Posts by Astral_Codex

    TL;DR: The new avatars and planet skins are ugly AI art incongruent with existing OGame art. Gameforge should at the minimum, include all existing OGame icons as avatars and planet skins, and allow players to turn off viewing other people's skins.


    What am I talking about?

    The OGame v12 content update was just announced: Avatar & Achievements


    I have many thoughts about the system and how it's implemented, but one issue stands out above all: the low-effort, low-quality AI art. Here's some examples from the above post:


    Look, I like AI art as much as the next guy, but this AI art is just shoddy -- almost shitty NFT-game territory, tbh. It'd be a real pity to be forced to look at these whenever you go onto the high score page (another thing made substantially uglier by the update, btw) or navigate in galaxy view. Not only are the examples uncreative and boring, they're also completely thematically inappropriate for OGame, which already has its unique art style. For example, compare the planets above, which look both fiddly and artificial, to existing OGame planet art, which is simple and realistic:

    Or, compare the avatars above to existing in-game icons:

    If you're going to use AI Art, at least put in the effort to make it both congruent with the rest of the art in the game! The actual avatars (you can view them on the PTS) are actually worse, because they're not even consistent in terms of what sort of style they're going for (except that every image is _very_ AI art).


    Worst of all, there are already many assets that have a unified thematic style would look better and would be thematically appropriate: the OGame in-game icons! It's really silly that (for example), unlocking slot 18 as Human gets you this AI-generated slop, instead of the Human Lifeforms image:


    qPcpBRP.png


    Or, imagine getting a metal 62 (which would cost something like 100-150 euros by itself, by the way), and unlocking this, when there's no way to unlock a metal mine or metal pack avatar:

    mOrJtlJ.png


    The only in-game icons you can use as images right now are the avatars for Rock'tal, Mecha, and Kaelesh (fittingly unlocked by unlocking slot 18 on a planet with the respective race) as well as the icons for Collector, Discoverer, and General (also fittingly unlocked as being the relevant class). I'd much rather look at the OGame icons, than shoddy AI art that often looks worse worse than what I can do with a few calls to ChatGPT.

    Suggestions and Possible Fixes

    Here's some suggestions that would alleviate or address this problem, in descending order of effort:

    1. (High Effort) Commission art that fits into the existing OGame style. This would be the best option in my opinion.
    2. (Medium Effort) Use AI generated art that fits into the existing OGame style. This isn't that hard, but it does take some amount of effort in terms of both prompt engineering and giving few-shot examples. It's also possible (albeit incredibly unlikely) that you can't easily get open source imagen models to mimic the OGame style.
    3. (Low Effort) Allow players to disable viewing other players' avatars and planet skins. If you're going to keep the ugly AI art, at least don't force players to view them every time they go to the Highscore or to Galaxy View. Even if the art was better, you probably still want to give players this option, since both avatars and planet skins serve no in-game function and are just clutter that distract from mechanics. If Gameforge can't do this, then I imagine someone (perhaps myself) will just make a script to hide avatars and display default planet skins.
    4. (Low Effort) Replace Avatars and Planet Skins with the appropriate existing OGame assets. I'd like to be able to use existing OGame icons such as the various researches, ships, and defenses, and I'd like to pick which of OGame's existing planet style I see on each planet (also on resources/facilities too!). But there's currently not an option to this! It'd also make a lot more sense to unlock the appropriate icon by building the corresponding research or building, instead of random AI art.

    I understand if GF doesn't want to go through the effort to commission art, and if they don't have the requisite skill or patience to generate AI art in the appropriate style. But the last two suggestions seem very easy to implement, and would greatly alleviate the issue of NFT shovelware-tier AI art.

    It's the exact same as the General +moon fields. If you have say, 250 fields base and get +30 fields from the bio modifier building, and then build to 270 fields, then swap to Rock'tal, you'll see that your planet has 270/250 fields. You won't be able to make more buildings until you swap back to Kaelesh. But you will keep your buildings that you already made.

