December 17th, 2023, 02:34
(This post was last modified: December 17th, 2023, 02:35 by GMBarak.)
Posts: 289
Threads: 20
Joined: Jan 2022
To COM23:
What do you dislike about COM II and what are the problems in it?
What would you like changed?
December 17th, 2023, 13:26
(This post was last modified: December 17th, 2023, 17:18 by Desertfox.)
Posts: 52
Threads: 2
Joined: Jul 2019
- the graphics are more modern but the portraits of the female wizards follow the woke push in video gaming to make women look uglier, and I don't like that.
- in strategy game design there is a constant conflict between realistic graphics for immersion, and abstract symbolic representations that are as barebones as possible to allow for an at-a-glance overview of the situation. The computer versions of the board games Chess and Go, still have very simple graphics, and making those games more graphically complex won't improve those games. I feel that the new graphics sometimes lose overview due to overcrowding of details. The original game had a near-perfect balance between the two (realism and symbolism).
- on the other hand, replacing the little dolls in the city screen that represented the farmers and workers of the population in the original game, with numbers you can +/-, is too much in the other direction, i.e. too much abstraction.
December 19th, 2023, 21:57
Posts: 12
Threads: 0
Joined: Jul 2020
(December 17th, 2023, 02:34)GMBarak Wrote: To COM23:
What do you dislike about COM II and what are the problems in it?
What would you like changed?
AI improvements are coming. Besides general competence, I don't want the AI to just be scripted to spam the best human dirty tricks to the point their bonus advantages + dirty tricks = near unbeatable (that's COM essentially), but for them to be different. Some will adopt rush strategies, some will turtle, some will want the world to burn, some are just peaceful etc.
They have done a good job in adding information like the combat logs (you can see how much damage comes from thrown, life steal ,gaze etc), city interface showing breakdown of info (production, gold, mana production) that MoM doesnt have, but more is even better.
More DLCs.. that are more radical.
Some of the new wizard retorts in the DLC are just too weak, while others are just right or amazing.
December 20th, 2023, 08:55
(This post was last modified: December 20th, 2023, 09:11 by WhiteMage.)
Posts: 634
Threads: 13
Joined: Nov 2010
In 4X TBS game development AI development was notoriously poor up until now and I predict that this will continue in the foreseeable future mainly due to the time, money, and expertise required to develop strong AI. AI is the biggest problem with 4X TBS games in general. That is the hardest part to develop by far. Meaningful AI development starts when the game rules are settled, and the game is in reasonably stable condition, game balance is set, bugs were fixed, exploits were removed. Typically, only about 5% of 4X TBS game development budget pays for AI development and is done by a person who knows very little about AI. A good AI developer earns over $300,000 per year and game development companies are unwilling to pay that. It would be financially a bad decision for them to hire some serious AI developers and paying them for a long time. And even groups of expert AI developers would take decades to develop a strong AI for a complex 4X TBS game like this.
Game development companies don’t give the AI developer enough time to develop a strong AI. When a game no longer generates enough revenue for the company, all development stops. AI suffers the most from this standard business way of thinking, but even the bugs, poor design decisions, balance problems, etc. remain in the games forever. Once a project is declared closed by a company, most original source codes are never going to be looked at again by anyone and there are additional copyright laws that apply making problems even worse, even harder to fix.
There are some volunteer AI developers out there who would happily work on AI development for their beloved games for free, but they are almost always not allowed to do so, since there is no external AI hookup support for most games. They would have to be hired into the company, but that rarely happens and even then only for short term contracts.
So instead of even trying to develop a strong AI, 4X TBS game development companies keep developing so called “fun AI”. They develop cheap scripted predictable AIs to “entertain” the users, instead of AIs that try to win. I am only interested in strong AI. It is not fun for me to play against AIs that only exist to entertain me, but do not try to win.
Winning a TBS game can be done in any way as long as it is within the game rules. No matter how dirty, exploitative, human-like etc. it is, they are all OK (to me and as a principle of strong AI development). These issues should be fixed by the game rules if they get too boring, annoying, etc. Not by making AIs avoid the strongest moves and not by only allowing humans use those against the AIs. Etc.
Handicap is a separate topic from AI development. It is an easy, intuitive, and interesting way to correct for the skill differences between players creating games where each side has roughly equal chance of winning. They apply it between human players too in Go and other games.
December 20th, 2023, 11:56
Posts: 52
Threads: 2
Joined: Jul 2019
Well, the CoM AI does try to win. If you neglect the defense of your capital, for example, the AI will sometimes send a large army, beeline for your capital, and conquer it.
December 20th, 2023, 12:11
Posts: 634
Threads: 13
Joined: Nov 2010
(December 20th, 2023, 11:56)Desertfox Wrote: Well, the CoM AI does try to win. If you neglect the defense of your capital, for example, the AI will sometimes send a large army, beeline for your capital, and conquer it.
