If they reinitialized the state on reload, then you could also just reload until your 20% shot hits too
Man, if you want to sit there and reload the game until you get a hit on a 1% shot I really don't care because how other people play the game doesn't impact me, but how the game calculates hits absolutely totally does.
And it's irritating to find out that 55% doesn't really mean what I think of when I see 55%.
0
Munkus BeaverYou don't have to attend every argument you are invited to.Philosophy: Stoicism. Politics: Democratic SocialistRegistered User, ClubPAregular
Oh I know. But, there might not actually be a sequence that lets you do what you want whereas with reinit there could easily be. Basically, if you want to cheat you can cheat. I would image they did it for multiplayer and testing, both cases where you need to control randomness
What actually matters is how hilarious it makes watching Beagle-Live.
And all his strategy and talking about removing the 'randomness' by trying to make sure his shots are a success.
Early Impossible is a total crapshoot with the RNG like it is, and going at it carefully at the start is a mistake. Ram your head until you get past that first start and then it becomes a game.
Humor can be dissected as a frog can, but dies in the process.
Oh I know. But, there might not actually be a sequence that lets you do what you want whereas with reinit there could easily be. Basically, if you want to cheat you can cheat. I would image they did it for multiplayer and testing, both cases where you need to control randomness
I don't see how this is really a thing that affects multiplayer. It's not like you can reload your shot in multiplayer.
And I feel like testing isn't a great excuse either, but that may be my own previous experience in QA and the tendency my bosses had for brute-forcing shit over automating it, and may not really be applicable to Firaxis since I have no idea how they run their QA team.
If they reinitialized the state on reload, then you could also just reload until your 20% shot hits too
Man, if you want to sit there and reload the game until you get a hit on a 1% shot I really don't care because how other people play the game doesn't impact me, but how the game calculates hits absolutely totally does.
And it's irritating to find out that 55% doesn't really mean what I think of when I see 55%.
How does it impact you though? If it makes you feel better it's not like it actually pregenerates the numbers into a list and laughs at you when you make an impossible shot, it simply says "hmm, shot fired. CALCULATE! oh, it missed"
If you reload it simply reloads the calulation at the same point. The results are fixed, but not known. It literally doesn't effect you in any way unless you decide you want to reload and try a different sequence.
And, if this makes you mad every other computer game that deals with randomness behaves like this. The reloading to a known point in the RNG state is optional, but some games do do it (Civ games for example will sometimes offer an option for it). It just so happens that in a lot of games so many numbers are pulled out and in an often indeterminate order that it's not quite so easy to control
Oh I know. But, there might not actually be a sequence that lets you do what you want whereas with reinit there could easily be. Basically, if you want to cheat you can cheat. I would image they did it for multiplayer and testing, both cases where you need to control randomness
I don't see how this is really a thing that affects multiplayer. It's not like you can reload your shot in multiplayer.
And I feel like testing isn't a great excuse either, but that may be my own previous experience in QA and the tendency my bosses had for brute-forcing shit over automating it, and may not really be applicable to Firaxis since I have no idea how they run their QA team.
It affects multiplayer because instead of one player saying "Oh yeah I totally made that 20% shot, trust me", both computers agree beforehand the scores of each shot, which is then used to determine hits or misses without any outside influence.
0
AegeriTiny wee bacteriumsPlateau of LengRegistered Userregular
edited May 2013
Actually all of you are wrong about why the game saves seeds, it's not to do with testing and not to do with multiplayer or anything similar. It's the preference of the guy who runs Firaxis: Sid Meier. So basically XCOM (and Civilization) uses a pRNG that does pretty much what I explained in a post on the previous page, it draws a list of numbers right at the start of the mission (not even on your turn) then saves those numbers. This has two primary effects, for one it prevents a player from saving/reloading to try to make a low percentage shot hit (as it keeps drawing the same number). You can get around this - but you have to use those poor rolls in another way such as taking a different shot first or similar. On ironman, you are basically screwed and you just have to live with whatever bad hand you are given. The second is it actually reduces the amount of memory the game needs by just pregenerating all the numbers it wants, saving them and then just calling on them whenever it needs to (as opposed to generating them on the fly). So there's that as well.
