Impressions after playing 10 years in game
Moderators: Balthagor, Legend, Moderators
- Zuikaku
- General
- Posts: 2394
- Joined: Feb 10 2012
- Human: Yes
Impressions after playing 10 years in game
So, these are mynotes taken during the 1914-1924 gameplay.
1. Research center building is doing well for now. US got 7 of them, China 4, UK 7, France 5... I just hope AI will now stop building them at some number or we will have a problem. How about setting limit on research centers if AI don't know how to stop building them. Is the number of research centers AI wants dependant on something (population?) or are they just randomly built?
2. Researching of the units is now satisfactory. AI do research new unit designs. Nice change, no complaints here for now.
3. Building of unit production facilities. AI is doing well by building them, but there are same concerns as with #1.
4. AI is somewhat bad at building airfields. It does build aibases when air production facilities are being built, but I'm not sure if it builds airfields at all.
5. Naval AI shows signs of intelligence. Much better than disaster before, but still needs improvements. Nice changes, though...
6. Espionage is great, especially AI activelly (over)using spies. I'm having more troubles with sabouters than bombardments
7. Military goods production is inadequate. This needs some serious tweaking. Wars were left dead in the water by production at 1/10 of the needs. I used trainer ( ) to add MGs to all AI major powers and wars immediatelly turned out to be very interesting, vivid and surprisingly good. So, I strongly suggest increasing MGs output.
8. AI not building roads (issue in '14/17 games after longer gameplay)
1. Research center building is doing well for now. US got 7 of them, China 4, UK 7, France 5... I just hope AI will now stop building them at some number or we will have a problem. How about setting limit on research centers if AI don't know how to stop building them. Is the number of research centers AI wants dependant on something (population?) or are they just randomly built?
2. Researching of the units is now satisfactory. AI do research new unit designs. Nice change, no complaints here for now.
3. Building of unit production facilities. AI is doing well by building them, but there are same concerns as with #1.
4. AI is somewhat bad at building airfields. It does build aibases when air production facilities are being built, but I'm not sure if it builds airfields at all.
5. Naval AI shows signs of intelligence. Much better than disaster before, but still needs improvements. Nice changes, though...
6. Espionage is great, especially AI activelly (over)using spies. I'm having more troubles with sabouters than bombardments
7. Military goods production is inadequate. This needs some serious tweaking. Wars were left dead in the water by production at 1/10 of the needs. I used trainer ( ) to add MGs to all AI major powers and wars immediatelly turned out to be very interesting, vivid and surprisingly good. So, I strongly suggest increasing MGs output.
8. AI not building roads (issue in '14/17 games after longer gameplay)
Please teach AI everything!
-
- Brigadier Gen.
- Posts: 768
- Joined: Jun 27 2015
- Human: Yes
- Contact:
Re: Impressions after playing 10 years in game
I was playing S. Africa 2018 Trump scenario. Blietzkrieged to Egypt annexing everything in my way. Time i was in Egypt i had 64 research centers... No AI dont know when to stop, went thru Israel, Iran, Pakistan, then had 114 research centers. Russia got Europe and about the same number of research centers.... You only need freaking 10 with the right priorities for your region. Iran had like 30 research centers 4 years in the game.
Gameplay 1st
- Zuikaku
- General
- Posts: 2394
- Joined: Feb 10 2012
- Human: Yes
Re: Impressions after playing 10 years in game
That is not good. So, I can expect some sort of unbearable AI research frenzy in my game. What a waste of AI resouces!! I guess AI will be researching '90s techs by '50sYoMomma wrote: ↑Oct 14 2018 I was playing S. Africa 2018 Trump scenario. Blietzkrieged to Egypt annexing everything in my way. Time i was in Egypt i had 64 research centers... No AI dont know when to stop, went thru Israel, Iran, Pakistan, then had 114 research centers. Russia got Europe and about the same number of research centers.... You only need freaking 10 with the right priorities for your region. Iran had like 30 research centers 4 years in the game.
Please teach AI everything!
-
- General
- Posts: 131
- Joined: Oct 17 2013
- Human: Yes
Re: Impressions after playing 10 years in game
Thanks for the feedback after such a long game. I will pass it on to the team and see if I can get you answers to some of your questions. We did do an update to the AI building research centers over the summer.
