Left 4 Dead 2

Left 4 Dead 2

Freddy Fazbear's Pizzeria
This topic has been locked
Olde 17 Dec, 2016 @ 4:05pm
Functionally Unbeatable
There must have been a corruption with the newest update (2.9) because once the game gets to night 3, the campaign becomes functionally unbeatable. Let me tell you how I drew this conclusion.

I played through this campaign on Normal with one friend, then two friends, and finally with three friends. Each time we never made it past the third night, even though we had beaten the FNAF2 and FNAF3 campaigns by the same author on Normal on the first tries each, without any difficulty. But in this campaign, the same thing kept happening each time. We would stay in the main room (the one with the radio, plushies, doors, and light switches), with two people strategically covering each hallway. At a certain point, the special infected rate along with nearly constant hordes and double tank spawns means that someone is going to go down, and anyone who tries to help will go down too. This isn't a resource problem, meaning it can't be fixed by throwing medkits and pills around. This is just the way the campaign operates at scaling up the difficulty. Additionally, we never have any problems surviving nights 1 or 2, it only becomes insurmountable at night 3.

So I came to the comments looking for similar situations. I didn't find any that coincided with mine until recently. One comment by the author told me to watch the video of the author beating the campaign with a friend (so, only two players and two bots) and I jumped ahead to night 3. It was very dissimilar to our experiences; in that scenario it played much like night 2 for us, nothing like our night 3. So I did a little bit of investigative work.

What I did was play through the entire campaign with God Mode enabled to see if my number of total infected would differ from the author's in his video. There were restarts in his video so what I did was I added up all the Special Infected and Common Infected Kills (discerned by subtracting special from total kills) and divided it by the number of minutes we each played. That is, my playthrough was 1h3m while the author's was 3h26m. The reason why mine was still so long even with God Mode enabled was because I was actually exploring the map, gathering items and such, so even though the author's video omitted roughly 2h30m of game time, my playthrough should match the same ratio of time played per night with exploring, unless the author and his friend went AWOL for an hour or so.

What I discovered when tallying up the kills confirmed my suspicion. The Special Infected kill rate for my playthrough was 6.25/min and Common Infected kill rate for my playthrough was 50.9/min. Conversely, the Special Infected kill rate for the author's playthrough was 2.78/min and Common Infected kill rate was 25.9/min. What this means is that the infected rate is essentially doubled in the current version of the campaign.

This kind of discrepency can't be explained away by chance, bad luck, or Director anger. But you might ask, so what? The issue is that this spawn rate is so hight that there is no realistic way that the player can be expected to survive, even with the doors (which occasionally get bugged, as in one case where the door on the right closed but wouldn't re-open until the night was over). There was a point in time where every kind of special infected spawned except for a hunter whilst the group was being attacked by a tank and horde. That is a functionally unsurvivable situation. Other facts that may be interesting to note are that spitters spawn most frequently, followed very shortly behind with boomers. Chargers are third most likely and hunters fourth. Smokers and jockeys spawn the least frequently.

In short, my point is that the new version becomes unbeatable without abuse of the doors, and even with closing the doors, the spawn rate of spitters discourages staying in the room. Other minor problems that appear in the current version but not in the version in the video have been noted, such as the Usage and Charge Level graphics not appearing.

If the author still claims that the campaign is still beatable on Normal difficulty, then I implore him/her to make a new video showing this possibility. I don't have a problem with campaigns being difficult. I do have a problem when there is a horde, tank, and five special infected attacking everyone at once in a location whose implementation the author defends yet the gameplay actively discourages. We already know that the gameplay getting ramped up is only difficulty in its artificiality, since the players hold out in the same environment each time and the hordes only differ in amount and duration of time. But this put a really sour taste in my mouth and I implore the author to take another look at the newest version and fix the mistake, since as of right now I don't believe this campaign is realistically beatable.
< >
Showing 1-7 of 7 comments
Deep Thunder  [developer] 17 Dec, 2016 @ 5:17pm 
https://www.youtube.com/watch?v=7utDl1KA9tY

^ incase anyone is wondering which video he's talking about
Last edited by Deep Thunder; 17 Dec, 2016 @ 5:39pm
Deep Thunder  [developer] 17 Dec, 2016 @ 5:28pm 
I'll read the entire post in more depth later, but at this moment any future updates have been abandoned. The community and other players have more than once indicated that it simply isn't worth it.