But essentially, it boils down to Sid Meier disliking the idea players will sit there and reload a billion times to get something impossible to happen. You can see this emphasized several times by him:
This discussion by Sid is about how players perceive randomness and I highly recommend @Fishman and @Phyphor listen to this (this should be a different timestamp to the first one, my apologies if this fails to work. EDIT: And of course it didn't and I can't remember the specific place it's discussed - whole thing is still worth watching).
We use synchronous random in combat so the player can’t just reload when they miss a shot. Now, obviously there are ways around this, but this is a decent way of ensuring that the player’s choices do matter.
Basically it's to ensure the choices you make matter and it's not a case of just save/reloading until you get a result you want.
And as for the discussion about the pRNG not being "random" due to the fact it saves the results first, this is absolutely not true. There is no distingusihable test - unless again you abuse saving and reloading - that would differentiate between the game generating the numbers in advance or when required. Here is an extremely good analysis of how the pRNG works in XCOM. This is the one I mentioned earlier but I didn't have directly on hand for the previous discussion.
If you're wondering why I have all this stuff more or less memorized, I really am not kidding when I say this was the main discussion (about the game cheating on higher difficulties mostly) for almost 2 months on the main forums.
Edit: Also I absolutely 100% agree that the game should have the option of generating numbers on demand. Civilization IIRC has an option to allow for seed state saving, which is a bit perplexing why XCOM doesn't.
Fishman:
Alright, some back-of-the-napkin maths. I hate this shit, by the way, I really think what we'd really need is a proper Monte Carlo simulation to work out the affect and impact of the pRNG modifier, but let's do some stupid calculations.
Sometimes a 55% shot is actually 55%. But sometimes it's 95%. So actually, the problem is that a 55% is actually, on average, some value higher than 55%, secretly.
Yes and you need to remember on normal, if you lose 1 soldier (which does happen with newer players) you actually gain a permanent +20% boost. Add to that a single missed shot over 55% and now it's +40%. So it becomes immediately clear how early on the game becomes substantially easier than it would otherwise seem, especially when that safety net is then gone on classic/impossible (where 1 soldier is not getting a whopping +60% bonus!).
It's only 95% about 25% of the time (give or take, but that will do), because to get that bonus you had to have missed two shots in a row.
My apologies for poorly explaining how the game worked, but actually you can end up with +40% as a permanent bonus if down to two soldiers, or get +40% for having 3 soldiers and then missing a single shot with one of them. So it can occur extremely frequently in the early game on normal.
After all, we're only interested in the delta of the pool of shots as it influences player behaviour and perception. And the answer is... the same as the size of the bonus, 20%. Which gives us 20% of 20% as 'influencing player perception'. Or a 4% difference.
Alternatively I can just give you links like this:
seriously i got the crap beat out of me on classic to the point a lot of times when my guys died it would feel cheap or bull ****y when they died half the time...(due to the insta moves when finding aliens, RNG in their favor to the point it gets ridicules , or insta panic which kills all my guys.) i couldn't even get past the sectoids before i threw in the towel.
How many more would you like me to quote? I can find tons of these discussions early on in the games life and all of them are the same: The games RNG is in the aliens favor on classic! or IT CHEATS!
Frankly unless you haven't read a lot of what other players complained about early on about the game, there is absolutely no way your argument that the RNG isn't making a demonstratable effect on players on normal vs. higher difficulties makes sense. People picked up on this straight away and hence there were endless threads about it, until the discussion showing that:
1) The pRNG was making truly random results (and was seed saving)
and
2) The pRNG was cheating in the players favor on normal.
4%, I feel, is enough to be statistically relevant, but not really 'OMG UR GAEM IS BORKEN' territory. As a step between difficulties, it's probably one of a number of factors that influences the perceived large step in difficulty, but I doubt it's the only one.
I never said it was the only one and given the way it actually works, the effect will be significantly greater than 4% (plus you don't consider how it penalizes the aliens shooting as well). But again just google "RNG in XCOM cheats" for all the reading material you could ever want about why people thought classic/impossible were cheating.
Well, no they're actually "drawn" when needed, but the distinction is academic because there's no difference between pre-building a list and computing on the fly given the seeds are saved properly.
No programmer is going to pre-build a list of random numbers when the state is saved anyway. A high quality prng state is going to be around 256 bits. Their prng is a 32-bit linear congruential generator, so the memory required for it is 4 bytes anyway.
