|
Post by Kerr on Nov 4, 2017 18:14:50 GMT
I'm 90% sure the game runs using Unity, if that counts. Not sure which version though. Really? Should there be some unity files in CDE folder then? Or are they in some kind of sub folder or not there entirely?
|
|
|
Post by AdmiralObvious on Nov 4, 2017 18:18:42 GMT
I'm 90% sure the game runs using Unity, if that counts. Not sure which version though. Really? Should there be some unity files in CDE folder then? Or are they in some kind of sub folder or not there entirely? When I go through the game directory, it looks very similar in structure to a Unity designed game. Also, when you get a specific crash, it gets a C+ Runtime error, which might lend to it being a Unity game.
|
|
|
Post by Kerr on Nov 4, 2017 18:25:37 GMT
Really? Should there be some unity files in CDE folder then? Or are they in some kind of sub folder or not there entirely? When I go through the game directory, it looks very similar in structure to a Unity designed game. Also, when you get a specific crash, it gets a C+ Runtime error, which might lend to it being a Unity game. Good point, the crashlog provides some insight into the engine.
|
|
|
Post by 𝕭𝖔𝖔𝖒𝖈𝖍𝖆𝖈𝖑𝖊 on Nov 4, 2017 19:56:18 GMT
I second this... qswitched should read this I doubt that it would help, the engine seems very limited. It would require the entire game to be made from scratch. not really. It would require minor changes and possible a new game mode, For example a Total war game mode where you can spawn ships with bases on your planet and if you attack a planet with enough nukes, the bases will no longer operate and you will loose the material generation and the credit generation. It would be cool
|
|
|
Post by 𝕭𝖔𝖔𝖒𝖈𝖍𝖆𝖈𝖑𝖊 on Nov 4, 2017 19:59:47 GMT
dichebach Less dense fuels require more mass/cost for armoring them; which is why most players use methane or the higher alkanes when building armored vessels. The more armor you slap on, the more attractive the denser fuels become. So, my chemistry is pretty remedial. Is Decane (which always seems to require Lox for a fully functioning system) pretty much the heaviest in the game? ADDIT: Also, "decane" a hydrocarbon. Hydrocarbons come from geological deposits "where decomposed organic matter provides an abundance of carbon and hydrogen." Are these things being artificially fabricated from hydrogen and carbon sources off Earth? Is that even energetically viable? I know there are synthetic oil processes which can be used in desperate circumstances (e.g., Germany in WWII) but even those require coal. Its not energy efficient to convert raw carbon and hydrogen into hydrocarbons, but it is the best way to do it in space, as you are converting electric energy to chemical potential energy with a bit of waste heat, and if you're using a nuclear reactor you can do this practically forever.
|
|
|
Post by Rocket Witch on Nov 4, 2017 23:18:50 GMT
I'm 90% sure the game runs using Unity, if that counts. Not sure which version though. I had guessed that before too, but CDE apparently runs on its own engine. I guess it could've been derived from Unity and modified beyond recognition though, you never know. See: childrenofadeadearth.com/FAQs.html
|
|
|
Post by dichebach on Nov 5, 2017 1:48:34 GMT
Finally, after a ridiculous amount of fiddling have a ship design that does fairly well in Vesta Overkill. It is almost like I'm playing another game than what you guys (and many Youtubers) seem to be referring to in their descriptions of that scenario. I've tried at least a half-dozen different ship designs, and before this one, they were all slaughtered, as in: nothing left after a few missile/drone fleets hit. This one still is not skating all the way through yet. The missile and drone fleets generally seem to cause minimal harm. But in the handful (3?) where I've endured the lag and stutter long enough to get to the final moments of the scenario, the fleet carrier always seems to outlast my last ship. This has got to be one of the fatuous hard game scenarios I have ever encountered. The game was not overly difficult up to the point of Vesta Overkill, but in all honesty, sorting out what works and what doesn't (much less why, which remains partly a mystery based on the variety of solutions I saw suggested by various others and which did not work for squat for me) was about equal parts pain and fun. Had I tried to do this without unlocking the module designer, downloading a handful of modules and studying how those were put together then fiddling with alternative designs, and reading up about the stuff, i.e., how the game assumes these things will work, I very likely would have just stopped playing as it seems many users must have. This is not meant as an attack on anyone, but merely as a factual description of my experience as a user. The beauty is, it doesn't have to be that way and