If I decide to come back to this map, I'll fully analyze your post. But in the meantime, I'm VERY well aware of the balance issues. Version 3 had plans to rectify that issue. If there were checkpoints for 2.9, I guess it would have been a different story. From a technical point of view, implementing checkpoints is a royal pain in the ass, and quite frankly I do not have the horse power or budget to continue. Personally it's getting really old with these wait times and I don't have the patience for it.

I'm getting a much better PC very soon. Whether or not that'll motivate me to come back to this map, I don't know but I'm leaning towards no.

You are right that it's unrealistically beatable, absolutely. I would, however, hate to make things too easy. It would be boring too easy. It depends if I really want to extensively test the difficulty; taking all gameplay variables into consideration. I guess I could recycle one of valve's custom scripts would work, but, what's the fun in that, ya know? Also, I would play the map again to prove your point but, I just really don't feel like.

Thanks a ton for the in-depth feedback, pretty awesome you went so in-depth with those numbers. Only every so often this will happen so I thoroughly appreciate it.
Last edited by Deep Thunder; 17 Dec, 2016 @ 5:28pm
Olde 17 Dec, 2016 @ 5:45pm 
Originally posted by Deep Thunder:
The community and other players have more than once indicated that it simply isn't worth it.
I disagree with that completely. Perhaps your perspective has been tainted by a few sour dissenters, but from what I've read from commenters, the majority of reception has been overwhelmingly positive. Even before I gave it a proper review, I had said that despite my not liking the FNAF series, that this was worth playing. The issue is the balance. If it were better balanced, it would be a standard worth many replays.

Originally posted by Deep Thunder:
If I decide to come back to this map, I'll fully analyze your post. But in the meantime, I'm VERY well aware of the balance issues. Version 3 had plans to rectify that issue. If there were checkpoints for 2.9, I guess it would have been a different story. From a technical point of view, implementing checkpoints is a royal pain in the ass, and quite frankly I do not have the horse power or budget to continue. Personally it's getting really old with these wait times and I don't have the patience for it.
I had read you mentioning that you abandoned the new version idea; I did not even think checkpoints were even possible. In any case, they are not necessary, the only thing necessary is the rebalancing. If you already really know how badly it gets from night 2 to 3, then I'm at least glad you're aware of that.

Originally posted by Deep Thunder:
You are right that it's unrealistically beatable, absolutely. I would, however, hate to make things too easy. It would be boring too easy. It depends if I really want to extensively test the difficulty; taking all gameplay variables into consideration. I guess I could recycle one of valve's custom scripts would work, but, what's the fun in that, ya know? Also, I would play the map again to prove your point but, I just really don't feel like.
The third night's intensity is what should be on night 5, in my humble opinion. The escalation is what needs rebalancing. My friends and I would be happy to entertain you for a night as we play it; we are very nice and team players.

Originally posted by Deep Thunder:
Thanks a ton for the in-depth feedback, pretty awesome you went so in-depth with those numbers. Only every so often this will happen so I thoroughly appreciate it.
You're very welcome. I only went into as much detail as I did because I know the campaign has so much potential and ultimately only needs minor tweaking. I would not have wasted my time on such an assessment with a lesser map. It is also a rarity that an author actually responds with feedback respectfully and in a timely manner, and even considers updating their work based on player input.
Last edited by Olde; 17 Dec, 2016 @ 5:45pm
[DT] GoblinKing 29 Dec, 2016 @ 10:21am 
If the map is too difficult on Normal, why not try it on Easy?

I've beaten the map twice, it can definitely be done. The main trick is to close the doors when the tanks come & shot them through the doors.

Also, throwing molotov in a hallway is invaluable to block it and give time to take out the SIs.