I am well aware that the majority of people simply don't get probability, it's why casinos exist
AegeriTiny wee bacteriumsPlateau of LengRegistered Userregular
edited May 2013
I also forgot about this excellent article: How to test XCOM dice rolls for fairness. The cliff notes conclusion being that unless you're trying to save scumm the game, there is no difference in fairness between the game doing what it does or generating a number on the fly. It makes no difference (other than psychologically).
I am well aware that the majority of people simply don't get probability, it's why casinos exist
I once saw someone put $3000 on the next roll in Sic Bo being low as the previous 5 rolls were all high. He was 100% confident there was no way that the sixth number could be high after that streak (noting that this was real dice in a real casino). I put 5$ on it being high. I ended up with $10.
Alright, some back-of-the-napkin maths. I hate this shit, by the way, I really think what we'd really need is a proper Monte Carlo simulation to work out the affect and impact of the pRNG modifier, but let's do some stupid calculations.
Sometimes a 55% shot is actually 55%. But sometimes it's 95%. So actually, the problem is that a 55% is actually, on average, some value higher than 55%, secretly.
Yes and you need to remember on normal, if you lose 1 soldier (which does happen with newer players) you actually gain a permanent +20% boost. Add to that a single missed shot over 55% and now it's +40%. So it becomes immediately clear how early on the game becomes substantially easier than it would otherwise seem, especially when that safety net is then gone on classic/impossible (where 1 soldier is not getting a whopping +60% bonus!).
It's only 95% about 25% of the time (give or take, but that will do), because to get that bonus you had to have missed two shots in a row.
My apologies for poorly explaining how the game worked, but actually you can end up with +40% as a permanent bonus if down to two soldiers, or get +40% for having 3 soldiers and then missing a single shot with one of them. So it can occur extremely frequently in the early game on normal.
Okay, this is a new factor that wasn't accounted for because this is new information. Yeah, this would swing the numbers considerably higher, although now we really need Monte Carlo because we can't really accurately model the loss of soldiers cleanly in the methodology I was using. But at a complete, stick my finger in the air and pull a number out of my butt guess, if we say (roughly) half of all shots are made when you're a man down, we're doubling the effective influence of the bonus, which would put us in the 8%-better-than-we-think-we-are range.
That's considerably more relevant than what was initially presented. That's one in 12 shots. Not inconsiderable at all, although these numbers are incredibly shaky.
4%, I feel, is enough to be statistically relevant, but not really 'OMG UR GAEM IS BORKEN' territory. As a step between difficulties, it's probably one of a number of factors that influences the perceived large step in difficulty, but I doubt it's the only one.
I never said it was the only one and given the way it actually works, the effect will be significantly greater than 4% (plus you don't consider how it penalizes the aliens shooting as well).
The Alien side would be included provided they were using the same RNG sequence and had similar penalties. It's a measure of influence of the modifier, or incidence of affected shots, which is to say how often the secret bonus/penalty hits in favour or misses in favour. So long as the Aliens have a corresponding penalty, the number wouldn't change. So yes, it was considered. Indeed, you'd want to half that number if it wasn't. But given the additional information that has now been presented, I can see where the RNG likely contributes disproportionately to the perceived gulf between Normal and Classic difficulties.
If they reinitialized the state on reload, then you could also just reload until your 20% shot hits too
Man, if you want to sit there and reload the game until you get a hit on a 1% shot I really don't care because how other people play the game doesn't impact me, but how the game calculates hits absolutely totally does.
And it's irritating to find out that 55% doesn't really mean what I think of when I see 55%.
How does it impact you though? If it makes you feel better it's not like it actually pregenerates the numbers into a list and laughs at you when you make an impossible shot, it simply says "hmm, shot fired. CALCULATE! oh, it missed"
If you reload it simply reloads the calulation at the same point. The results are fixed, but not known. It literally doesn't effect you in any way unless you decide you want to reload and try a different sequence.
Knowing it's fixed triggers my "bullshit" sense.
It's 100% a matter of perception; doesn't make it any less irritating.
And, if this makes you mad every other computer game that deals with randomness behaves like this. The reloading to a known point in the RNG state is optional, but some games do do it (Civ games for example will sometimes offer an option for it). It just so happens that in a lot of games so many numbers are pulled out and in an often indeterminate order that it's not quite so easy to control
I always play with that random-seed option checked.
