My first thoughts about EB

[QwertyMIDX]

You are aware that there is an element of randomness in the calculations that RTW uses right? In the exact same situation different things will happen if you play it over and over again. I've broken enemy armies with cavalry charges to the flanks more times than I can count, but there have been a couple of times when the charge just wasn't that effective. Of course the way we had to compensate for the broken charge in 1.2 exacerbates this, and hopefully cavalry will be a bit more well balanced in 1.5 when we actually have a charge stat to work with. But really, would you want or expect the same thing to happen every time?

As for missile units, firing in the rear or the right flank (shield-less) is certainly quite effective. I just tested it again and my greek slingers can kill 6 or so TH’s a volley. Either way though the port to 1.5 will require rather serious missile rebalancing to deal with the bug/feature in 1.5 with archers/slingers and heavy infantry, so you may get a bit of your wish.

[fallen851]

You are aware that there is an element of randomness in the calculations that RTW uses right? In the exact same situation different things will happen if you play it over and over again.

This is not intended to change anyone on the EB team's opinion. It won't no matter how I phrase it because my opinion holds no weight, partly because of how I have treated the EB team before, and partly because of the way they view me. So this is intended for anyone else who cares.

QM and Ludens you're wrong. That isn't a personal attack, it is a fact. There is little randomness (at least in 1.5). I did many, many tests and the same thing happens, every time.

For instance, due to the fact I've decided to rebalanced EB when it comes out for 1.5, I've done a lot of research, particularly into the charge bonus, and why armor acts differently vs missiles some of which can be found here https://forums.totalwar.org/vb/showthread.php?t=62744

And what I found was that the same thing happened over and over again. For instance I've since done about 10 more runs, but these 6 runs showing a unit of 100 gauls in a warband with 5 armor factor (0 in the defensive skill and shield factor) standing still and take 5 shots from a chosen archer warband with an attack of 20 is not signficantly different:

Computer Controlled Archers:(units stood still)
Run #1: 81 Gauls remaining after 5 volleys
Run #2: 78 Gauls remaining after 5 volleys
Run #3: 78 Gauls remaining after 5 volleys
Run #4: 80 Gauls remaining after 5 volleys
Run #5: 82 Gauls remaining after 5 volleys
Run #6: 80 Gauls remaining after 5 volleys
Average: 79.833 repeating

When I ran an ANOVA test, there was no signifcant differences (though I knew there would not be before I ran the test...). Sure there is obviously some randomness, because during one test they lost 18, while another 22, but I'm assuming these small differences are not what you speak of.

The results for the charge bonus aren't up there yet, but my testing there (where I setup my unit in a set spot and let the AI cavalry charge) shows exactly the same thing, no significant difference between what happens in the 15 tests I've done charging from the rear. The unit loses about the same about of men every time, and another ANOVA test shows no significant difference between tests. Please note some of those tests are not valid (the ones comparing armor types in melee) because I had the warcry on the units before I retested, and warcry appears to lose its effects after a certain period of time, so the side what warcries first does worse in battle.

I invite you to test what I did. If it comes out differently, send me the replays. The differences are in what happens in the battle, not some magic randomness.

[econ21]

Fallen851 the post you cite from QM talks about randomness in the result of cavalry charges, not archery. He is not wrong about that, since melee combat in EB is essentially a moral contest. Whether cavalry breaks an enemy with a flank attack will depend on a lot of morale modifiers such as command stars, relative casualties, fatigue, positional factors etc. These things will not be fully captured by a simple cold "one-on-one" custom battle experiment. The actual initial casualties are almost secondary - they are certainly much lower than I've experienced in Goth's all factions mod for BI[1] where a whole unit can die in 3 seconds in similar circumstances. In my experience with EB, either the defenders break quickly or it soon gets ugly for the cavalry.

[1]Goth's mod is excellent, BTW, although as you can tell the combat model differs substantially from EBs.

[QwertyMIDX]

I would say something, but I seem to have been beaten to the proverbial punch.