the game has so much magic and wonder in it, it would be a literal travesty to NOT revise the game design to make this an addictively fun game. I really hope it happens. ADDIT: I also feel the need to say: I've played a lot of computer games over the years. I've been military board games, and roleplaying games since 1983, and computer games for nearly that long. I am pretty good at estimating what is going on with the algorithms that comprise my computer opponents under the hood, and I tend to suspect that Admiral Voitenko is a big fat cheater! There is as yet, no such thing as "AI" and certainly not in the emergent decision-making patterns which can arise from the algorithms in a little game app for PC. With the exception of specialty opponents like chess opponents or other relatively simple board games, MOST so-called "AI" in computer games cannot possibly pose a credible competitive threat to a typical human player if they meet on a level playing field. In short, I understand that "cheating" computer opponents are in fact, the norm and I have been party to them myself. But there is blatant "this don't seem like we are being bound by the same rule set" cheating and then there is "Hmmm, how did he do that so fast/easy/direct?" Vesta Overkill falls into the former, more blatant category and I can imagine that many would be fans did not take well to that experience.
|
|
|
Post by AdmiralObvious on Nov 5, 2017 2:51:47 GMT
But there is blatant "this don't seem like we are being bound by the same rule set" cheating and then there is "Hmmm, how did he do that so fast/easy/direct?" Vesta Overkill falls into the former, more blatant category and I can imagine that many would be fans did not take well to that experience. I'm not sure specifically what you mean? As far as I can discern, the AI won't cheat. However, it's exceptionally good at maintaining the "perfect" intercept for basically everything, as it is the PC, and automatically calculates. Care to explain specifically what makes you believe this? P.S. That ship can't fire all weapon mounts at once, even 3 is pushing it. That's probably partially part of the problem. Try adding a few drones/small nukes.
|
|
|
Post by treptoplax on Nov 5, 2017 3:45:35 GMT
But there is blatant "this don't seem like we are being bound by the same rule set" cheating and then there is "Hmmm, how did he do that so fast/easy/direct?" Vesta Overkill falls into the former, more blatant category and I can imagine that many would be fans did not take well to that experience. I'm not sure specifically what you mean? As far as I can discern, the AI won't cheat. However, it's exceptionally good at maintaining the "perfect" intercept for basically everything, as it is the PC, and automatically calculates. Care to explain specifically what makes you believe this? P.S. That ship can't fire all weapon mounts at once, even 3 is pushing it. That's probably partially part of the problem. Try adding a few drones/small nukes. Remember that Vesta is just an asteroid; the gravity is quite low, so the fleets can easily fly straight at each other in a way they can't around Venus or even Luna. And the AI has something like 2-3X the funds you have, so such a direct fight is problematic. I do wish you could use stock-component drones/missiles without unlocking module design - you can build them, but not a launcher, which is a bit silly. It's possible to improve hugely on the Stinger even with stock modules and such a drone would be invaluable in this scenario. To elaborate on Admiral Obvious' suggestions: Each laser can only route through one laser mount at a time. Usually you just have 3-4 per laser, at most, to cover multiple facings. Personally I find the 13Mw is the best of the stock lasers; maybe use multiple of those instead, especially as you're power-limited in this design anyway. I was just testing this again myself and found that 3 of the stock cutters (so 9 13Mw green lasers in all) can destroy all the incoming drones/missiles pretty effectively if you specifically target the guns. The lasers probably won't do to kill the capital ships, though. Broadside-radiators-behind is a pretty standard design type. Remember that damaged weapons will leave holes in armor; you might want to mount them off center (if they have enough angle) or over spacers (so nothing is behind them). For a more extreme variant, put some or all armor layers only on the frontal side (see the stock Sentinel for an example of this).
|
|
|
Post by treptoplax on Nov 5, 2017 14:42:20 GMT
Also, I think some of the AI fixes made Vesta Overkill even harder since most of the people here beat it. Although I think some of the stock missile launchers have issues now; occasionally the Siloship seems to be doing heavy maneuvering while launching and collides with it's own missiles, so you may have that going for you...