Oh I know. But, there might not actually be a sequence that lets you do what you want whereas with reinit there could easily be. Basically, if you want to cheat you can cheat. I would image they did it for multiplayer and testing, both cases where you need to control randomness
I don't see how this is really a thing that affects multiplayer. It's not like you can reload your shot in multiplayer.
And I feel like testing isn't a great excuse either, but that may be my own previous experience in QA and the tendency my bosses had for brute-forcing shit over automating it, and may not really be applicable to Firaxis since I have no idea how they run their QA team.
It affects multiplayer because instead of one player saying "Oh yeah I totally made that 20% shot, trust me", both computers agree beforehand the scores of each shot, which is then used to determine hits or misses without any outside influence.
You're saying it's a way to prevent bugs stemming from either client not agreeing on what the outcome of a given roll was?
That makes a ton of sense.
Like, no sarcasm, that is a super solid reason to do it this way for multiplayer.
This is a SUPER interesting, really great post, thank you Aegeri.
+1
AegeriTiny wee bacteriumsPlateau of LengRegistered Userregular
edited May 2013
The aliens do share the same pRNG sequence, this is even easy to prove for anyone who wants to confirm this themselves. Hit with a ridiculously low percentage shot, like 25% and then miss with another higher shot - maybe around 75%. Reload the game and do nothing, the aliens will hit with the first shot (that 25%) and then probably miss on the second shot. Then reload the game, take the first shot and do nothing. The alien that hit in the second test, then promptly misses - so they definitely do use the exact same numbers as you do and yes, will pick up similar penalties to hit you. I actually think I might have the facts slightly wrong, I think it might be that the aliens get a -20% penalty to hit you when you are down below 4 soldiers and not the other way (giving you a bonus). I will need to check that actually, but the aliens can be packing some serious penalties on normal if they hit you and minimizing the damage you take is pretty significant.
While if you even just read this thread, you'll see people complaining about how the aliens are one shotting them in full cover on classic (with the implicit assumption that they weren't doing that on normal). Well that's because they aren't getting penalized on their shooting anymore and have slightly better aim and crit (but the 10% difference would not be enough to explain why it's suddenly much harder). In fact when you consider the aliens are penalized for hitting high % shots, take penalties when you are down below 4 soldiers to all their shooting AND have a +10% bonus, you're making a much bigger gap in difficulty than giving them +10%. Removing the cheating in the players favor AND the +10% bonus is a significant leap, much more significant than a simple +10% aim bonus should suggest. Nobody argues that every soldier should have a scope (+10 aim) as mandatory must have equipment, because while +10 helps it isn't always the biggest advantage over say having a medkit or a grenade or whatever.
So the difference is something other than giving aliens +10 aim, +10 crit, an extra pip of health and being able to engage you with more than 5 aliens simultaneously (I want to say on classic it's 10, but I am not sure). These are all increases on difficulty that to me, should be 100% manageable if you have played normal. But classic is frequently a brick wall for players going from normal -> classic, so the explanation has to be deeper and the fact the game cheats in your favor so heavily on normal is the best explanation (IMO).
And, if this makes you mad every other computer game that deals with randomness behaves like this. The reloading to a known point in the RNG state is optional, but some games do do it (Civ games for example will sometimes offer an option for it). It just so happens that in a lot of games so many numbers are pulled out and in an often indeterminate order that it's not quite so easy to control
I always play with that random-seed option checked.
I know they do this in Civilization, but I wonder why they didn't provide the option in XCOM as I don't give a shit if anyone wants to save scumm and technically speaking, you can anyway. It's just much much more awkward.
Edit2: Okay, I am semi-correct. You get +20% AIM when down to 3 soldiers or below on normal (permanent boost), +20% aim that is cumulative per miss that you make and aliens take a -10% penalty per hit they make (again, cumulative).
I assume (without any actual proof) that, much like the "Nobody cares about different accents/languages", it was an executive decision by the lead designer.
I assume (without any actual proof) that, much like the "Nobody cares about different accents/languages", it was an executive decision by the lead designer.
I suspect it's like the second wave options: They just didn't think of implementing it or just ran out of time before they decided to do so. I wonder if the expansion (or whatever they are doing) will implement it.
Edit: I really want to watch your stream, but unfortunately it's a bit too laggy where I am. Good luck with those floaters.
I assume (without any actual proof) that, much like the "Nobody cares about different accents/languages", it was an executive decision by the lead designer.