[fallen851]

Fallen851 the post you cite from QM talks about randomness in the result of cavalry charges, not archery. He is not wrong about that, since melee combat in EB is essentially a moral contest. Whether cavalry breaks an enemy with a flank attack will depend on a lot of morale modifiers such as command stars, relative casualties, fatigue, positional factors etc. These things will not be fully captured by a simple cold "one-on-one" custom battle experiment. The actual initial casualties are almost secondary - they are certainly much lower than I've experienced in Goth's all factions mod for BI[1] where a whole unit can die in 3 seconds in similar circumstances. In my experience with EB, either the defenders break quickly or it soon gets ugly for the cavalry.

[1]Goth's mod is excellent, BTW, although as you can tell the combat model differs substantially from EBs.

This is exactly what I mean with EB's response to criticism, it is like you guys don't even read or think about people's posts.

Whether cavalry breaks an enemy with a flank attack will depend on a lot of morale modifiers such as command stars, relative casualties, fatigue, positional factors etc. These things will not be fully captured by a simple cold "one-on-one" custom battle experiment

I hate to quote myself, but if you had read you would have seen:

The differences are in what happens in the battle, not some magic randomness

Everything of what you said, is not randomness, it is what happens on the battlefield. Morale, fatigue, positional factors and command stars are not random conditions. Your general's command stars do not randomly go up and down during battles do they? I dare you to answer that question, but I don't think you will.

I shouldn't even have repond to the first part about cavalry charges because it was in my last post. I did charge bonus tests, and guess what? There was no randomness. Every time barbarian cavalry hit a warband from the rear there was a certain number of losses. There was a little bit of variance, but not much.

I would really love to see some replays of all this randomness, it simply doesn't exist. If a battle goes a certain way, and everything happens the same the next time in terms of movements and decisions, the outcome will be the same. It is how replays. If you take a replay, and then go and change the stats of units, the replay will look totally different. That is because the replay repeats all the movements made, losses will differ slightly. Replays are not movies, they are repeat battles fought in the same way. Now take that replay and watch it over and over it will differ slightly, but the result will always be the same (unless it is so close that a couple of men makes the difference). If you watch a replay of a cavalry charge into a flank over and over, one time it won't kill 200 men, and the next time only 5 men, that randomness does not exist in the RTW engine, I'm sorry, but it doesn't.

As a final response, anticipating your next post, your not actually going to tell me that men dying has nothing to do with morale? Deaths effect morale, perhaps it is the most powerful factor? Thus if cavalry is only killing 5 men hitting the rear of a unit, it isn't doing as much damage to the morale as it did 25 kills.

[QwertyMIDX]

I'm going to go out a limb here, but somehow I don't think all of these people (many of whom are not EB members) are all in some giant conspiracy to thwart your rational and level-headed attempts to change something. You seem much less willing to hear people come out against your stances than anyone else involved in this, and when anyone doesn't agree with you accuse them of not listening. It's really coming to the point where it's just not worth saying anything to you, unless I being worshipping you as the center of all truth in the universe I don't think you'll be satisfied. You just yelled at econ21 (better know as Simon Appleton) for being a partisan of EB soley for not agreeing with you. This is just getting out of hand.

[Kull]

This is exactly what I mean with EB's response to criticism, it is like you guys don't even read or think about people's posts.

Being the huge stat freak that you are, perhaps this data will shed some light on the real problem:

1) Number of EB downloads: 29,000

2) Number of non-sticky threads in the EB Org Forum in the last 30 days: 103

3) Number of posts in those threads: 2027

4) Number of posts complaining that EB's "entire unit system is flawed": 4 (all by you)

Draw your own conclusions.

[econ21]

Everything of what you said, is not randomness, it is what happens on the battlefield. Morale, fatigue, positional factors and command stars are not random conditions. Your general's command stars do not randomly go up and down during battles do they? I dare you to answer that question, but I don't think you will.

Fine, I'll concede your semantic point. It is not randomness. But it seems like that to the casual player (e.g. me) because we cannot directly observe the quantitative strength of these battlefield conditions. Sometimes a retreating phalanx charged in the rear will fold quickly, sometimes it will stand, turn and maul the attackers. Through casual play, I cannot directly observe the factors that account for the variation - hence my sloppy use of the word randomness.

