorcing the espionage system from being dependent on disposable spy characters would be a great thing. Treat spies in the way that diplomats (or were they called ambassadors?) currently pseudo-function as spymasters- perhaps keep a chance that they get caught, but make it much less likely.

That otherwise mentioned change to characters getting fewer negatives would be a great pair with that.

Nightskies wrote: Tue Feb 07, 2023 8:43 am I believe this- espionage- was said to be slated for another pass at balancing. I had believed this was the case because of that above all else.

Yep. #11 on the roadmap:
"Espionage needs a balance pass. It’s no longer too easy as it was at initial release, but the range of success or failure chances is currently too wide in some cases. The spy academy facility also needs to do a bit more."

Feedback sent before balancing passes are carried out is useful, no?

Nightskies wrote: Tue Feb 07, 2023 8:43 am While Storming isn't taking into account a few important factors (and I would question the validity of the credit estimation), he is right that spies are effectively practical only for stealing tech - and map/territory data.

Though we already understand it, I think its worth repeating... As the system stands, the races with bonuses have a tendency to end up with less spies because they will send them on missions, which inevitably ends up with them getting captured eventually. While having the espionage bonus grants those races a higher likelihood of getting more spies, to doesn't keep up with the losses.

All models are wrong, some models are useful.

The model of estimating how much it costs the empire to lose the spy versus the benefit the spy gains for the empire is a vast improvement on the previous paradigm: subjective and evidenceless theory-craft based on wishful thinking.

Nightskies wrote: Tue Feb 07, 2023 8:43 am Then... In my last play, I shamelessly took advantage of a 1 colony rebel empire to pump my spies with free experience, as the Teekans. I didn't even need ambassadors to have a 100% chance of stealing tech or territory data with 1 month missions. And then promptly lost them in 80%+ missions and was left with but a single, albeit pretty good, spy for a long time.

If you send a spy on five 85% success chance missions, the probability, p, the spy will fail at least one mission is 55%. Not sure of the other probability, k, that a single failure will be a "career-ending failure" for the spy (double-agent trait, captured, killed), but k seems to be pretty high.

Nightskies wrote: Tue Feb 07, 2023 8:43 am I think Jorgen is onto something. Perhaps divorcing the espionage system from being dependent on disposable spy characters would be a great thing. Treat spies in the way that diplomats (or were they called ambassadors?) currently pseudo-function as spymasters- perhaps keep a chance that they get caught, but make it much less likely.

That otherwise mentioned change to characters getting fewer negatives would be a great pair with that.

Well, this is rude.

Give credit where credit is due, please.

That idea was generated in the post that Jorgen replied to, which was submitted 3 days ago to this thread

StormingKiwi wrote: Fri Feb 03, 2023 12:20 pm I'm not sure that ambassadors should be used to support espionage.

There should be a spymaster character type you either keep at your capital or send to the capitals of empires and pirates. That character type should be rare. Spies who do missions should be common.

"Feedback sent before balancing passes are carried out is useful, no?"

Absolutely! I also think this is the most constructive discussion you and I are both in a while... (Good thing)

"Give credit where credit is due, please."

Well... Jorgen was the one who mentioned the idea of abandoning characters for espionage, which is the really good idea IMO. Which could have been built on your idea, though it seems an original thought. Regardless, we are of like mind that ambassadors serving as pseudo-spymasters does feel a bit off.

I admire your effort in making the visuals to support your position. Even if it frustrates me sometimes ::haha::

Oh, regarding the 80% missions: failure doesn't equal capture. The significant point is that even attaining useful spies still ends up with having very few at all. A very frustrating thing given how much effort it may take to even get somewhat useful spies.

Nightskies wrote: Tue Feb 07, 2023 12:43 pm "Give credit where credit is due, please."

Well... Jorgen was the one who mentioned the idea of abandoning characters for espionage, which is the really good idea IMO. Which could have been built on your idea, though it seems an original thought. Regardless, we are of like mind that ambassadors serving as pseudo-spymasters does feel a bit off.