I suspect it's like the second wave options: They just didn't think of implementing it or just ran out of time before they decided to do so. I wonder if the expansion (or whatever they are doing) will implement it.
Edit: I really want to watch your stream, but unfortunately it's a bit too laggy where I am. Good luck with those floaters.
I got hit with the bug where they act on discovery. Lost everyone that mission because of it.
AegeriTiny wee bacteriumsPlateau of LengRegistered Userregular
edited May 2013
That is the biggest bullshit bug. I got that on the rainy map a while ago and it was a total pain in the ass. Thankfully I was far enough back they couldn't get flanks (and thus hit my hunkered down forward scouts) and then I rocketed them to oblivion. Was wondering why those floaters were acting so aggressively immediately.
I assume (without any actual proof) that, much like the "Nobody cares about different accents/languages", it was an executive decision by the lead designer.
I suspect it's like the second wave options: They just didn't think of implementing it or just ran out of time before they decided to do so. I wonder if the expansion (or whatever they are doing) will implement it.
Edit: I really want to watch your stream, but unfortunately it's a bit too laggy where I am. Good luck with those floaters.
Actually I don't think it was the bug, I think I just forgot I had moved. I probably shouldn't have played right now, but I think I screwed the metagame on that go anyway.
I think I might get slingshot for my next run though.
It was still bullshit that I still got 2 groups of floaters and one group of chrysalids on the first turn though.
I love capture-heavy playthroughs, even after you have all the research credits. why yes, I'll gladly spend $200/20 ele/20 alloys on a plasma rifle to finish your 6 rifle order and get $3000. come back soon, don't be a stranger!
I think after this playthrough I'll probably try I/I marathon, even though trying to think of a tactical build order that'll have enough engineers/money/time to avoid losing 5+ countries by the second month seems...not fun.
Torgairon on
0
Brovid Hasselsmof[Growling historic on the fury road]Registered Userregular
edited May 2013
People say these numbers are fixed but I dunno. I thought they were, but I've had a few occasions in my current LP game where I've reloaded in order to try and get a screenshot I missed the first time, and done EXACTLY the same move, and missed shots I just hit.
I also had one bit on the Trainyard of Death map when Rank was critically wounded, where if I reloaded a save Bogey started panicking, even though he wasn't panicking when I saved it and nobody got shot at or anything when I reloaded.
AntimatterDevo Was RightGates of SteelRegistered Userregular
every single unit from that concept art had better be added in EW, balance be damned
(jokes)
0
Munkus BeaverYou don't have to attend every argument you are invited to.Philosophy: Stoicism. Politics: Democratic SocialistRegistered User, ClubPAregular
I like the little pink guys with the big guns.
Humor can be dissected as a frog can, but dies in the process.
0
Caulk Bite 6One of the multitude of Dans infesting this placeRegistered Userregular
Somehow, I get the feeling that whatever the plan was for the giant bunny, it would have been just as terrifying as the Chrysalid.
EDIT: In fact, looking at the fact that they have a small bunny there too, my guess would be that it was another self-multiplier -- except instead of converting soldiers and civilians, it would have just continually bred with others of its kind in dark corners of the map.
Gandalf_the_Crazed on
+1
Caulk Bite 6One of the multitude of Dans infesting this placeRegistered Userregular
Ugh, I'm trying to start a classic ironman run, but all of my rookies suck. None of them can get more than 50% hit chance even when flanking and they seem to love sticking their heads out of full cover.
I've already started over 4 times. Two of which started with failed opening operations.
Posts
Man, if you want to sit there and reload the game until you get a hit on a 1% shot I really don't care because how other people play the game doesn't impact me, but how the game calculates hits absolutely totally does.
And it's irritating to find out that 55% doesn't really mean what I think of when I see 55%.
What actually matters is how hilarious it makes watching Beagle-Live.
And all his strategy and talking about removing the 'randomness' by trying to make sure his shots are a success.
Early Impossible is a total crapshoot with the RNG like it is, and going at it carefully at the start is a mistake. Ram your head until you get past that first start and then it becomes a game.
I don't see how this is really a thing that affects multiplayer. It's not like you can reload your shot in multiplayer.
And I feel like testing isn't a great excuse either, but that may be my own previous experience in QA and the tendency my bosses had for brute-forcing shit over automating it, and may not really be applicable to Firaxis since I have no idea how they run their QA team.
