Ah, that's good. My mistake then. I only glanced at the system overview and forgot that the (16) is the expected growth turn, not the cap. Sorry for this mishap, having way too many things on my mind to handle concurrently these days. Thank you for clearing this up. Looking forward to start playing properly now.
No worries, it happens, (I actually didn't realize you picked that trait till you mentioned it). Enjoy the session, I'm not going to spoil it for you but I have already begun, and while I know our events will be different, there are too may spoilers if I post anything now. I will wait then, perhaps until you are done and give an abbreviated account of how my own game went using your setup. I've taken notes and screenshots along the way so I don't forget anything substantial. Enjoy the game
Cheers. Feel free to begin your own AAR right away. It's better that way to report findings while they are fresh on the mind. As our AARs are in separate threads we should be able to handle it intelligently. If will refrain from reading your AAR too early too far. Thinking about that, I will also put my own AAR in a separate thread in the subforum, so that we may be free to continue discussing high-level aspects here.
Turn 1 to 68 As things turned out to be tight in this scenario, I don't think your game will be much different than mine, unless you have a lucky strike. Your empire is pretty much boxed in, and in need of a creative strategy to gain an edge. I guess many players will not enjoy a challenge like this, but the fun might start once you have established a sound base.
Turn 68 to 108 @Konstantine I enjoyed your well written report. It's nice to see how you tackle the same challenge with a more militaristic outlook. The constrast and alternating viewpoint should also make for nice reading if visitors stumble upon it. My next report will take a bit more time, as this evening is a different game where Mbappe & co kick forward the ball.
Well sir, you are about to enjoy it even more as we diverge greatly. The beauty of this is that we will both have viable ways to win. Enjoy the game!
Turn 122 to 143 Recommendations (and bugs) Allies with a mining treaty settle planets in your empire. This doesn't seem right. Tech breakthrough chance doesn't seem to work any longer (1.0.4). My faction has now a tech breakthrough chance of around 20%, but it didn't even fire once in 143 turns. @Adam Solo By now I have seen enough to give you a solid first assessment how ISG fares in comparison to MoO2, if you like.
Turn 154 to 167 @Adam Solo I can now give you my final feedback on how ISG fares in comparison to MoO2. I don't think it will change after this point. First, do not be disappointed if there is a gulf between your vision of ISG and what I have seen. It is evident that ISG is a labor of love, it shows in so many aspects of the game, from the music and sound design, over the detailed lore and characteristics of the leaders, the events galactic and colonial, the exploration of astronomic phenomena, to the overall design of the game's systems itself. A lot of thought and love has gone into this, undeniably. Yet, MoO2 is a masterpiece that stood the test of time. I could go back and play it on the hardest difficulty, and it would give me a fair but harsh challenge which I may loose or win. Nothing of this is true for ISG. If you had a peek at Konstantine and mine AARs, you will notice that the computer empires on the hardest difficulty, with their most favourable starting conditions, underperform vastly in comparison to a skilled player. This goes for everything, economy, production, research, colonization and military. The latter is the most severe drawback. I don't fault you for not implementing an advanced AI (hardly anybody can do this today), but I fault you for claiming that ISG is a grand strategy game in the vein of MoO2, because this does require a competent AI. You should be honest with your customers and tell MoO2 veterans with skill that they won't find a challenge here. Yet you have the basis for a wonderful game that is just as worthy to play as Firaxis' Civ V for example. The other thing is the UI design. Maybe I am a bit harsh here. When I studied computer science I worked for a lab that did software ergonomics. I simply was a programmer, but they explained things to me, and eventually their way of thinking stuck with me. While the field has advanced far since these days, the general principles hold true. There are so many instances of poor UI design in ISG that I felt uncomfortable at times. For instance, that screen that notifies you of a possible combat engagement when your fleet enters a new system, and that then requires you to leave the screen to cancel the engagement and then again requires you to confirm this action, with a long wait state to boot, is poorly thought out. It is completely redundant, you could easily replace it with an icon on the right hand side that notifies you of a possible combat engagement. I will leave it at that. There are a lot of technicalities I could explain. If you are interested, send me a PM, because I wouldn't want to go into this in public. At any rate, I didn't feel that my time spent playing ISG was wasted. I enjoyed figuring out how to play the game. But as a single-player strategy game with replay value it does not work for me. @Konstantine I will gladly accept your challenge, but we should do it a bit differently. Here is my idea: You design our starting race. I set up the game. For the most part it will be a huge galaxy with mostly vanilla rules similar to the last one, but with more customized factions. Then we will agree on a set of victory conditions. Not the game ones, but more involved conditions that will also provide for a more fair competition. Once we agree on the victory conditions the game starts in earnest. The player who meets the victory conditions first wins (turns count, not real time). AAR rules: a requirement to document all galactic, colony and survey events, failed or otherwise, so that we have some understanding how each other fares. All other AAR documenting is voluntarily. I certainly will do more, although this may help you, but it will make the game more fun. I think we both agree that the AI will not present a formidable obstacle to either of us. But with a setup as explained above, we can still play and have a real challenge. Thoughts?