But I am not sure where this pedantry takes us. The thread originally was about whether archery and cavalry kills too few people in EB. Honest people are going to disagree about that. I'd be interested in any test data you produce based on EBs EDU. But whatever the numbers are, since we lack corresponding real world data from ancient battles, people will still differ over whether the results are realistic. From casual play I am quite happy with EBs balance. I don't expect missiles or cavalry to kill many heavy infantry (except cav vs routers). But they do contribute to victory through other means. The missiles harass and pressure the enemy, providing the slight edge that can tip a tough fight. The cavalry frequently has decisive effects via morale and their superior mobility (although it doesn't always work - hence QMs and my inexact use of the term "randomness"). That just feels right to me, but I know others such as yourself may disagree.

However, what you seem to want to do with the thread is use it to attack the way the EB team responds to criticism. Given that I am not affiliated with the mod and am just an interested passerby, you'll have to exclude my semantic error from your indictment of the team. However, in my experience QM and others in the team have been friendly, informative and reasoned in dealing with questions and comments. I know from personal experience, some other mod teams are nowhere near so tolerant or mature.

[QwertyMIDX]

Thanks for the support Econ, I really do try and be reasonable.

[[TAG]ImperialMarch]

not to jump into the flame war or anything... but...
am i the only one that noticed that he's testing using 1.5, and EB is 1.2? I thought that made a fairly significant difference in the cav charges, and somewhat to the missle stuff...

[Ludens]

QM and Ludens you're wrong. That isn't a personal attack, it is a fact.

Your accusation that the EB team is unable to rationally discuss their system is a personal attack.

The EB team has clear ideas of how thing should work, so they are not going to be swayed by a few observations that something is wrong. Only when you can prove that a situation regularly gives illogical results then they might change their mind. For example after repeated complaints that peltasts were overpowered compared to hastati and thurephoroi Qwerty reexamined their stats. I don't know if they have been corrected in 0.74 but he did announce they are going to be weakened. However, AFAIK no-one ever did systematic archery tests to prove they were indeed underpowered. If you did I, for one, would be happy to see them. Unfortunatly, it is very hard to balance cavalry charges in 1.2; and this is a rather futile excersize anyway since the mod is going to move to 1.5 in the next build.

Lastly, if I interpret Qwerty and econ21 correctly, they did not say that under similar circumstances unit performance can differ significantly (which is obviously wrong) but that in a campaign differing circumstances can have a very pronounced effect on performance.

[fallen851]

I'm not the only one to complain about the EDU file, it is a common complaint. How many threads have we gone over this?

I do not think there is a conspiracy against me to thwart my great ideas. I don't care if you like what I think. What upsets me is how you guys respond to criticism. It is not consistent, it is done on an ad hoc basis (see above quote by Telesos and The Wizard). Your vague, you change the subject, and claim authority (we've done the testing so we know what happens and you don't).

So what did I do? I went out and did the tests, did the research, and you're previous statesment are incorrect. But you guys never give up. QM has claimed in the past there is randomness in how the shield armor is applied (Macroi's thread in which he stated talking to EB memebers was like "pulling teeth"... I'm not the only one). I must have done at least 20 tests with the shield armor, and I found it is equal in melee combat to both the normal armor factor and the defensive skill factor, and 7/8ths as effective against missiles. That is not random. Is it weird? Sure, but if you do the tests, you figure out what is going on. Later in the conversations (ad hoc) QM stated he hadn't done tests in awhile, and he wouldn't have the time to do them, so the shield stats stayed the way they were, flawed. Why are they flawed? Because they are based on the untrue premise that the shield application has randomness.

Oddly enough, those who have criticized EB have began very cordial, but when the EB simply argues with the methods I outlined above, people become irrate. Then of course the EB team lays on the "personal attack!" key they have must have on their keyboard. Either that is happening (what I described above), or (as you might argue) people are coming on here and demanding things be changed, and when you prove them incorrect, they resort to personal attacks.

I don't care if you change things because I can change it myself, in fact I don't think many people care that much, they come on here to contribute to making this mod more realistic, and when they are on the recieving end of the personal attack key, they simply change it themselves and leave.