I also recommend you stay in the hallway near the main room, and only retreat to the main room to avoid acid spits & to hide from tanks.
Olde 21 Jul, 2017 @ 4:38pm 
Okay, I'm back after playing this map again with two other players on Normal difficulty and we did in fact manage to beat it. Night 3 is still by far the hardest as it seems to immediately spawn at least one of every type of special infected. On our first run we got utterly decimated right at the start of night 3, but the second time, we had one AK user constantly firing down each hallway, one SPAS user and the bot used the hunting rifle. The fourth and fifth nights saw a lot of specials and tanks getting stuck in walls/rooms and just dying. Furthermore, when one player on the right stays right by the vent (which leads to the kitchen), it prevents certain spawns, making the hordes and specials more manageable. Because of this, the fourth and fifth nights were about as easy as the second night, which makes the map much more beatable. The part outside is so much easier that it's a joke by comparison. By that point you have the best weapons (with laser sights), significant vantage spots, no tank spawns until the end, and an ammo pile close by.

I'm very torn on what this means for the map. On the one hand, it's a good thing in that it shows that the campaign is beatable, granted you have at least two other experienced players. On the other hand, I think it needs more rebalancing, not least of all because when I played it with three other experienced people, we always got rekt at the same part. Plus, I'm not sure why the third night would have to be the hardest, unless this is unintended. Ever since my review, I've eagerly awaited an update for this map to make it more realistically beatable. Now that I've beaten without even a full team...I don't know. I think it still has room for improvements because it's got a pretty huge and unexpected difficulty spike on night 3, which then seems to lower on nights 4 and 5.

Functionally, I feel like the campaign reaches its peak too early, making the rest of the map drag, rather than building up to a climax throughout. Again, I'm not sure if this was intentional or not, but either way, I think it could be addressed in an update.

I'm posting our game stats because some people may find them useful or interesting:

http://gtm.steamproxy.vip/sharedfiles/filedetails/?id=1085269278
http://gtm.steamproxy.vip/sharedfiles/filedetails/?id=1085269360
http://gtm.steamproxy.vip/sharedfiles/filedetails/?id=1085269416
http://gtm.steamproxy.vip/sharedfiles/filedetails/?id=1085269790
http://gtm.steamproxy.vip/sharedfiles/filedetails/?id=1085269826
Last edited by Olde; 21 Jul, 2017 @ 4:43pm
Deep Thunder  [developer] 21 Jul, 2017 @ 10:27pm 
Ever since my review, I've eagerly awaited an update for this map to make it more realistically beatable.

I'm going to address this first because it's the easier to "answer". I'm sure you know this but I want to make it very clear. Personally I've been eagerly awaiting to finish up this map.

I won't release a "singular" update to address balance. It's more than just moving around numbers. Off the top of my head, I don't think I could come up with precise numbers that demonstrate an actually satisfactory gradual progression of difficulty for each night (which is indeed what others and I will be playing around with in the future). And, like you mentioned, special infected and tanks will die randomly. This may or may not be happening to common infected, which I'll verify at a later date.

Due to that huge issue needing to be fixed, I'd much rather release it with everything else. Dealing with the AI director is arguably the most frustrating and boring task, and as a result, I usually get to it last. On the brightside, I've developed/polished/optimized other areas of the map. Even by my own standards, I accomplished a ton of work in this area. I've taken my sweet time making sure such functionality works, and I'll make sure I'll do the same with balancing and the AI director. My new desktop makes this workload 20x easier.

You've eagerly been waiting, and I agree it's been way too long since then. But since I am not receiving monetary benefits in return for making this map, this is the price others will pay.

Night 3 is still by far the hardest as it seems to immediately spawn at least one of every type of special infected...Plus, I'm not sure why the third night would have to be the hardest, unless this is unintended.

I can assure you that's 100% unintended. It was intended to be a gradual progression of difficulty.

I always felt like there was a good enough gradual progression. Due to my teammates and I beating the map, I thought it was a solid enough benchmark anyway. Overtime, the general consensus became more clear to me from people's comments.

Since there will be checkpoints now, I wonder if it'll even be worth it. If you die on Night 3, you'll restart on Night 3, instead of Night 1. Not a bad deal, eh?