    First, the effect of temperature on the production of colonies is greatly overstated. In terms of MSU, the difference in mine production between the average slot 15 and the coldest slot 15 is on the order of 2-3%, and mine prod is a small fraction of total income for discoverers and generals anyways. So it's adding a new, unnecessary item that has little effect on the game. (Keep in mind most players are way, way further away from 2-3% from the "optimal" build.)


    More importantly in my opinion: This screws over all the accounts that spent a lot of time colonizing planets over and over to get cold temperatures, in pursuit of "true" optimality. Some people like hedgefundpoe or Asto Vidatu have spent hundreds or even thousands of colonies in order to get big planets with minimum temperatures. Others have relocated their planets several times each in pursuit of lower temperatures.


    I mean, to be clear, it's fully in line with screwing over people who optimize their accounts. (See hedgefundpoe's recent experiences with screwing over his "perfect" 30C max temp slot 8 account. Or halving the price of metal packs after adding the imbalanced slot 8 production bonus. Or combat expos. Or allowing people to relocate to slot 15 from anywhere in the first place.) But I'll oppose this item every single time it's brought up.

    Thanks for writing this up. I like the suggestion of adding the eco multiplier to general/collector, albeit with a much lower cap. It's quite creative! If I were to implement it, I would suggest capping it at the 25kk/50kk level, instead of adding two new levels for discoverer. This would alleviate the early game issues of collector/general without seriously affecting late game balance. I also obviously really like the ACS Def wreckfield (I've already posted a suggestion that made it to GF).


    The 20% Phalanx for general instead of discoverer also makes sense to me.


    That being said I disagree with most of the changes. I'd go so far as to say that I think implementing them would probably be bad.

    1. I don't think we should do tiny number changes like 20% -> 25%. Ditto for small speed speed buffs (50% is small when lifeforms lets you get +500% or more from drives + lf techs). Why change things for the sake of changing things?
    2. Unless the two new levels on discoverer are ridiculous (e.g. 10m for 10T+), your current changes make collector even more overpowered in the late game, since losing 1/3 of expo finds in exchange for increasing your prod by a factor of 1.5x is almost always going to be worth it. Disco is barely worth it late game compared to collector when collector gets no expo finds, how would it compete if collector gets 2/3 as much expo finds?
      1. I also oppose changes that lead primarily to number inflation without addressing balance. Servers can barely handle big hits these days. So I'd prefer to keep discoverer as is, slightly nerf collector late game, and give collector/general the eco multiplier but with a much lower max find cap.
    3. The cost reduction on general is too good: no other classes have **cost reduction**. Rock'tal (the only race with cost reduction) is by far the best for most players in large part because of its cost reduction. In general, I oppose all cost reduction changes for ships/mines/etc.
    4. In general, the General changes (2x fleet spots?? 75% loot on all actives) are probably too good late game, while not helping much early game. The fundamental problem is that Disco and Collector both get more income with little effort from the start, but OGame's dying population means that there's not many targets early. General's primary benefits of -25% deut and offensive WF also doesn't matter early game when fleets are small.

    I think this post comes from the perspective that all classes need to be equally balanced for ~all players. But instead, I think it's better to balance the classes so each has a distinct use and encourage players to swap between them as necessary.


    Currently, all three classes have a use. Collector is the choice of most players late game, discoverer early game. General is more niche, but it's indispensible late game to save deut on fleetsaves and offensive wreckfield. Most big fleets in older universes are general for a reason. (Also we might expect fewer players to pick General since more players are miners than fleeters.)


    I think my actual suggestions for class balance look like:

    1. General gets ACS Def Wreckfield. I outlined the arguments here.
    2. Give out more class change tokens in reward events (ideally at least one a month). This is good for new players in that they won't be screwed over if they picked the wrong class by accident. It's good for older players in that they can easily swap between classes as needed to make ships or adjust their time investment in the game. I'd rather give players more flexibility, than homogenize the classes in the name of balance.
    3. Update the class choice menu for new players. For example, simply stating "we recommend you pick discoverer, it grows the fastest" would help a lot. Or linking to a good guide/FAQ on basic game mechanics would help as well.
      1. This isn't just a complaint about classes, I have the same issues with Lifeforms, where people get tricked into picking Kaelesh because of the misleading description.