Now I will do what you said Ludens, and do comprehensive testing of the archery system, but let me tell you right now what I'm going to get for an answer "the archer stats are fine, they shouldn't be powerful". This is purely a judgement call. You may think it is realistic, you may think cavalry charging the rear of a unit and killing 5 men is realistic, I don't know what you think, but as long as you have that arguement, you guys have proven a willingness to use it. And you will always have that arguement, so all the criticism raised against you, by many others and myself, falls on deaf ears.

I personally believe many of the people who have criticized EB to the extent I have learned this lesson long before I did to simply give up. To them it simply isn't worth the time, and we have gone through thread after thread, and most them aren't around anymore. Perhaps I was harsh when I said people were "zealots" of EB, but it seems they come into a thread and then start ranting about "Your not using the units correctly!" completing disregarding the idea the unit could be flawed!

If your not willing to see a problem, you won't see it.

You guys don't want to see the problems in the EDU text, and even more importantly, you want to think you guys are very cordial when dealing with criticism. Well, then where are all the critics?

[QwertyMIDX]

I actually didn't say that shield points had any randomness I said "It's not a simple point for point relationship." Which apparently you now agree with, as your tests seem to have proved it doesn't work the exact same way as armor.

I must have done at least 20 tests with the shield armor, and I found it is equal in melee combat to both the normal armor factor and the defensive skill factor, and 7/8ths as effective against missiles.

Please try not to misrepresent me while you attack me.

[Ludens]

:Sigh: fallen851, you have adressed none of the arguments I brought against your postition, instead arguing again that cavalry charge is not correct, even though I already explained why it is that way. You are actually acting like you accuse the EB-team to act.

[fallen851]

you have adressed none of the arguments I brought against your postition

Now I will do what you said Ludens, and do comprehensive testing of the archery system

Did you read that part?

Now what are your other arguements? I addressed your first sentence, essentially stating that many critics of EB have given up because the EB team doesn't argue in a rational way. I can point to several people who no longer post on this forum that have criticized EB. I don't see many critics of EB anymore, which could mean one of two things #1 EB is perfect, #2 EB silences critics. You don't see a lot of critics of the government in North Korea do you? Those people gave up criticizing, because it wasn't worth it.

Did you read those parts?

QM, you're correct...

(I have no problem admitting I'm wrong, I've done it on several occasions with you, but in the past you did claim and I quote "'Your' overhand spear argument wasn't yours" to which I responded "Ugh. I never said the overhand arguement was mine. I never even used 'mine'. Read it this time." to which you simply ignored in your next post. When you're wrong, you sweep it under the rug like nothing happened. I have other post of you been shown you were wrong and not admitting it, which of course is a misrepresentation. So I guess it is ok for you to misrepesent?)


...you did say it wasn't a simple point for point relationship and you didn't say it was randomness. I shouldn't have put it that way, and should have antcipated your response. However you said it was something quirky about the way the shield values are calculated (also correct), leading me to believe you didn't understand the system clearly. So did you base armor values on a system you didn't understand clearly? Or if you did understand it clearly why didn't you tell people that then and end the discussion?

"I have not had time to re-run dozens of tests to demonstrate this, and I won't anytime soon, feel free to run your own if you'd like."

Did you know the shield provides 7/8th's the armor against missile than normal armor? My whole problem with you during that thread is that you wouldn't tell me why the shield is less effective beyond the AP, left, and rear issues, which is why I ran my own tests. The best thing about my case on this topic is I was right then about EB withholding information, or I am right now about the EDU system being flawed.

So if you didn't know the exact relationship the problem remains, if you didn't fully understand the system, the system is flawed.

None of that really matters though, that arguement was to point out I am not alone in criticizing the EDU text.

[Ludens]

Did you read that part?

I did. You did not, however, adress my argument cavalry charged is bugged, nor any of the points I raised earlier in this thread.

[Teleklos Archelaou]

I can point to several people who no longer post on this forum that have criticized EB. I don't see many critics of EB anymore, which could mean one of two things #1 EB is perfect, #2 EB silences critics. You don't see a lot of critics of the government in North Korea do you?

Hmmm. Now that you mention it...