How does it impact you though? If it makes you feel better it's not like it actually pregenerates the numbers into a list and laughs at you when you make an impossible shot, it simply says "hmm, shot fired. CALCULATE! oh, it missed"
If you reload it simply reloads the calulation at the same point. The results are fixed, but not known. It literally doesn't effect you in any way unless you decide you want to reload and try a different sequence.
3DS: 1289-8447-4695
It affects multiplayer because instead of one player saying "Oh yeah I totally made that 20% shot, trust me", both computers agree beforehand the scores of each shot, which is then used to determine hits or misses without any outside influence.
But essentially, it boils down to Sid Meier disliking the idea players will sit there and reload a billion times to get something impossible to happen. You can see this emphasized several times by him:
This discussion by Sid is about how players perceive randomness and I highly recommend @Fishman and @Phyphor listen to this (this should be a different timestamp to the first one, my apologies if this fails to work. EDIT: And of course it didn't and I can't remember the specific place it's discussed - whole thing is still worth watching).
Here is an article with a quote from Jake Solomon about why they did this in XCOM:
Basically it's to ensure the choices you make matter and it's not a case of just save/reloading until you get a result you want.
And as for the discussion about the pRNG not being "random" due to the fact it saves the results first, this is absolutely not true. There is no distingusihable test - unless again you abuse saving and reloading - that would differentiate between the game generating the numbers in advance or when required. Here is an extremely good analysis of how the pRNG works in XCOM. This is the one I mentioned earlier but I didn't have directly on hand for the previous discussion.
If you're wondering why I have all this stuff more or less memorized, I really am not kidding when I say this was the main discussion (about the game cheating on higher difficulties mostly) for almost 2 months on the main forums.
Edit: Also I absolutely 100% agree that the game should have the option of generating numbers on demand. Civilization IIRC has an option to allow for seed state saving, which is a bit perplexing why XCOM doesn't.
Fishman:
Yes and you need to remember on normal, if you lose 1 soldier (which does happen with newer players) you actually gain a permanent +20% boost. Add to that a single missed shot over 55% and now it's +40%. So it becomes immediately clear how early on the game becomes substantially easier than it would otherwise seem, especially when that safety net is then gone on classic/impossible (where 1 soldier is not getting a whopping +60% bonus!).
My apologies for poorly explaining how the game worked, but actually you can end up with +40% as a permanent bonus if down to two soldiers, or get +40% for having 3 soldiers and then missing a single shot with one of them. So it can occur extremely frequently in the early game on normal.
Alternatively I can just give you links like this:
The difference between normal and classic is incredibly jarring
And then quote where it matters:
How many more would you like me to quote? I can find tons of these discussions early on in the games life and all of them are the same: The games RNG is in the aliens favor on classic! or IT CHEATS!
Frankly unless you haven't read a lot of what other players complained about early on about the game, there is absolutely no way your argument that the RNG isn't making a demonstratable effect on players on normal vs. higher difficulties makes sense. People picked up on this straight away and hence there were endless threads about it, until the discussion showing that:
1) The pRNG was making truly random results (and was seed saving)
and
2) The pRNG was cheating in the players favor on normal.
I never said it was the only one and given the way it actually works, the effect will be significantly greater than 4% (plus you don't consider how it penalizes the aliens shooting as well). But again just google "RNG in XCOM cheats" for all the reading material you could ever want about why people thought classic/impossible were cheating.
No programmer is going to pre-build a list of random numbers when the state is saved anyway. A high quality prng state is going to be around 256 bits. Their prng is a 32-bit linear congruential generator, so the memory required for it is 4 bytes anyway.
I am well aware that the majority of people simply don't get probability, it's why casinos exist
Edit:
I once saw someone put $3000 on the next roll in Sic Bo being low as the previous 5 rolls were all high. He was 100% confident there was no way that the sixth number could be high after that streak (noting that this was real dice in a real casino). I put 5$ on it being high. I ended up with $10.
Okay, this is a new factor that wasn't accounted for because this is new information. Yeah, this would swing the numbers considerably higher, although now we really need Monte Carlo because we can't really accurately model the loss of soldiers cleanly in the methodology I was using. But at a complete, stick my finger in the air and pull a number out of my butt guess, if we say (roughly) half of all shots are made when you're a man down, we're doubling the effective influence of the bonus, which would put us in the 8%-better-than-we-think-we-are range.