- Zuikaku
- General
- Posts: 2394
- Joined: Feb 10 2012
- Human: Yes
Re: Impressions after playing 10 years in game
Looking forward for any feedback.
Please teach AI everything!
-
- Brigadier Gen.
- Posts: 511
- Joined: Apr 05 2012
- Human: Yes
- Location: Italy
Re: Impressions after playing 10 years in game
yes...now it work!! also AI work in this way!!
now the problem is : when region annex other, the number of research centers increases a lot.
Con forza ed ardimento
- George Geczy
- General
- Posts: 2688
- Joined: Jun 04 2002
- Location: BattleGoat Studios
- Contact:
Re: Impressions after playing 10 years in game
Thanks for the feedback! Some notes below:
-- George.
The AI building of research centres is generally tuned to AI regions on a region-by-region basis, based on their population, GDP, treasury, etc. So if France continues to be just France, the number of research centres should make sense for a long term game. The issues mentioned above by dax1 and yomomma are what happens when regions fall and are combined - if this happens to a human player, yes, you will end up with a lot of excess research centres, and it is up to you to scrap the ones you don't need. (I suppose we could add a minister priority to manage this automatically, but there isn't one at present.) For AI regions that end up with merged regions (annexation etc) this is an issue because I don't think that the AI considers scrapping them. So yes, an aggressive AI that takes over a number of other regions in later games could end up with too many.Zuikaku wrote: ↑Oct 13 2018 So, these are mynotes taken during the 1914-1924 gameplay.
1. Research center building is doing well for now. US got 7 of them, China 4, UK 7, France 5... I just hope AI will now stop building them at some number or we will have a problem. How about setting limit on research centers if AI don't know how to stop building them. Is the number of research centers AI wants dependant on something (population?) or are they just randomly built?
2. Researching of the units is now satisfactory. AI do research new unit designs. Nice change, no complaints here for now.
Given the fact that bigger regions often have more bases, the combining of unit facilities in annexed regions hasn't generally been a problem, but I suppose in some cases it could be. I've made a note to take a look at how the AI handles "too many" bases.3. Building of unit production facilities. AI is doing well by building them, but there are same concerns as with #1.
Hmmm... good question, I've made note to look at it. Airfields do make nice supply sources.4. AI is somewhat bad at building airfields. It does build aibases when air production facilities are being built, but I'm not sure if it builds airfields at all.
We've worked on it in the summer 2018 updates so glad it's paying off a bit. Yes, still more work to be done, there is very poor fleet/escort behavior, and no effective use of aircraft carriers yet.5. Naval AI shows signs of intelligence. Much better than disaster before, but still needs improvements. Nice changes, though...
6. Espionage is great, especially AI activelly (over)using spies. I'm having more troubles with sabouters than bombardments
Seems like we spend a lot of time tweaking output up and then down and then up and etc. We made some major improvements in the summer updates for games that went from 1936 into the modern era by fixing the progression of game engine values (the Cold War maps use a slightly inconsistent set of values because of their earlier design). Since this WW1 era you're talking about, there was an intention to have finished goods shortages because that was a real world thing as nations were in the midst up modernizing, but maybe we should take a closer look.7. Military goods production is inadequate. This needs some serious tweaking. Wars were left dead in the water by production at 1/10 of the needs. I used trainer ( ) to add MGs to all AI major powers and wars immediatelly turned out to be very interesting, vivid and surprisingly good. So, I strongly suggest increasing MGs output.
I still would prefer more of these to be added using scripted events so that they could be realistic to history, but yes this is a popular request. It is expected to get some attention at some point.8. AI not building roads (issue in '14/17 games after longer gameplay)
-- George.
-
- General
- Posts: 2550
- Joined: Dec 08 2007
- Location: Tipton, UK
Re: Impressions after playing 10 years in game
Nice to hear that the naval AI is getting some love and we may see proper carrier use at some point. I had noticed the AI being different in its behaviour at sea, and even the odd carrier with planes on it. Have you seen my suggestions on potential behaviour elsewhere on the forum?George Geczy wrote: ↑Oct 15 2018
We've worked on it in the summer 2018 updates so glad it's paying off a bit. Yes, still more work to be done, there is very poor fleet/escort behavior, and no effective use of aircraft carriers yet.5. Naval AI shows signs of intelligence. Much better than disaster before, but still needs improvements. Nice changes, though...