Yes. That is one of many reasons why I rank COM II as the best game ever made up to now. And I rank 4X TBS as the best genre due to this genre prompting for the deepest thinking and providing diverse game experience through replayability and diversity of game options and actions. AI should pick up much of that but sadly it does not.
December 20th, 2023, 12:17
Posts: 12
Threads: 0
Joined: Jul 2020
(December 17th, 2023, 00:18)WhiteMage Wrote: (December 16th, 2023, 23:43)COM23 Wrote: (December 16th, 2023, 22:45)WhiteMage Wrote: (December 16th, 2023, 01:11)COM23 Wrote: On a larger scale, looking at these reviews I am happy.
Pretty much everything mentioned are minor details or nitpicks that can be easily fixed some even by Moddershall.Heck a lot are already fixed in the beta!
If these are the best objections to the game by diehard CoM fans , the remake has succeeded fundamentally with further improvements
To me these are not nitpicks. These are major flaws that make their tested version unplayable for me.
Also, these are not my biggest problems with the game. These are just the first 104 problems after my first 92 hours of game play. The game reached a point that I no longer see reason to waste my time on that version.
I agree that while almost all other 4X TBS games are unfixable, this is a rare 4X TBS game that appears to be fixable, although only theoretically. In reality most likely it will not be fixed by anyone it in the next 30 years to a barely playable level.
I am not happy with what they have done so far. As it stands it is a major failure with a very low probability to bring it up even to the level of MOM 1.31, let alone to the current COM II level.
Also, they brought additional shame and disappointment for both MOM and for 4X TBS. Both new and seasoned players are not likely to benefit from this version.
Please do not think you speak for everyone. CoM diehards are a small minority vocal camp. You guys are mostly bitter the remake isn't using CoM as a base and despite that this is the best you can say against it? Lol
The remake is doing very well and Sitherine even announced the sales of the base game and last DLC topped a past steam sale by revenue AND units.
If you go outside your echo chamber and see what people are saying you will realise the Remake is catching on despite attempts by people like you to damage the IP
I never said or thought that I speak for everyone. Quite the opposite. Also, I stated about 5% of my opinion so far on this topic.
I did not compare the new MOM to COM II. I evaluated it on its own merits.
I just spent a large part of my life playing, evaluating, developing, and publishing games; also on scientific materials like research, peer reviews, publishing, books, grant proposals, journal articles, etc.
What I see is a mixture of good and bad everywhere. As a scientist, I believe in free speech, data-based opinions, an open mind, probabilities, quantum thinking, constructive criticism, perfection, and dedication to the truth even if I am alone against everyone else, which comes up quite often.
I am not bitter. I am just both unhappy and happy in the same time based on the large amount of data I processed. The truth and science are not a democracy. I refuse to be intimidated into submission. Just because the majority has an opinion, I am not required to agree with it. I am not required to keep silent.
On a side note, I am not die hard for COM II. COM II has numerous problems and is far from my desires, far from perfection. But overall, I perceive it as the best game ever made up to now.
Also, I was not speaking against the new MOM. I just stated my review, the same way as I do for journal reviews, book reviews, grant proposal reviews, and in promotion reviews of other professors when I need to write their evaluations. Good and bad are all there then I pass it up for someone else to aggregate my review with other reviews.
Among other things one of my missions in my life is to create the best game possible or die trying.
Also, please don't wrap me together with other so called COM diehards. I am very different from the others here in many aspects including my opinions.
To me a game's value is not based on how much money they make with it. Greed is exactly the main reason why we have never seen any excellent game developed, at least in the 4X TBS genre. Excellent game would take knowledge, passion, stamina, domain expertise, coding expertise, and more than a just a 1 person effort. Not the capitalist view they push down on our throat over and over how to make the most money with the least effort and how to destroy the source code when the company goes bankrupt, and how to resell the same thing over and over again with different packaging, that we have seen so many times before.
Also, I have no echo chamber. No one echoed me and I echo no one, unless in rare cases it is scientifically justified and even then I provide data to back up my stated opinions.
Maybe I should ask my $40 back for this new MOM, since it does not even worth $1. But I decided not to do that, since I am very well aware about the huge challenges and difficulties it takes to develop an excellent 4X TBS game. For their effort, I respect the Polish team that developed this new MOM. And I am happy to support them financially. If I would see more than just effort, I may even offer them a few hundred thousand dollars like I offered before to other similar game developers...
I love a pompus ass who thinks he's the only academic here.
The fact he compares playing and talking about the game at the same level.as doing science is telling .. Probably some third rate "scientist"
And threatening to ask back $40 ? Lol. Are "scientists" so poorly paid in wherever you claim to be to need to do that?
December 20th, 2023, 12:27
Posts: 12
Threads: 0
Joined: Jul 2020
(December 20th, 2023, 08:55)WhiteMage Wrote: <rambling nonsense snipped>
So instead of even trying to develop a strong AI, 4X TBS game development companies keep developing so called “fun AI”. They develop cheap scripted predictable AIs to “entertain” the users, instead of AIs that try to win. I am only interested in strong AI. It is not fun for me to play against AIs that only exist to entertain me, but do not try to win.