Both spymasters as characters and the concept of a spy organisation (an organized group of people with a particular purpose) were present in my post.

Elimination of spies completely, with players instead managing an anonymous espionage system, is a solution in search of a problem.

Yes, we are of like mind about ambassadorial involvement in espionage.

While I agree that character-driven systems are overdone in the genre and can be incongruous with their impact on the civilization scale, if they are executed well, they do have distinct advantages over other systems. Espionage and diplomacy are areas where characters have the potential to be used well, and currently are in Distant Worlds 2. The problems with espionage are related to balance.
(Civilizational Leaders, Scientists, Admirals and Generals on the other hand...)

There is a concept of a concrete unit, e.g., a ship, or a colony. They have properties (cost, location, strength, etc.), and exist in the map layer independently.
There is also a concept of an abstract unit, e.g., a facility or a character. These also have costs, locations, strengths etc, but they don't exist in the map layer independently - instead, they exist within concrete units.
Units affect player capabilities. You can't do more stuff than you have units.

Characters are the better choice for spies.

I believe espionage is most appropriately implemented using abstract units, not concrete ones, as most of the missions are nebulous and strategically flexible. It makes sense to send a spy agent on a mission to find the territory map of some empire you've only heard of. They go to space bars, they search the star lanes, and in a year they return with something. The next mission, they go to perform a different mission against a different empire. It doesn't make sense to do this with either a ship (that invariably has to disappear from your view on the map layer) or a facility.

For both Espionage and Diplomacy, Facilities don't make as much sense as Characters. While in the modern sense, an embassy is only a building, if the people aren't there, there is no embassy in the historical sense.
Facilities don't move. The idea of shifting priorities by shifting funding to build a network somewhere else over time seems interesting but always ends up with annoying gameplay (and a sunk cost of time that motivates the player not to give up all their progress against one empire to start again at square one).
Finally, because the characters are divorced from the economy, similar empires have similar capabilities. Espionage in Distant Worlds 2 is an area where automation does perform with reasonable levels of competence, even if its execution does not match my strategic preferences.

I'm sure others see it differently.

Nightskies wrote: Tue Feb 07, 2023 12:43 pm I admire your effort in making the visuals to support your position. Even if it frustrates me sometimes ::haha::

The visuals take very little effort, the effortful part is gathering the evidence and collating the data to support the visual.

However, both are much less effort than the difficulty of engaging in conversations and discussions with fuzzy, abstract, concepts where people's mental models don't align.

Nightskies wrote: Tue Feb 07, 2023 1:06 pm Oh, regarding the 80% missions: failure doesn't equal capture. The significant point is that even attaining useful spies still ends up with having very few at all. A very frustrating thing given how much effort it may take to even get somewhat useful spies.

Failure does not necessarily mean it is career-ending. However, I suspect the chance of a career-ending failure is a much higher value than I think it reasonably should be.

The career-ending failure is not a good mechanic either. You send a spy on a mission with a 95% chance of success, and 3 months in the other empire starts counterespionage, and the success chance immediately adjusts. There's no skill involved in being smitten by a Random Number God.
Yes, spies earn experience far too slowly.

I did want to estimate k but ran out of energy and time. If someone could help me by gathering data, or looking at files, that would be appreciated. For example, sending 10 spies on 1 low-success chance mission each, recording how many failed and how many of those who failed suffered a "career-ending failure”, then posting results would be highly appreciated. Complicating matters, I think there is a difference in chance of a "career-ending failure" between highly risky (red) and somewhat risky (yellow) missions.

Further complicated by how most agents can be recovered with a large cost via diplomatic exchange. Or militarily, through invasion of the word they end up on, though this act can easily kill them in the process. If one were to do the above data gathering, I think they should also include the results of getting the spies back. In my experience, more often than not, a recovered character is immediately dismissed on grounds of incompetence. Even in times were I did this because the spy was valuable, rather than because I had so much money that I simply bought everything they could offer aside from absurdly priced location information.