Thanks for your feedback. I'm glad you enjoyed your time ISG so far. As for the lack of challenge you report, please see my reply below. Regarding challenge, or lack thereof in your opinion, and the state of the AI, we have to be clear about one thing. The AI in ISG does not cheat and does not get bonuses. Apart from some advantages at game start, the AI never gets any bonuses or special treatment of any kind during the entire game. The AI may provide a challenge with that, but with time skilled 4X gamers and MoO2 veterans may report the same you do, that MoO2 offered more challenge than ISG on Impossible. The thing is, and I don't know this first-hand but only by reading forums about this, is that MoO2's AI (as for any AI in 4X games I know of), all of them cheat in the sense of getting insane production, research bonuses and other restrictions that do not apply to the AI and they get more leeway with things. MoO2 AI's seem to enjoy from all this and in specific they also seem to get more points for extra modifiers and special abilities for their races, more space in their ships, more command points (so they can field way more ships), you name it. Check here and here as examples where people describe these bonuses. Again, I do not have first-hand data, I'm just relying on what people report. Seems convicing that MoO2's AI "cheat" (aka receive bonuses and special treatment), but I have no way to confirm this so I apologize in advance, especially to MoO2's devs if that is not the case. Our idea for ISG was to offer a competent AI that did not cheat. We trusted our abilities to program a decent AI and we took the challenge to offer an AI that didn't cheat (didn't receive bonuses during the game or any special treatment of any kind). We accomplished that in my opinion. The AI does not cheat. However, that will not be enough for the "more skilled" and "veteran" 4X players, or to put it simply, people that seek the extra challenge. This need, to offer the extra challenge was also reported by others, and that's why we'll be offering a way for you to set the bonuses the AI receives, or penalties, as you see fit. We'll start with research and production bonuses. We'll also keep improving the AI to bring it to its best in all areas of the game, especially on the higher difficulty settings. Many players don't seek the extra challenge though, so having a way to set economic penalties to research and production speed to the AIs will also suit these players, so in the end everybody wins. We look forward to offer the research and production efficiency sliders soon, so by then I recommend you try Impossible again in ISG setting the bonuses to the AI you feel are fair (like supposedly MoO2 also does), and then please report back to see if the challenge is enough now. We're aware the UI needs improvement. That long wait when cancelling engagements is a case of poor feedback we already idenfied, along with poor turn processing feedback and other areas that need better feedback. I can say that the UI will keep receiving love and that we'll get there both in terms of polish and usability to what the UI is concerned. If you have other tips on where you think the UI needs more work please send me a PM as I'm very interested to know your recommendations. Sent you a PM just now. I want to hear all your thoughts on what could be improved. Thanks for your time. Thanks again for your feedback and I look forward to hear your further thoughts when we offer the economic efficiency modifiers, as that should provide a huge boost to the AI challenge.
A few... I propose a compromise I design the custom race You generate the start for both of us, huge map, Im[possible/Emperor Default races plus the custom race. 7 total Few events (they help too much) We agree on victory conditions Agreed on documenting all event types How does that sound?
@Quantomas, Your most significant complaint seems to be the difficulty levels not being difficult enough. As @Adam Solo pointed out, I'm not aware of any other 4x game that doesn't give the AI significant cheats--MoO2 was no exception, massively so at the highest difficultly levels. The simple fact is the state of gaming AI will never (or at least not anytime soon) come close to an experience player in complex games like this--there are simply too many variables to consider. Attempting to implement difficultly level simply by starting conditions with no AI cheats is an admirable goal--but it's simply not possible yet. @Adam Solo is on the right track with introducing cheats, though I would suggest picking them be limited to a "Custom" difficulty level. For standard difficulty levels I'd suggest just offering a checkbox for "Allow AI bonuses" which would pick some determined by the code but not shown to the player--players tend to be more accepting when they vaugely know the AI is cheating but not the details (which would also allow more easily adjusting to them). Some ideas would be--first the obvious ones of just taking the current race customizations and adding bonuses on top of them--stuff like +4 research, industry, culture, taxes per pop; growth bonuses; attack and defense bonuses; etc. Others along the lines of things often implemented are techs discovered by the player as 1/2 cost of the AI; instant free retrofitting of ships where ever they are; building that doesn't even rely on construction; automatic sentry fleets with a random element around ever system that stay there, so no system left defenseless; increased hostility towards the player and friendliness to other AI; increased chance of negative events for the player, possibly including the appearance of a large pirate force that only targets the player; a "virtual leader" with bonuses to initiate, etc., for all AI fleets scaling with game progression, etc.; automatic building of outposts on strategic resources without needing ships to go there; etc. In short, the problem isn't with the AI itself--it is quite good. You're just comparing it MoO2 with cheats at higher difficulty levels, which Adam says are planned as an option. So I wouldn't make a final judgement yet.