That's considerably more relevant than what was initially presented. That's one in 12 shots. Not inconsiderable at all, although these numbers are incredibly shaky.
The Alien side would be included provided they were using the same RNG sequence and had similar penalties. It's a measure of influence of the modifier, or incidence of affected shots, which is to say how often the secret bonus/penalty hits in favour or misses in favour. So long as the Aliens have a corresponding penalty, the number wouldn't change. So yes, it was considered. Indeed, you'd want to half that number if it wasn't. But given the additional information that has now been presented, I can see where the RNG likely contributes disproportionately to the perceived gulf between Normal and Classic difficulties.
Knowing it's fixed triggers my "bullshit" sense.
It's 100% a matter of perception; doesn't make it any less irritating.
I always play with that random-seed option checked.
You're saying it's a way to prevent bugs stemming from either client not agreeing on what the outcome of a given roll was?
That makes a ton of sense.
Like, no sarcasm, that is a super solid reason to do it this way for multiplayer.
This is a SUPER interesting, really great post, thank you Aegeri.
While if you even just read this thread, you'll see people complaining about how the aliens are one shotting them in full cover on classic (with the implicit assumption that they weren't doing that on normal). Well that's because they aren't getting penalized on their shooting anymore and have slightly better aim and crit (but the 10% difference would not be enough to explain why it's suddenly much harder). In fact when you consider the aliens are penalized for hitting high % shots, take penalties when you are down below 4 soldiers to all their shooting AND have a +10% bonus, you're making a much bigger gap in difficulty than giving them +10%. Removing the cheating in the players favor AND the +10% bonus is a significant leap, much more significant than a simple +10% aim bonus should suggest. Nobody argues that every soldier should have a scope (+10 aim) as mandatory must have equipment, because while +10 helps it isn't always the biggest advantage over say having a medkit or a grenade or whatever.
So the difference is something other than giving aliens +10 aim, +10 crit, an extra pip of health and being able to engage you with more than 5 aliens simultaneously (I want to say on classic it's 10, but I am not sure). These are all increases on difficulty that to me, should be 100% manageable if you have played normal. But classic is frequently a brick wall for players going from normal -> classic, so the explanation has to be deeper and the fact the game cheats in your favor so heavily on normal is the best explanation (IMO).
I know they do this in Civilization, but I wonder why they didn't provide the option in XCOM as I don't give a shit if anyone wants to save scumm and technically speaking, you can anyway. It's just much much more awkward.
Edit2: Okay, I am semi-correct. You get +20% AIM when down to 3 soldiers or below on normal (permanent boost), +20% aim that is cumulative per miss that you make and aliens take a -10% penalty per hit they make (again, cumulative).
I assume (without any actual proof) that, much like the "Nobody cares about different accents/languages", it was an executive decision by the lead designer.
I begin my first terror mission of this Ironman Impossible game. Will my squad survive?
Also I'm not completely sober.
I suspect it's like the second wave options: They just didn't think of implementing it or just ran out of time before they decided to do so. I wonder if the expansion (or whatever they are doing) will implement it.
Edit: I really want to watch your stream, but unfortunately it's a bit too laggy where I am. Good luck with those floaters.
I got hit with the bug where they act on discovery. Lost everyone that mission because of it.
Video is fine for me, but the sound is horrible
I think I might get slingshot for my next run though.
It was still bullshit that I still got 2 groups of floaters and one group of chrysalids on the first turn though.
I think after this playthrough I'll probably try I/I marathon, even though trying to think of a tactical build order that'll have enough engineers/money/time to avoid losing 5+ countries by the second month seems...not fun.
I also had one bit on the Trainyard of Death map when Rank was critically wounded, where if I reloaded a save Bogey started panicking, even though he wasn't panicking when I saved it and nobody got shot at or anything when I reloaded.
Game's weird.
(also giant bunny rabbits)
(jokes)
He looks a sight intimidating
Twitch (I stream most days of the week)
Twitter (mean leftist discourse)
JESUS CHRIST
EDIT: In fact, looking at the fact that they have a small bunny there too, my guess would be that it was another self-multiplier -- except instead of converting soldiers and civilians, it would have just continually bred with others of its kind in dark corners of the map.
I've already started over 4 times. Two of which started with failed opening operations.
I would chat this, but chat is for-ever loading for some stupid reason.