There's another thread active with a YouTube video of a VO win with custom modules and an insanely cheap ship - it has no armor or engines (!), just a bunch of really cheap gundrones. Shows good interception techniques...
|
|
|
Post by qswitched on Nov 5, 2017 19:05:41 GMT
ADDIT: I also feel the need to say: I've played a lot of computer games over the years. I've been military board games, and roleplaying games since 1983, and computer games for nearly that long. I am pretty good at estimating what is going on with the algorithms that comprise my computer opponents under the hood, and I tend to suspect that Admiral Voitenko is a big fat cheater! There is as yet, no such thing as "AI" and certainly not in the emergent decision-making patterns which can arise from the algorithms in a little game app for PC. With the exception of specialty opponents like chess opponents or other relatively simple board games, MOST so-called "AI" in computer games cannot possibly pose a credible competitive threat to a typical human player if they meet on a level playing field. In short, I understand that "cheating" computer opponents are in fact, the norm and I have been party to them myself. But there is blatant "this don't seem like we are being bound by the same rule set" cheating and then there is "Hmmm, how did he do that so fast/easy/direct?" Vesta Overkill falls into the former, more blatant category and I can imagine that many would be fans did not take well to that experience. The AI in Children of a Dead Earth is bound by the exact same rule set as the player. An AI that doesn't cheat was a major rule I followed when developing the game. The only unfair aspect is that the AI always starts with more units than you.
|
|
|
Post by dichebach on Nov 5, 2017 22:10:13 GMT
ADDIT: I also feel the need to say: I've played a lot of computer games over the years. I've been military board games, and roleplaying games since 1983, and computer games for nearly that long. I am pretty good at estimating what is going on with the algorithms that comprise my computer opponents under the hood, and I tend to suspect that Admiral Voitenko is a big fat cheater! There is as yet, no such thing as "AI" and certainly not in the emergent decision-making patterns which can arise from the algorithms in a little game app for PC. With the exception of specialty opponents like chess opponents or other relatively simple board games, MOST so-called "AI" in computer games cannot possibly pose a credible competitive threat to a typical human player if they meet on a level playing field. In short, I understand that "cheating" computer opponents are in fact, the norm and I have been party to them myself. But there is blatant "this don't seem like we are being bound by the same rule set" cheating and then there is "Hmmm, how did he do that so fast/easy/direct?" Vesta Overkill falls into the former, more blatant category and I can imagine that many would be fans did not take well to that experience. The AI in Children of a Dead Earth is bound by the exact same rule set as the player. An AI that doesn't cheat was a major rule I followed when developing the game. The only unfair aspect is that the AI always starts with more units than you. Well that is edifying! I was considering recreating the scenario to see if it has some "tweaked" math under the hood, but I"ll take your word for it that a copy of the scenario would perform (statistically) identically to the one in the campaign. The one game I've been involved in design work on: the computer opponent HAD TO CHEAT. There was simply no other way to make the singleplayer mode challenging. In fact, there are effectively two different sets of equations for most math-bound processes in that game (a grand strategy title whose name I will not reveal, because the fact that the "AI Cheats" is a big "secret"): one for the player (whether playing "Blue" or "Red") and one for the computer (whether playing Blue or Red). Which is to say: there is no shame in making a game in which the computer opponent "cheats." There is also no shame in "lying" to players about that! But I do believe you! My observations are probably too casual. Apologies if anything I have said or say is irritating Qswitch. I really only mean to help if that is even possible
|
|
|
Post by dichebach on Nov 5, 2017 22:28:08 GMT
