Hi All !
What is the reliability of the replays, in your experience ?
I've found that the replays don't work 100% as the battle was played out and in the replay there are some changes to unit behavior.
Has anyone else experienced this ?
Salute !
Printable View
Hi All !
What is the reliability of the replays, in your experience ?
I've found that the replays don't work 100% as the battle was played out and in the replay there are some changes to unit behavior.
Has anyone else experienced this ?
Salute !
Not only have I seen changes to unit behavior, I've sometimes gotten entirely different battles with no relation to the one I fought other than the units involved.
Yes - my understanding is that rather than save the entire battle as like a movie type thing, what it actually saves is the units and then the moves you make, like your orders. but sometimes I've seen it so confused that what was a close victory for me replays and becomes a loss...
No wonder I can confuse the computers troops, apparently even after watching me it still has no idea what I did, much less trying to counter it at the time on the field of battle :)
Now this is interesting. Does this not imply that if a human were to do the exact same thing, issue the same orders in the same conditions, that the outcome would also be different there? Does it not imply some bizarre randomisation of unknown factors? Strange.Quote:
Originally Posted by DesertEagle
Quote:
Originally Posted by Quickening
Note that I'm not 100% sure on the assertions I made, it's just my understanding based on things I've heard and more important (to me, but more anecdotal to everyone else) what I've seen in my own experiances.
I couldn't testify to any of this, other than really close outcomes can change. And to be sure, I haven't had a heroic victory turn into a crushing loss or anything. If the outcome changes it's changed by a tiny amount, just enough to effect the end result. Maybe a unit that barely didn't rout in the game, routs in the replay, or maybe the arrows do a few less casualties and the extra 10 dismounted knights are just enough to turn the tide. Small things like that.
And I really think it's down to animations. Now, I use fairly missle-heavy armies too, so that may add extra-randomness to mine that someone who focuses more on melee/cav wouldn't have.
A quick test if you want to see if it happens to you, do a battle with 1/2 stack retinue longbows and 1/2 DFK, vs whatever. Note your actual battle results as far as men killed by the bowmen. Less important the Then watch the reply a few times. You'll (hopefully) note a difference in casualties. Or at least I did, last time I bothered to try a replay, but that's been a long while. Maybe it's been fixed in the 1.1 patch. Once I saw the uselessness of it, I didn't bother with replays anymore. Also of note, I found the same results in RTW and even MTW2. Didn't think to look for it in Shogun but I believe that if you went back it probably would hold true there.
My guess is that a "perfect" video type replay where the EXACT same thing happened down to the individual man on the field, the replay files would have a very large footprint, so they pare down the info to a "do-over" of the battle where both forces are controlled by the AI, rather than a replay in the conventional sense.
That´s not new and, as a matter of fact, has ben the case for all of the Total War games. Someone from CA confirmed once that the replays record the units and the commands given, not the actual movements.
And since there is a certain chance element in the battles (think of the stats in a Pen-and-Paper way, in a way they represent dice) the results in a replay will alter slightly.
We have heard that before and where it is the case that a series of commands are what is recorded, I can definitely state that on both STW and MTW any replay deviation was slight, barely noticeable, not noticeable or not present.
I had loads of STW and MTW replays and they were perfect replays of the actual battles. The only problems found with those games (in my experience) was if you did not have the map the battle was fought on or if you ran a replay using the wrong stats (eg mod replay on vanilla stats)
With both RTW and M2TW I have seen anything from minor differences to HUGE differences.
As an example. I ran a 3v3 custom battle Turk v Egypt, experimenting with army set ups. My 2 Turk allies were quite quickly attacked by all 3 Egyptian armies (it was really laughable how fast they were routed) I was hurrying to assist but they were both gone in a matter of a minute or so. I got into formation and prepared to go down fighting, already firing on the army which had been opposite me. As I was cutting this army down, I watched the other two regrouping. There goes army number one, my lone HA cutting down routers as they fled. Army number two was at me now, my Ottoman inf on guard mode let fly with volley after volley; all out assault countered by my Alans and army number two begins to flee. My lone HA has more heads to collect. Army number three is now in all out attack and my Ottomans ditch their bows and enter the frey, Afghan javelins pile into Egyptian lines and my 2 Jannisary units keep shooting from the flanks.My Alan cav reforms and charges into the Egyptian left flank and army number three is rolled up and chased off the field.
I had over 1800 kills, my lone HA did not lose a man yet killed nearly three times their number. Needless to say I was happy with my Turkish army.
Then the replay.....
All ran correctly until the point where my army formed up ready for the onslaught. Then it was a total mess. Why are my cav charging straight through my lines? Why are my infantry lines charging into melee? Why is my army routing?
It was total pants! Heroic victory reduced to clear defeat!!
Ask me my opinion and I am convinced it is engine related and there's just too much wrong with it to expect any decent fix with a patch
........Orda
I would imagine a proper replay of a battle shouldn't be too tough to make for CA considering other companies' games do not have such troubles.