Jump to content

Yekaterina

Community Members
  • Posts

    2.919
  • Joined

  • Last visited

  • Days Won

    48

Everything posted by Yekaterina

  1. Yes, but only in early game. The small houses cost only 75 wood, which means for 300 starting resources you can build a farmstead, a storehouse and a house all at the same time -> good efficiency. However, in late game, the small houses become an absolute pain because their building them is less efficient at creating population space than large houses. They also take up more land and hence you might get yourself in an awkward housed situation. You either micro more or you have to set more house builders (they have to walk more). It's often the case that you boom faster and gather faster than you can build houses. @MarcAurel Good idea. But this tech should have a very short research time. Gauls and brits desperately need this. Building houses at the right place at the right time is the hidden key to achieving a good boom. It's often overlooked.
  2. Now let's test a mainland medium temperate 4v4: Using Vanilla A27: OpenGL: ~70, fluctuates between 40 and 80. Vulkan: about the same as OpenGL, but the 1% low is 60-70. No dropping below 50. OpenGL ARB: ~90-100 significantly better than the other two backends. Using abstractGUI: Vulkan: ~200 OpenGL ARB ~210, but much more stable than Vulkan. OpenGL ~190, but very unstable compared to the other two. Conclusion: in the vanilla case, the performance differentces are quite visible, but not so much for AbstractGUI. AbstractGUI doubles the frame rate on Mainland temperate medium 4v4. The biggest cause was the trees - ondulating canopy and complicated leaf details consumed much processing power. The cubes of abstractGUI has only 6 polygons to render so it's much lighter on the GPU and CPU. There are many trees on Mainland, so it makes a huge difference. The trees are also a difficult task for the GPU that enables us to compare the performance of each renderer against each other, but in real gaming, it's best to have simple models.
  3. These were the settings. I checked that all settings were kept the same for each test. It was consistent (at least for me) that OpenGL ARB was the best and normal OpenGL was the worst. Vulkan is close to OpenGL ARB but is obviously inferior.
  4. The more archers in a crowd the more deadly, because they become simple unapproachable. I saw a player called Kafkas who made 180 archers with the Persians, and nothing else. It took two 1600 players to hold against the crowd. Eventually a third player came and they were able to crush him 3v1. Melee units die before they can reach the crowd, and at that number, you lose very little so it's easy to reinforce with more. Only Persians and Mauryas can do this, Kush and Carthage have the low populationn limit so they can't play the strategy. I still think Carthaginian archers deserve a range tech, because they do have the perfect setup for archer play.
  5. The issue with both of these civs is that they can't play archer spam, even though they are archer civs. You are better off doing cavalry or mercenary, especially in the case of Carthage who only has one type of citizen melee unit. Archer civs really need additional numbers or some alternative citizen ranged unit. The biggest difference between archer and skirmisher is that 150 skirmishers is not significantly better than 120 skirmishers in battle, because the short range means many of them spend time pathfinding around each other before reaching the front line. However, 150 archers can just stand still in one crowd and continuously shower the enemies with arrows in all directions without stop. There is little pathfinding around needed.
  6. Mauryas and Persians have the best archer spam, because of the additional pop space. Archer spam is all about numbers, so the additional 20 units is extremely important. If you want Kushites to have the best archer spam, give them additional pop allowance. Difficult question. We can make them have access to different kinds of melees and other units. I think the archer civs are already quite different from each other.
  7. After activating my AbstractGUI mod bundle, the frame rates were quite even for all rendering processes. The testing platform was Arch Linux 6.3.2-zen1-1-zen, 12th Gen i7 with Alder-Lake P integrated graphics. Vulkan: averaged around 260 to 270, but occasionally reached 303 but dropped as low as 160. OpenGL ARB: averaged around 270-280, with fluctuations between 250 and 330 OpenGL: averaged around 220-230 but fluctuations were large. An additional observation: killing Google Chrome in the background really helped with frame rates. Just closing 2 tabs increased my frame rate by 30. Conclusions: OpenGL ARB > Vulkan >> OpenGL. Even though it's ugly, AbstractGUI really reduces lag and increases your frame rate, especially for players with weak GPUs.
  8. Testing Vullkan vs vanilla A26: I hosted a multiplayer game in Acropolis Bay and stared at a Han CC, doing nothing. On vanilla A26, the average frame rate was around 120fps but with huge fluctuations, from 70 to 150 On A27 with Vulkan, the average frame rate was 130 but with smaller fluctions between 110 and 160. The frame rate did not drop below 100, unlike A26. On A27 with OpenGL, the average frame rate was 100 but the values always fluctuate from 80 to 120. On A27 with OpenGL ARB, the average frame rate was 220. In my particular case, OpenGL ARB >> Vulkan > OpenGL. I am using intel integrated graphics.
  9. @Boudica You might want to see this: the ProGUI mod that is under heavy scrutiny You can make mods undetectable by deleting their mod names in mod.json. But HerkEule with BoonGUI was able to see through this. You need mods to defeat other mods...
  10. Camels not so effective against Kush, as the Kush arches do have ranged upgrade and all sorts of other bonuses. Camels don't. So camels will be outranged and lose to the Kush infantry (assuming some melee is involved)