I believe it is possible to check what the experience did to the character before getting them back if you have sufficient detail on the world they're on. Given the questionable usefulness of a spy and how often they are simply truly worthless after being captured, this will often be a wasted effort.

I am not sure such information contributes to the point that really needs proved: spies are only practical for tech and map stealing. The cost chart was a good attempt, but there's too many abstract numbers being drawn in that, as if the presented data was a purposefully skewed to represent your truth rather than fairly representing the information that could be presented. I believe we need to fairly present more simple derived data.

That's a lot of work for us unpaid testers though.

Personally, I think spies as spymasters abstractly representing a generation of spies available with different strengths suffices better than having both spy and spymaster characters, much like how a single scientist and the leader represents far more than themselves (even if that is not the intent).

Nightskies wrote: Wed Feb 08, 2023 1:48 pm Further complicated by how most agents can be recovered with a large cost via diplomatic exchange.

If one were to do the above data gathering, I think they should also include the results of getting the spies back.

I think the cost of recovery via a diplomatic exchange is lower than a military campaign.

That's exactly what the cost value of ¤35,000 for one lost spy represents.

A career-ending failure is death OR capture OR gain of a double-agent trait.
Recovering a double-agent trait spy doubles the cost of the failure.

Nightskies wrote: Wed Feb 08, 2023 1:48 pm I believe it is possible to check what the experience did to the character before getting them back if you have sufficient detail on the world they're on. Given the questionable usefulness of a spy and how often they are simply truly worthless after being captured, this will often be a wasted effort.

You can deduce whether they have the double-agent trait by how much they are valued in diplomatic exchange - ¤30,000 or less is a double-agent spy.

Nightskies wrote: Wed Feb 08, 2023 1:48 pm 1. I am not sure such information contributes to the point that really needs proved: spies are only practical for tech and map stealing.
2. The cost chart was a good attempt, but there's too many abstract numbers being drawn in that, as if the presented data was a purposefully skewed to represent your truth rather than fairly representing the information that could be presented. I believe we need to fairly present more simple derived data.

1. I proved spies were only practical for tech and map stealing by showing the low success chances. This guarantees that automation would only consider those missions to be viable, in the majority of cases.
The net value calculation shows that alternative mission types are not rewarding.

2. The cost chart is the net value of a spy's successful mission. Net Value = Benefits - Costs. This is simple and is not a skewed representation of reality.

The benefits are all generated by finding the monetary value of a successful mission. These are concrete values.
The costs are all generated by finding the cost of a lost spy - again, a concrete value - and modifying that by the number of attempts needed to achieve one success - yet another concrete value.
All three values are derived from in-game data.

Furthermore, while the previous method overestimated the cost of lost spies and therefore, underestimated the net value, it did not change the conclusion to repeat the calculations with a correction incorporating k-values for the three thresholds (low risk, somewhat risky, highly risky). It's the sign of the net value that matters, not its magnitude.


Re: insubstantial critiques: These happen often, and every time, I am reminded of "The Man in the Arena" speech by Roosevelt:

Theodore Roosevelt wrote: It is not the critic who counts; not the man who points out how the strong man stumbles, or where the doer of deeds could have done them better. The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood; who strives valiantly; who errs, who comes short again and again, because there is no effort without error and shortcoming; but who does actually strive to do the deeds; who knows great enthusiasms, the great devotions; who spends himself in a worthy cause; who at the best knows in the end the triumph of high achievement, and who at the worst, if he fails, at least fails while daring greatly, so that his place shall never be with those cold and timid souls who neither know victory nor defeat.

Nightskies wrote: Wed Feb 08, 2023 1:48 pm That's a lot of work for us unpaid testers though.

Agreed - so why should anyone do it alone?

Nightskies wrote: Wed Feb 08, 2023 1:48 pm Personally, I think spies as spymasters abstractly representing a generation of spies available with different strengths suffices better than having both spy and spymaster characters, much like how a single scientist and the leader represents far more than themselves (even if that is not the intent).

I think scientific leaders suck because only their existence matters, and there are other ways to get the same bonuses, so I often forget them completely. Occasionally I notice when one dies, but I could not remember who or what benefit they brought.