It's important to exactly pinpoint what is so "spikey" about Night 3. That way, I can tone it down, whatever it is, and thus, we'll be able to achieve that gradual progression. It seems having all Special Infected at once is the problem (in addition to the extra 2 minutes) so do correct me if I am wrong.

Allow me to show you a comparison between Night 2 and Night 3. Difference are underlined:

Originally posted by Night 2's numbers:
DirectorOptions <-
{
CommonLimit = 40
MobSpawnMinTime = 25
MobSpawnMaxTime = 30
MobMinSize = 40
MobMaxSize = 40
ProhibitBossess = 0
MaxSpecials = 3
ShouldAllowSpecialsWithTank = 1
ShouldAllowMobsWithTank = 1

//Time: 6 minutes
}

Originally posted by Night 3's numbers:
DirectorOptions <-
{
CommonLimit = 40
MobSpawnMinTime = 25
MobSpawnMaxTime = 30
MobMinSize = 40
MobMaxSize = 40
ProhibitBossess = 0
MaxSpecials = 5
ShouldAllowSpecialsWithTank = 1
ShouldAllowMobsWithTank = 1

//Time: 8 minutes
}

The rest of the nights will be displayed further down this post.

This information is taken from the current release of the map (2.9). The differences between Nights 2 and 3 is that Night 3 has two more special infected than Night 2. In addition, Night 3, at 8 minutes, lasts two more minutes than Night 2. In the upcoming update, Night 3 will last 6 minutes.

It seems that the extra two minutes is articificially contributing to that difficulty spike. The extra two special infected is most likely causing all of the special infected to appear. I wonder, though, how to address this? Should I at all? Should new special infected gradually appear at each night? And if so, which ones, and in which order? How does one determine that?

I'm sure there are other factors in the process, such as if the infected are dying randomly or even spawning at all. In other words:

The fourth and fifth nights saw a lot of specials and tanks getting stuck in walls/rooms and just dying.

Yeah, that's not intended. I intend to address it.

In any case, that about summarizes my thoughts in response to yours in the quickest manner possible. Essentially, the spike on Night 3 throws off the rest of the map's balance, and I think your post eloquently details how and why that happens. Yes, not even just Night 3 that needs to be addressed, but the rest of the nights as well.

Lastly, since 2.9 will be obsolete in the upcoming future, here are the infected numbers for nights 1-5. May these numbers assist your endeavors of further breaking down this map:

Originally posted by Night 1's numbers:

DirectorOptions <-
{
CommonLimit = 40
MobSpawnMinTime = 25
MobSpawnMaxTime = 33
MobMinSize = 40
MobMaxSize = 40
ProhibitBossess = 0
MaxSpecials = 2
ShouldAllowSpecialsWithTank = 1
ShouldAllowMobsWithTank = 1
}

Originally posted by Night 2's numbers:

DirectorOptions <-
{
CommonLimit = 40
MobSpawnMinTime = 25
MobSpawnMaxTime = 30
MobMinSize = 40
MobMaxSize = 40
ProhibitBossess = 0
MaxSpecials = 3
ShouldAllowSpecialsWithTank = 1
ShouldAllowMobsWithTank = 1
}

Originally posted by Night 3's numbers:

DirectorOptions <-
{
CommonLimit = 40
MobSpawnMinTime = 25
MobSpawnMaxTime = 30
MobMinSize = 40
MobMaxSize = 40
ProhibitBossess = 0
MaxSpecials = 5
ShouldAllowSpecialsWithTank = 1
ShouldAllowMobsWithTank = 1
}

Originally posted by Night 4's numbers:

DirectorOptions <-
{
CommonLimit = 45
MobSpawnMinTime = 25
MobSpawnMaxTime = 33
MobMinSize = 45
MobMaxSize = 45
ProhibitBossess = 0
MaxSpecials = 6
ShouldAllowSpecialsWithTank = 1
ShouldAllowMobsWithTank = 1
}

Originally posted by Night 5's numbers:

DirectorOptions <-
{
CommonLimit = 49
MobSpawnMinTime = 25
MobSpawnMaxTime = 35
MobMinSize = 49
MobMaxSize = 49
ProhibitBossess = 0
MaxSpecials = 5
ShouldAllowSpecialsWithTank = 1
ShouldAllowMobsWithTank = 1
}
Last edited by Deep Thunder; 21 Jul, 2017 @ 10:39pm
Olde 22 Jul, 2017 @ 1:03am 
Thank you for your everly consistently well-detailed, informative, and positive response. I have my own opinions about what could potentially be improved, but of course, I'm not the author, you are. All I can do is offer my perspective and opinion, and maybe show you something that you might not have known.