Seems like we spend a lot of time tweaking output up and then down and then up and etc. We made some major improvements in the summer updates for games that went from 1936 into the modern era by fixing the progression of game engine values (the Cold War maps use a slightly inconsistent set of values because of their earlier design). Since this WW1 era you're talking about, there was an intention to have finished goods shortages because that was a real world thing as nations were in the midst up modernizing, but maybe we should take a closer look.7. Military goods production is inadequate. This needs some serious tweaking. Wars were left dead in the water by production at 1/10 of the needs. I used trainer ( ) to add MGs to all AI major powers and wars immediatelly turned out to be very interesting, vivid and surprisingly good. So, I strongly suggest increasing MGs output.
-- George.
I'm wondering if the military goods thing is related to the same problem I have in my long term games - no matter how many facilities I build, production doesn't seem to go up by much, even if I build several times the amount of facilities and put them in good supply etc. I think something must have happened along the way where what a building should produce is somehow fudged.
My SR:U Model Project, get the latest and post suggestions here:
http://www.bgforums.com/forums/viewtopi ... 79&t=28040
http://www.bgforums.com/forums/viewtopi ... 79&t=28040
- Zuikaku
- General
- Posts: 2394
- Joined: Feb 10 2012
- Human: Yes
Re: Impressions after playing 10 years in game
Thank you for the feedback, Balth. It looks promising and optimistic in the long run.
Please teach AI everything!
- Balthagor
- Supreme Ruler
- Posts: 22106
- Joined: Jun 04 2002
- Human: Yes
- Location: BattleGoat Studios
Re: Impressions after playing 10 years in game
Sadly, I can't take the credit
-
- General
- Posts: 2550
- Joined: Dec 08 2007
- Location: Tipton, UK
Re: Impressions after playing 10 years in game
I don't know about that, you usually pass on info for us.
My SR:U Model Project, get the latest and post suggestions here:
http://www.bgforums.com/forums/viewtopi ... 79&t=28040
http://www.bgforums.com/forums/viewtopi ... 79&t=28040
- Zuikaku
- General
- Posts: 2394
- Joined: Feb 10 2012
- Human: Yes
Re: Impressions after playing 10 years in game
Like SGTScuba said, when I see red letters in the nickname I automatically presume it is you (Balthagor).
Please teach AI everything!
-
- General
- Posts: 2550
- Joined: Dec 08 2007
- Location: Tipton, UK
Re: Impressions after playing 10 years in game
Is there any chance the Goats will look at the economy alongside the upgrades for the AI?
My SR:U Model Project, get the latest and post suggestions here:
http://www.bgforums.com/forums/viewtopi ... 79&t=28040
http://www.bgforums.com/forums/viewtopi ... 79&t=28040
-
- Sergeant
- Posts: 24
- Joined: Apr 11 2015
- Human: Yes
Re: Impressions after playing 10 years in game
I'm really hoping they eventually get around to adding the ability to release/create countries and rebellions, since that hurts replayability when you're trying to play a long game. :C
-
- Brigadier Gen.
- Posts: 884
- Joined: Nov 11 2011
- Human: Yes
Re: Impressions after playing 10 years in game
This is my no.1 concern as well, as I like to play really long games (10-50 years) and it usually becomes too static.
Perhaps certain scripts could be run to make countries release vassals or free states if their military/domestic approval is really low (or under other rare conditions)?
It would also be nice if smaller countries would be building more military goods and fabrication facilities as well. In certain scenarios some countries have 0 units (not sure if they traded the few they got off or what) and it quite breaks immersion. This was extremely apparent in my 1936 Lithuanian AAR for example:
viewtopic.php?f=94&t=23694
Latvia, Danzing and Estonia had absolute 0 of troops. I think they didn't have garrisons either. They probably had 0 fabrication as well or traded their units away? Anyhow, 0 shots fired when I annexed them.
I really like small countries giving a good fight. And small countries is what really makes the game fluid and dynamic, I strongly believe.