Much like in the real world, the civilizational leader has very little impact on society as a whole.

lol. You really do see yourself as a paragon? You think you're doing the good work while the likes of I point out that much of what you claim is regularly omitting other's Experience and preferences as invalid before your own? You think your work so elevated that criticism of it is mere unsubstantial commentary! Hah! I needn't present anything to inform you that these numbers you selected are arbitrary. Do a search for Planetary Weapons (surface-to-space weapons analysis). You'll see an example I made that shows a fair presentation of simple derived numbers. It attempts to show as many relevant factors, calculated for (almost) every significant piece involved. Its excessive, perhaps, but it didn't take arbitrary values and made up expressions (even if simple) to support my argument. There's some substance for you. Now. This unnecessary bickering about... Ego... aside... we should be working together in this.

The point of my criticism is not to discredit you. Its that we should do better. For now, I don't know what values would be less arbitrary, but I'll try to think of something this morning.

I believe we can present a better argument to Erik that spies need a hard rework largely because they're practically only useful for tech and map stealing. I think he believes that they're still useful for the other things they're practically not useful for.

What about taking the Zenox, and figuring the chances of pulling off the various more difficult missions with a pool of... say... 10 spies, including missions meant just for experience to make them able to do so? Not just one-off attempts. I mean the chances of making an effective sabotage force under the best circumstances.

edit: Thinking about this more, it appears that you're expecting that someone just doesn't understand why you said the things you said when they don't think what you say is right/justified. This is not the case. Reiterating myself, I believe it was a good attempt, you have the right idea in showing value of spy actions to demonstrate how ineffective it is. Cost, however, has little to do with the viability of spies. And everyone knows spies have a low chance of success at the other missions...

edit2: I think the idea of tracking 10 Zenox spies and using them with maximum care under Spymaster style automation settings and presenting that as raw data will be quite useful. Then calculating how often one devoted to this style can be expected to do *something* with a major impact (aside from tech/maps) with *any* of the 10 spies against a rival empire or stronger. If the probability is less than 50%, I think that should constitute proof that espionage needs rework?

Nightskies wrote: Thu Feb 09, 2023 2:22 pm lol. You really do see yourself as a paragon? You think you're doing the good work while the likes of I point out that much of what you claim is regularly omitting other's Experience and preferences as invalid before your own? You think your work so elevated that criticism of it is mere unsubstantial commentary! Hah! I needn't present anything to inform you that these numbers you selected are arbitrary. Do a search for Planetary Weapons (surface-to-space weapons analysis). You'll see an example I made that shows a fair presentation of simple derived numbers. It attempts to show as many relevant factors, calculated for (almost) every significant piece involved. Its excessive, perhaps, but it didn't take arbitrary values and made up expressions (even if simple) to support my argument. There's some substance for you. Now. This unnecessary bickering about... Ego... aside... we should be working together in this.

Um, no I don't see myself as a paragon. You clearly do.
Insubstantial critique: When I said "These happen often" - I am not referring to you as a general case. Often, in the public arena, someone will bother to do some work and the "critical" response to it will be sneering without substance.

"Made-up expressions" - What are those? A career-ending failure? That's merely the chance of the spy failing in such a way that it means you cannot use them anymore.
"We should work together" - sure. I'm waiting for you to do some work.

Nightskies wrote: Thu Feb 09, 2023 2:22 pm The point of my criticism is not to discredit you. Its that we should do better. For now, I don't know what values would be less arbitrary, but I'll try to think of something this morning.

You don't know what values would be less arbitrary? Then mine are, by definition, the most systematic. I could have randomly generated the targets chosen - and that would be more systematic than no method. It's this attitude of sneering ignorance without being able to contribute something better that makes the post an insubstantial critique.

As it is, I systematically selected targets for everything. I had to start again with destroy base because some empires did not have a spaceport, and the display of highest incite rebellion at colony success chance for the entire dropdown only updates when you change your selection, so while in the first attempt, I selected that with the highest success chance I could see on the initial screen (which was the success chance for the first colony listed alphabetically), in the final version I selected capitals.