Numbers are one thing. If it were up to me, I would probably lower the 6 specials to 5 on night 3, but that might not ultimately make a difference. The point I probably didn't get across is that there are a lot of issues that compound the difficulty, it's not just sheer numbers: nearly endless streams of common, door energy, specials, running out of ammo, tanks (single and double spawns), tanks hitting players through the doors, bot idiocy, and lastly, human error. The map script may be a series of numbers but gameplay and our experiences are all different.

I did manage to record what happened when two others and I played night 3 and died. Upon reviewing the footage, this is what happened:

At 1:00 AM, two tanks spawned. We plan on closing the doors and getting one tank on each side. However, both one player and a bot are trapped outside when we close the doors. I and the other player inside the room kill both tanks; the player outside the room gets pounced by a hunter. The bot is already dead from the tank. We rush out to help the downed player but a spitter spits on him; that, with his low incap health and the large number of zombies on him, makes rescue unfeasible. Another tank spawns. The other player and I both rush back to the room. I get jockey'd, which promptly gets shoved off. My friend gets smoked, the tongue of which I promptly break. We close the left door before the tank gets to it, but it manages to incap my friend through the door even though he's standing in the middle of the room. At this point I'm the only one not dead or incapped, killing common, but then turn my attention to the tank and kill it through the door. Just as I'm about to help my friend up, I get vomited on and a spitter spits in the room. My incapped friend kills the boomer as I immediately open the door. The spitter starts clawing at me while my incapped friend gets pounced on by a hunter. As soon as I kill the spitter, a jockey leaps on me and a smoker shoots its tongue out at me. Even if I had successfully deflected the jockey, there would have been no way for me to avoid being smoked.

The tl;dr version: one player died while trapped on the other side of the door, the other player was incapped by a tank through the closed door, and ultimately we died to a boomer + spitter + hunter + jockey + smoker + human error.

If it's one thing I think this campaign doesn't accommodate most, it's human error. Because players are going to make mistakes. The question is should players be punished for making mistakes that may or may not be in their control, such as hitting the light switch instead of the door, locking someone out, assuming spit won't get under the door or on the table, going to the ammo pile at the wrong time? Should players be punished for healing bots? For failing to heal bots? Should they be punished for not tackling the map with three or four seasoned players? The map is a challenge, yes, and it is beatable. But I wonder, for every player who beats it, how many never even find a way to unlock the ammo pile? How many players don't know the importance of carrying defibs from the cellar to the room? How many bots have wasted the pills and adrenaline when still in the green or failed to get inside the room when there was a tank? As an author I can imagine it's a predicament wherein it's seemingly impossible to satisfy the average L4D2 player as well as the well-seasoned one. But while the question of how realistic it is to beat may not be of concern to a map maker, it certainly is to the player.

I still can't really voice my opinion about what to do for this campaign. Should it be easier? Harder? I can't quite say, nor does my opinion matter because I'm not the author. All I can say is I'm happy that you're still working on this campaign and I have no reason to doubt that the newer update will be worth the wait.

Edit: My thoughts as I'm re-watching some footage. It seems as though if everyone keeps constant pressure on the hallways, there's very little actual risk of something going wrong. However, if even one special gets through--say, if someone is going for ammo--it can be a death sentence for everyone. If one person gets jockey'd away from the room or someone in the hallway gets charged, that might let a spitter through to make the room unavailable or might allow a smoker to drag another survivor into the main dining room, which then will probably lead to an immediate pile-up of attacks of specials until it's a total disaster.
Last edited by Olde; 22 Jul, 2017 @ 3:02am
< >
Showing 1-7 of 7 comments
Per page: 1530 50