    All said, I intend to vote against this suggestion and encourage others to do the same.

    I notice you're using default OGame without plugins. On the default OGame fleet page 2, it's really easy to accidentally change the speed by just overing over the speed bar. And since the speedbar is between target selection and clicking confirm, it's super easy to hover over it by accident.

    Given that the GAs may not understand how "OR" interacts with "NOT", the rule as stated is clearly too complicated.


    I'd be in favor of a change, albeit not necessarily this one. Here's one that's even simpler and does most of what you'd want anyways:

    Rationale: clearly the extra condition added was to save GA/GO labour. But the top 20 eco condition is pretty much unnecessary, as the main abuse happens in old unis with massive accounts, or at the start of new unis. So just make a rule that cuts those off, that is simpler than the currently stated rule.


    Also, I think it's very important to make it clear when rules are conjunctive or disjunctive. I've tried to make that explicit in my phrasing.

    Thanks for the response!


    Is there a reason this wasn't this communicated before the merge like the other settings? While I doubt this would've changed anyone's decisions, it's still pretty bad for people not to know what the settings of the universe they're merging into will be.

    It's like this:

    This flowchart is wrong as written or at least ambiguous. If anything your flowchart makes the rule more ambiguous/confusing and not less.


    Assuming "/" means "OR" in the figure, then by the second set of arrows, someone has 2bn points they can't pull. But if at the same time, if they are rank 21 eco they can pull by the third set of arrows. NOT (X OR Y) =/= (NOT X) OR (NOT Y), but the flow chart separates out the cases as if the equality were true.


    If you read the wording of the rule, it explicitly says that accounts who are either rank 21+ in eco OR <2bn in points are exempt. So this suggests that someone with 2.1bn points but is rank 21 in eco should be exempt.

    Quote

    For all accounts below TOP 20 Economy OR less than 2bn points this rule does not apply, except on the start of a universe.

    I presume the intention is presumably to reduce the work of GOs in monitoring the universe. You'll have way more accounts if you have to look at the top 20 eco in all unis plus the hundreds of accounts in .org that are >2bn points, than if you only had to look at the top 20 eco people who were also >2bn points. In the second case you'd only have to consider 20 accounts per universe at most.


    Please clarify.

    That is actually a good suggestion. Only thing the original suggestion does is render mds impossible for anyone besides whales.
    I understand the op's intend, and I do agree that mds need to be nerfed given the recent state of the game, but for people who insta 100k rips daily, 300 rips or 2400 don't matter, it does to the rest of us trying to counter-play.

    You keep saying that this would make MDs impossible. But before we had lifeforms or even player classes, 300 rips were worth maybe 3000 if not 6000 rips today, and yet MDs still happened.


    Back when I was actively fleeting in that era, the standard was to do 5/6 waves of 10-20 RIP MDs. A few years before that, people were MD'ing with waves of 3-5 rips.


    MDs _should_ be expensive. You _should not_ be able to force pop a player who has only 50b fleet and profit. If you still want to MD them after the change, either RNG with your 10-20 rip waves (which would become 100-200 RIP waves in Thuban, where you are), or admit that you aren't in it for profit.

    Given how they botched the Vega vote, I would imagine they'd split this one too, which means you'll just get 10x eco and not the other two.

    I feel kind of bad for Prongs, because he's clearly not the person who made this shitty choice. Unfortunately, those actually responsible hide behind CoMAs like Prongs.

    This is a nice patch that doesn't fix the underlying issue. Still in favor though, MDs are way too cheap these days.


    To clarify: the main problem is that MDs are too fast. You can't sleep properly if you want to FS safely in higher speed unis. The suggestion from OP raises the costs a bit, but 1.5-2.5k RIPs is still nothing for people with hundreds of billions or trillions of fleet. You still can't sleep properly in faster universes as a fleeter. But at least it'll stop people from MD'ing miners for 20-30b profit.