Yes, we are both aware it could be improved to perfection - for example, currently, the technology target was selected by the highest success chance. I could improve that by making a list of every technologies value and then selecting a target by the highest product of value and success chance.

But that won't change the conclusion.

Perfection is the enemy of the good.

Nightskies wrote: Thu Feb 09, 2023 2:22 pm I believe we can present a better argument to Erik that spies need a hard rework largely because they're practically only useful for tech and map stealing. I think he believes that they're still useful for the other things they're practically not useful for.

I don't think we can present a better argument than what has already been presented. Success chances below the minimum mission assignment limit prove that.

Nightskies wrote: Thu Feb 09, 2023 2:22 pm What about taking the Zenox, and figuring the chances of pulling off the various more difficult missions with a pool of... say... 10 spies, including missions meant just for experience to make them able to do so? Not just one-off attempts. I mean the chances of making an effective sabotage force under the best circumstances.

Well, the success chance of each mission is provided on the UI, so I'm not sure what you are trying to achieve by doing a lot of work to replicate what is already given to you.

Nightskies wrote: Thu Feb 09, 2023 2:22 pm edit: Thinking about this more, it appears that you're expecting that someone just doesn't understand why you said the things you said when they don't think what you say is right/justified. This is not the case. Reiterating myself, I believe it was a good attempt, you have the right idea in showing value of spy actions to demonstrate how ineffective it is. Cost, however, has little to do with the viability of spies. And everyone knows spies have a low chance of success at the other missions...

No.
I expect someone engaged in a discussion would be able to provide something more robust if they say effort and output are invalid.

Cost - This gets back to this point you made: "I believe we can present a better argument to Erik that spies need a hard rework ... I think he believes that they're still useful for the other things they're practically not useful for."
I believe Erik believes spies are useful for the other missions because of rewards and risks (aka the benefits vs the costs) I think he is of the opinion that while the risk of a failed attempt is high, the potential rewards are also so high that the mission is worthwhile. There are larger benefits associated with higher failure-chance missions, and so they are worthwhile because they will cost you less than you are rewarded by them.
Additionally, I think he believes spies gain sufficient experience at a fast enough rate to convert even the lowest success chance missions into viable ones quickly enough to be relevant for gameplay.

Nightskies wrote: Thu Feb 09, 2023 2:22 pm edit2: I think the idea of tracking 10 Zenox spies and using them with maximum care under Spymaster style automation settings and presenting that as raw data will be quite useful. Then calculating how often one devoted to this style can be expected to do *something* with a major impact (aside from tech/maps) with *any* of the 10 spies against a rival empire or stronger. If the probability is less than 50%, I think that should constitute proof that espionage needs rework?

You could have looked at the raw data for Arramy Kodansta performing missions provided in an earlier post in this thread.
Obviously, you cannot have a non-integer success, and fractional achievement won't give you results, so decimals should be rounded down. So in 10 attempts, this is the number of successes you would actually expect for one spy doing each type of mission.
As closing thoughts, from previously in the same speech:

Theodore Roosevelt wrote:Let the man of learning, the man of lettered leisure, beware of that queer and cheap temptation to pose to himself and to others as a cynic, as the man who has outgrown emotions and beliefs, the man to whom good and evil are as one. The poorest way to face life is to face it with a sneer. There are many men who feel a kind of twister pride in cynicism; there are many who confine themselves to criticism of the way others do what they themselves dare not even attempt. There is no more unhealthy being, no man less worthy of respect, than he who either really holds, or feigns to hold, an attitude of sneering disbelief toward all that is great and lofty, whether in achievement or in that noble effort which, even if it fails, comes to second achievement. A cynical habit of thought and speech, a readiness to criticise work which the critic himself never tries to perform, an intellectual aloofness which will not accept contact with life's realities - all these are marks, not as the possessor would fain to think, of superiority but of weakness.

While I agree with Jorgen's excellent suggestions, I wish there was a Space Empire's 4 type option to just turn off espionage. It's pretty much just a player exploit at this stage anyway, that the AI gets far less benefit from.