Few events is ok, although I like their flair. The default races are not optimized well. Is it possible to add more races and/or customize the existing ones?
Yes, I like their flair too, (I wrote some of them), but they tend to help far more than they hurt, i want to diminish the player advantages this time. You cannot customize the existing races, but don't worry about the AI races not being optimized well, neither are my Terrans, it should be a fair contest. Edit Up to 8 races may play at the same time, but I prefer we leave it at seven, the custom race against all standard AIs
@Adam Solo I will say more relating to ISG in private. In general, a 4X game employing cheats is not attractive to me, except for the sightseeing and initial exploration. There were two high profile grand strategy games that stood above the rest over time: MoO2 and Heroes of Might and Magic III. It is a myth that you cannot write an advanced non-cheating AI. These two games had exactly that, and this was what made them great, because it gives them an integrity that permeates to all features of the game. The AI of Heroes III doesn't cheat. I have seen the source code, it's a beauty to behold. There are smaller things that could be seen as cheats, but these are more like rounding off the edges. The same goes for MoO2. I'd treat comments from players that play with 100+ ship fleets with caution. That's not the way to play the game, you never need fleets bigger than 20 ships, even to defeat the Antarans or capturing Orion. The one thing that MoO2 did, and this is very smart, on the higher difficulties it added more race pick points to your opponents, essentially giving a randomized extra trait which is different every game. Now, this is very powerful, as this allows your opponents to grow and build up considerably faster. But still, for this to work, it requires a top-notch AI. I haven't seen the source code for MoO2, but based on my experience of countless matches on the highest difficulties, and being a guy who is highly aware of anything that remotely resembles a cheat, I have severe doubts that MoO2 used anything to cheat beyond the race pick setup. For me, it didn't feel like cheating but rather encountering a superior race. Again, the reason why they could hold up against me wasn't these picks, it was that the AI knows the ropes. You will be surprised.
Yes, You can create as many customized races as you wish, set their biome preference and race pic among the existing choices. For example, you could create a custom race based on the existing races, then when setting game parameters, disable the stock races, and choose how many players you have in the game. In this instance, only custom races would be in play. For our own exercise, I don't recommend this, as we should receive a far greater challenge from the stock races than the Lemurians did, and I want more data on how the stock races perform when the player uses a race that doesn't really excel at anything all that much. To be honest, I can create a massively powerful race, and set it against custom AI races that are also fully optimized... but it would dampen the experience for me. Knowing that the races I am up against are well optimised, I would be then tempted to optimise my own, as well as my game play, skipping parts of the tech tree, etc. I would then prefer we play with a good but flawed race, no more or less optimized than the AIs, while giving those AIs all the advantages allowed in 1.04 so we can both get an additional feel of the challenge posed, but not be reduced to playing a pure puzzle with a single solution.
On cheats in MoO2 @Quantomas @Adam Solo The AI in MoO2 did in fact cheat, though I am not certain to what degree. One recurring thing that always happened on the harder levels, was that your tech was stolen without the AI actually having spies assigned against you. A nice little cheat to give their research a boost. Command point limits may or may not have been ignored, as it is possible the AI simply switched over to producing money to cover the over cost. But, having played that game to death, I do not believe that it was a matter of the AI having a few extra picks only on the harder levels, far more likely, in addition to this, they had reduced research and production costs. The buffs given to ISG races on severe and impossible are actually quite potent, extra pop, extra infrastructure, free techs and on impossible, a mine and econ cap. Giving them production and research bonuses on top of this will only delay the inevitable, not alter it (for veteran players) The AI needs adjustments in other areas to become far more challenging than it is now, areas specifically dealing with war-time behavior and war making capability would be a great start.
My idea was we play with your Terran race, which I understand isn't heavily optimized, or even gives us a few burdens. As the AI is quite weak, playing on emperor/impossible is a must, but in addition to that I would like to have better optimized opponents, to give us at least a modicum of challenge. I don't think the data that we gather playing against the stock races has any meaning at this point, as the AI would need a massive upgrade to be even modestly challenging. If you like you can even give your Terrans a few penalties, while I optimize our opponents. One would be the Lemurians. This is the setup that makes most sense to me. I would also like to look at a game with twelve opponents in total, simply to see how that impacts gameplay, with space being more scarce.
Unfortunately 8 players is the maximum allowed, Very well, go for it, optimize them all then, though I do not entirely agree that the stock races would be so easily dispatched with a less optimized race, your Lemurians were very specialized and somewhat OP. Even though I played them in a different manner than you, they still kicked butt. Here is how you need to set up the Terrans Dictatorship Repulsive Warlord +1 research +1 production No unique racial traits at all Let's give it a go