  11. wow, you have been posting on this forum since before I was born! OP! Greetings from a new generation of 0AD players! (A26 at the moment)
  12. Sun Wukong wants A23 Brits and Gauls back, when these civs' buildings added some pop cap. This feature caused great balancing issues so it was removed since A24. On one hand, it makes booming easier. On the other hand, the implication of people sleeping in barns and barracks is not so nice. So if you want to implement this feature again, do it for every civ for balance.
  13. Ptolemies. They can beat the Kushites in 0AD, because of slinger + pike combinations.
  14. Some people think that Carthage is a weak civ in A26. If you play them with the correct strategy, they are not bad at all. These two replays might be helpful: carthaginians.zip Loadouts: 90 archers + 40 spearman + 10 cav Jav cav spam + spearman or champion cav Elephants + archers 40 archers + 40 cav + 50 spearman It's also a good idea to make a bit of everything with the Carthaginians. They are the most diverse civ so they do have the best potential to play any strategy you want. They are totally flexible and the mercenary units can save you in dire situations. The key to playing Carthaginians is always keep about 9 units on metal, as there are many opportunities to spend metal in late game. Archer spam with the Carthaginians is quite powerful. Try to get your enemy in front of some building or some obstacle then shower then with arrows. If they approach you, retreat immediately. The Carthaginian spearman has identical shape to a hoplite so you can do the trick of shrinking 50 spearman into one tiny unbeatable singularity, using formations. Enemy pikes are countered by your elephants and hero dancing, meanwhile your archers snipe enemy ranged. One thing that can be improved is to give Carthaginian archers the additional range tech. They are an archer civ after all so they do deserve this tech like any other archer civ.
  15. I'm not. You are misinterpreting. Firstly such a mod doesn't exist yet. If you want and CAN write it, be my guest. Secondly if you are paranoid about mods, then just don't play. I was not talking to you. As you can see it was separate post. I said this because Atrik's progui has been under much scrutiny. As someone who has tested the mod, played against users of the mod, I believe that it's not cheating. You probably haven't even heard about it yet because you were offline for too long.
  16. I think star diagrams with criteria for each player are more useful than just a single number. Some people can be really good at one thing and weak at others. Such diagrams will reflect people's strengths, weaknesses and where to improve on: http://3.bp.blogspot.com/_9uFou63ufEE/TTdLVWUSQxI/AAAAAAAAAMU/sTqtb9N67cE/s1600/star-diagram4.jpg Some criteria that we can put in there are: Average KD Average kills per game Average population count at 11:00 Average resource gathering rate (count/s) For each player's profile, we can add time of account creation total number of units killed total games played of any kind
  17. we can start from everyone being 1200, then climb up all again. That would be fair. Although one should include 2v2 4v4 3v3 in because most games take this format.
  18. @Grapjas look at yourself: you just joined the thread and started giving random, irrelevant remarks, without offering any helpful information regarding vinme and felixix. We we all talking about what happened between vinme and felixix. So who is off-topic here? You are the one who first steered the conversation away from vinme to completely irrelevant things. So it seems like you are the one trying to attract attention.
  19. You are the one steering the conversation. We were talking about vinme and felixix but you suddenly joined and accused me, rm -rf, snake of needing mental help. Then you steered the conversation off-topic. I know who I am but I can change the way I behave and talk as I want to ?
  20. You can now change the GUI scaling in the settings. Set it to x2.00 and it should be fine.
  21. yes. I was one of the last 10 posters, so I am included the range of people you are talking to. what is obvious? Yes K is me. Manipulating what? Be clear. You just randomly asserted that the 3 of us need professional mental help, as if we have mental health issues, in a thread that is completely irrelevant to you. So you probably missed some context. Maybe you are the one who needs some help, since you also posted here. You don't seem to be aware of who and what has been taking place in the lobby. My recommendation is you come to the lobby and play some TGs more often. I watched this conversation between vinme and felixix take place yesterday. If you were there, you would have understood better why people are saying seemingly strange things.
  22. Who are you talking to? I see 4 different people talking, you, me , rm-rf, snake. I think all of us are off-topic now. Unless vinme has anything more he wants to say, this thread can close.
  23. All things considered, rm -rf is the most OP here terminates the thread with one sentence one of the strongest players in the lobby gave me suggestions on how to improve my Python skills thanks man!
  24. I made videos teaching people to use xmpp to log into the lobby on their phone so that they can monitor the lobby all the time. I think there isn't anything wrong with that. I have been able to talk to friends in different time zones using android xmpp clients. If some misuse it, that's another story. But even xmpp users are under heavy monitoring by the moderators and changing name results in instant kick. So I can't see why you mentioned Xmpp Open source software dominate the world. Microsoft, Apple and Facebook die At this point, one dude behind berhudar is truly the best. Borg Vali feld and the other felixix is purely up to their luck and biological condition
×
×
  • Create New...