Diplomacy OTOH needs razed to the ground and rebuilt. Having more automation options, as per Jorgen's suggestions, would be great. As ever, the more granularity there is in any of the automated systems the better it is. You could have a training wheels option for people who are afraid of customising the automation settings; eg have a few diplomacy templates within the framework of each race/faction.

But the most sophisticated player diplomacy options are fairly pointless if enemy AI acts in a cookie-cutter manner. I appreciate that it's a deliberate game mechanic, but it'd be a treat if the enemy AI wasn't quite so keen to slit its own throat all the time.

But, fundamentally DW2 is a wargame, so lets get THAT fixed first.

Also, is "paragon" code for "asshole"? Asking for a friend.

maggiecow wrote: Fri Feb 10, 2023 3:22 pm Also, is "paragon" code for "asshole"? Asking for a friend.

Ok, personal insults are against the forum rules, so consider this an official warning to keep things civil. Per our forum system, you get a warning and after that comes a temporary ban. I'd rather not go there, so please stop with the insults.

While I'm at it, I'd also like to remind StormingKiwi of that. I've seen you dismissing and belittling other people's input in more than one thread and I think it ends up crossing the line as well. You have posted some excellent analysis, but your social skills need improving. Keep it civil and we'll get along fine.

Regards,

- Erik

maggiecow wrote: Fri Feb 10, 2023 3:21 pm While I agree with Jorgen's excellent suggestions, I wish there was a Space Empire's 4 type option to just turn off espionage. It's pretty much just a player exploit at this stage anyway, that the AI gets far less benefit from.

Either dial up the success chance to max, or disable all automatic spy missions - and the system will fade into the background as spies sit idle or merely do counter intelligence.

Personally, I would like both espionage and diplomacy to be more like research. Some (large) amount of budget can be devoted to espionage or diplomacy, a proxy for building of the necessary institutions and networks. Characters can have a specific additive or multiplicative impact on overall efforts or specific missions, but do not prevent you from spreading your "spy bandwidth" among your choice of target missions (like research projects). Your desired risk (e.g. 80%) determines the amount of time a mission "builds networks" before starting to fire attempts with risk of failure, with relative network vs network size impacting how long it takes. E.g. the prep to get to 80% risk with a comparable empire could take a few years, for a hard mission a decade or two, for a impossible one never. This could offer more options to pursue spy missions early on (at long, long duration), instead of just defaulting to counter intelligence indefinitely or until some facilities are available to help.

To avoid further personal exchanges, I'm going to lock this thread, but I want to say that a lot of the feedback, analysis and ideas above presented by various posters are very good. StormingKiwi's charts are always valuable, even if they don't encompass everything.

It is intended as Espionage works right now that you use your Ambassadors in conjunction with your Spies for the best results and that the most difficult missions should be difficult high risk/high reward endeavours that could cost you a good spy. As was pointed out, there are many possible ways to fail not all of which result in the loss of your spy.

The success chance is also different from the enemy's "interception" chance which is based on how long the duration of the mission is as well as how many agents they have on counterespionage.

In terms of adjustments in the relatively near future, I'm thinking we need to tone down the decrease on success chance caused by agents on counterespionage and instead have them mainly affect those intercept/abort chances.

In addition, we will likely buff the spy academy a bit and perhaps add some additional intelligence utility to the diplomatic techs (language/culture/etc.).

It's also pretty clear that the current policy risk threshholds are not properly adjusted to the success chances we have at this time.

Larger changes, such as turning spies into spymasters or so on would be beyond the current horizon. We do expect in the long run to take each "department" of your empire, such as say Intelligence or Diplomacy or Research and make it something that is run by a sub-leader of sorts (A Spymaster or a Chief Scientist, etc.) and give each such department its own budget and priority as well. That's much farther in the future though, but we'll get there eventually.

Right now, focusing on balancing the systems we have to a better point gameplay-wise and fixing any bugs in them is very realistic.

Regards,

- Erik