Winning a TBS game can be done in any way as long as it is within the game rules. No matter how dirty, exploitative, human-like etc. it is, they are all OK (to me and as a principle of strong AI development). These issues should be fixed by the game rules if they get too boring, annoying, etc. Not by making AIs avoid the strongest moves and not by only allowing humans use those against the AIs. Etc.
Clearly our "scientist" doesn't know how AI works. How do you think COM AI works if not by scripts? Or do you think under the hood our genius AI coder is using state of art Reinforcement learning? lol.
Any tactic is OK? Then the game rules wouldn't have needed to be modified so many times because the AI can;t handle human tactics. Most famously the can't get armor guild before Turn X nonsense because the AI is too weak to survive a common sense rush strategy!
December 20th, 2023, 12:29
Posts: 12
Threads: 0
Joined: Jul 2020
(December 20th, 2023, 12:11)WhiteMage Wrote: (December 20th, 2023, 11:56)Desertfox Wrote: Well, the CoM AI does try to win. If you neglect the defense of your capital, for example, the AI will sometimes send a large army, beeline for your capital, and conquer it.
Yes. That is one of many reasons why I rank COM II as the best game ever made up to now. And I rank 4X TBS as the best genre due to this genre prompting for the deepest thinking and providing diverse game experience through replayability and diversity of game options and actions. AI should pick up much of that but sadly it does not.
Wow shocker! I did not expect that. So much for unique opinions in COM...
December 20th, 2023, 13:17
(This post was last modified: December 20th, 2023, 16:48 by WhiteMage.)
Posts: 634
Threads: 13
Joined: Nov 2010
(December 20th, 2023, 12:27)COM23 Wrote: (December 20th, 2023, 08:55)WhiteMage Wrote: <rambling nonsense snipped>
So instead of even trying to develop a strong AI, 4X TBS game development companies keep developing so called “fun AI”. They develop cheap scripted predictable AIs to “entertain” the users, instead of AIs that try to win. I am only interested in strong AI. It is not fun for me to play against AIs that only exist to entertain me, but do not try to win.
Winning a TBS game can be done in any way as long as it is within the game rules. No matter how dirty, exploitative, human-like etc. it is, they are all OK (to me and as a principle of strong AI development). These issues should be fixed by the game rules if they get too boring, annoying, etc. Not by making AIs avoid the strongest moves and not by only allowing humans use those against the AIs. Etc.
Clearly our "scientist" doesn't know how AI works. How do you think COM AI works if not by scripts? Or do you think under the hood our genius AI coder is using state of art Reinforcement learning? lol.
Any tactic is OK? Then the game rules wouldn't have needed to be modified so many times because the AI can;t handle human tactics. Most famously the can't get armor guild before Turn X nonsense because the AI is too weak to survive a common sense rush strategy!
I choose not to respond to your personal attacks as I did not initiate personal attacks on you. That does not belong here.
But yes, COM II AI is hard coded (call it scripted) as far as I know. I did not look into it deeply to be sure. As a result, COM II AI is poor. COM II AI is not like the Go and Chess AIs. All 4X TBS AIs are poor as of now. I don’t know a single exception. Developing strong AI for such games is much harder than developing strong AIs for Chess and Go. Each of those took 40 years by dedicated teams, major financial support, tournaments, and more.
Other major problem in all 4X TBS games is that their designs are poor with a few exceptions that are mediocre. Other than the bugs and balance problems, the rules are easily exploitable making certain strategies required or useless, etc.
Also, with changing rules, the existing AIs become useless.
So in the last 30 years I was fighting (as a hobby that I spent my own money and time on) for a good 4X TBS game design to be developed that at least allows strong AI development. I was only partly successful in this effort. Sadly, I am not aware of any excellent 4X TBS game design, but C-Evo, COM II, and Dominus Galaxia are the closest to it, in my opinion.
With good rules, any tactic is OK, yes. Just like in Chess, Go, etc. It is very sad that some game developers chose to change the game rules just to make AI development easier for it. I fought for the other direction. I wanted to develop a game that makes AI development as hard as possible, but not at the cost of human fun playing the game.
Finally, yes, scientific principles of expert peer review and data-based evaluations are the best way to develop deep thinking strategy games, as opposed to the main stream resale of old ideas but in decreasing quality of each new product, and focusing on the profits only followed by fanatics trolling each other online and starting personal attacks on everyone else who dare to have a different opinion.
With good rules, any tactic is OK, yes. Just like in Chess, Go, etc. It is very sad that the programmers chose to change the game rules just to make AI development easier for it. It fought for the other direction. I wanted to develop a game that makes AI development as hard as possible, but not at the cost of human fun playing the game.
|