I'm not sure specifically what you mean? As far as I can discern, the AI won't cheat. However, it's exceptionally good at maintaining the "perfect" intercept for basically everything, as it is the PC, and automatically calculates. Care to explain specifically what makes you believe this? P.S. That ship can't fire all weapon mounts at once, even 3 is pushing it. That's probably partially part of the problem. Try adding a few drones/small nukes. Remember that Vesta is just an asteroid; the gravity is quite low, so the fleets can easily fly straight at each other in a way they can't around Venus or even Luna. And the AI has something like 2-3X the funds you have, so such a direct fight is problematic. I do wish you could use stock-component drones/missiles without unlocking module design - you can build them, but not a launcher, which is a bit silly. It's possible to improve hugely on the Stinger even with stock modules and such a drone would be invaluable in this scenario. To elaborate on Admiral Obvious' suggestions: Each laser can only route through one laser mount at a time. Usually you just have 3-4 per laser, at most, to cover multiple facings. Personally I find the 13Mw is the best of the stock lasers; maybe use multiple of those instead, especially as you're power-limited in this design anyway. I was just testing this again myself and found that 3 of the stock cutters (so 9 13Mw green lasers in all) can destroy all the incoming drones/missiles pretty effectively if you specifically target the guns. The lasers probably won't do to kill the capital ships, though. Broadside-radiators-behind is a pretty standard design type. Remember that damaged weapons will leave holes in armor; you might want to mount them off center (if they have enough angle) or over spacers (so nothing is behind them). For a more extreme variant, put some or all armor layers only on the frontal side (see the stock Sentinel for an example of this). Ah! Thank you for confirming my suspicions about the lasers! I was wondering about this and must have skipped over the suggested reading part that explained that. Right now, that design actually has an energy glut. I'm pretty certain I could squeeze in at least one more 100mW laser and possibly two. Some tweaking on the power plant (or addition of one or two small ones) would allow for more dedicated "Laser + Mount" pairs. I had heard of the "target the guns" technique, but I'm curious: why would that matter? Given that gun operation is supposed to be "automatic?" I know that ticking "ignore range" results in a great deal more fire being lain down, but it is not clear if this has much real benefit. I had also encountered various commentators say similar things about lasers "not doing to kill capital ships" and I don't quite follow why that is the case. Are the 9mm rail guns likely insufficient for killing the cap ships? In my experience, missiles (my missiles) and drones are extremely ineffective. I designed a very high explosive (like 9mt) nuke and a matching flak missile (same size and mass) with the intent to launch salvos of 30 (27 flaks and 3 nukes). Probably way too big. In the course of two or three test plays with Vesta Overkill, multiple waves had minimal effect. The other thing with missiles is: the "safest" intercept trajectories seem to be fairly curved, even around Vesta. Because the computer sends his missiles at me headlong (using relatively more dV) they are relatively easy to sucker off into dead end trajectories. But I've found that, when I have made extensive use of Missiles, the computer opponent ALSO is quite savvy about suckering my more cost-effective (and thus slower) approach missile fleets and the outcome of the scenario ultimately depends on the attrition effects of his more numerous waves of drones and missiles whittling me down and then his largely unscathed capital fleet blasting me to smithereens. The broadsider ship I posted up above handles his missiles and drones almost without suffering harm. Killing most of his cap ships is also quite typical for that fleet of 10 to 13 broadside gunboats. But his last ship so far has always outlived my last ship. Some great ideas on weapon mounting! Until this design, I had been trying the "nose forward" and let all weapons fire, but in my experience that just does not work.
|
|
|
Post by treptoplax on Nov 6, 2017 3:24:35 GMT
Very briefly (these subjects could support their own thread easily):
Lasers are very accurate, and more to the point, dodging has no meaningful effect, so you can hit tiny targets at long distance. Most weapon mounts aren't protected by ship armor, and are difficult/annoying to armor for various reasons, so laser-sniping enemy weapons is often a fast way to cripple/disable.
However, stock/core lasers will take a while to burn through serious armor (if you're deploying the Apophys Electrics designs, downloadable in another thread about here somewhere, that's a different matter).
The automated range estimation is pretty crude. There's rarely a reason not to go ahead and let lasers shoot, even if they're unlikely to hit much. Railguns often have plenty of ammo as well. It also doesn't take closing speed into account, so conventional cannon on rapidly closing craft often have far greater range than the automatic range estimate would calculate.
|
|
|
Post by AdmiralObvious on Nov 6, 2017 7:07:19 GMT
Right now, that design actually has an energy glut. I'm pretty certain I could squeeze in at least one more 100mW laser and possibly two. Some tweaking on the power plant (or addition of one or two small ones) would allow for more dedicated "Laser + Mount" pairs. I had heard of the "target the guns" technique, but I'm curious: why would that matter? Given that gun operation is supposed to be "automatic?" I know that ticking "ignore range" results in a great deal more fire being lain down, but it is not clear if this has much real benefit. I had also encountered various commentators say similar things about lasers "not doing to kill capital ships" and I don't quite follow why that is the case. Are the 9mm rail guns likely insufficient for killing the cap ships? In my experience, missiles (my missiles) and drones are extremely ineffective. I designed a very high explosive (like 9mt) nuke and a matching flak missile (same size and mass) with the intent to launch salvos of 30 (27 flaks and 3 nukes). Probably way too big. In the course of two or three test plays with Vesta Overkill, multiple waves had minimal effect. The other thing with missiles is: the "safest" intercept trajectories seem to be fairly curved, even around Vesta. Because the computer sends his missiles at me headlong (using relatively more dV) they are relatively easy to sucker off into dead end trajectories. But I've found that, when I have made extensive use of Missiles, the computer opponent ALSO is quite savvy about suckering my more cost-effective (and thus slower) approach missile fleets and the outcome of the scenario ultimately depends on the attrition effects of his more numerous waves of drones and missiles whittling me down and then his largely unscathed capital fleet blasting me to smithereens. The broadsider ship I posted up above handles his missiles and drones almost without suffering harm. Killing most of his cap ships is also quite typical for that fleet of 10 to 13 broadside gunboats. But his last ship so far has always outlived my last ship. Some great ideas on weapon mounting! Until this design, I had been trying the "nose forward" and let all weapons fire, but in my experience that just does not work. This reply might be difficult for me to write, since I'm quoting from my phone, however, in most cases, lasers alone are not sufficient to actually ablate the armor of a targeted capital ship, if it's got any type of reasonable anti laser armor in even modest quantity. An extremely thin layer of graphite, for example will negate all but the most powerful of lasers. Railguns on the other hand are often more than sufficient to deal with most types of armor, unless that particular ship is designed to overcome railgun fire, with either a lot of redundancy, or a stupid amount of armor to soak up enough impacts to survive. Lasing the guns on a drone, especially if armed with a conventional gun makes the ammo explode, unless you shenaniganed the ammo to be a seperate component from the gun itself. This will almost always disable the drone, and most conventional guns are made of solid steel types of alloy. They're tough to resist the stress of firing, but they will melt pretty easily. Is your nuke measured at 9mt, or 9Mt? Because it REALLY matters. I'm just going to assume capital Mega, instead of Micro. Regardless, nukes have to be relatively close to their target for them to do anything remotely useful, since shockwaves don't transmit much in space, your nuke is primarily reliant on heat and gamma radiation, first melts armor, second melts the DNA of the crew. You ideally want a high FLUENCY nuke (because of science reasons currently going over my head due to insomnia), as it is able to do more damage over distance, though I've not yet myself mastered the way to make a high fluency nuke, with high enough tonnage to be noteworthy. As for dodging missiles. I personally reccomend throwing singular nukes at the enemy missile fleets. Last I played, they were extremely efficient ways to remove enemy missile groups, if you make the nuke burst behind the missile fleet in question. As for my specific issue regarding power. It seems that you had too many guns to fire simultaneously. Your ship had insufficient power to fire any more than 3 of the 6 laser mounts you had at once. With the current setup, each laser and railgun would end up "taking turns" firing as power was available, which may work fine for stupidly high power, ponderous weapons, it won't work for most super high fire rate rails/lasers.
|
|