Forums  /  Left 4 Dead series  /  Left 4 Dead  /  Proposed changes to bring us in line with L4D2

Pixel walking, you mean on the fences around the Church? Yup. That part is actually by far the easiest, it's hard to fall off once you're practiced. The hard part is getting a Tank spawn so you can use setups to boost up to that part, and then the crouch hop to get to the saferoom's exit door and end the level. Granted I haven't tried it in L4D1 yet, maybe it's lucky and the last crouch hop is easier... I'm not saying it's reasonable to expect to execute this but the idea is out there ready to infuriate anybody that wants to try it. Heh.

Sounds good. Thanks for the input, Canterlott. I think I'm leaning on each campaign having its own IM section, that will probably look cleaner and more organized.

(Working on looking at your timing to see if it was 2:05, juggling a bunch of conversations ATM)

CanterlottCanterlott likes this. 

I'm fine with pretty much everything you suggested TheMaster.

Also @Deagle, should track all speedruns for a game. That's the point of the website.


Originally posted by Cyberdemon531Also @Deagle, should track all speedruns for a game. That's the point of the website.
Where does it say on the website that it's supposed to track all speedruns ever indiscriminately? I might as well speedrun a bunch of L4D2 mutations so that they can get added as more categories that virtually nobody cares about.


Alright, I think I'll move forward with the category shifting as planned. One is up now as an example of what it will look like, under Miscellaneous you'll see "No Mercy Individual Maps" with five subcategories. I will need some time to make these changes since I will have to enter runs again manually as I move them over along the way so stand by, and in this transition I ask everybody to please hold any submissions until it's complete.

As for tracking all speedruns for a game, that is true within reason. The communities decide what they want to run in a game and what makes the most sense as a category based on the content of the game. It isn't entirely black and white. Super Mario 64, for example, has the 100% category of 120-star, and other categories based on when you unlock each boss. Technically you may be able to say tracking all possible categories would add getting all stars in only Whomp's Fortress, another category for all stars in only Bob-Omb Battlefield, categories for getting specifically 19 stars, there are so many ways for it to go. That doesn't make the most sense for the game, though, so the community doesn't run it like that.

Categories can also be based on how different they are from one another, with two categories that are so close in gameplay sometimes discounting one or the other for lack of variety. That would be the case with mutations as many of them don't really change much, especially when it comes to speedrunning. Mutations would need to be picked far more carefully for what unique contribution they may bring to a speedrun that isn't there in a regular run of the game, which is something we could visit down the road if looking for something new to do, I'm not sure. That's getting off topic however and would be a discussion for another day.


I think the discussion is pretty much over but I want to vocalize my opinion anyway.

Having the Misc category for the map runs is great. That way you dont have 50 subcategories for a category
Then you can either do solo and coop subcategories for each category or do solo and coop categories with only the maps as subcategories. The first option would be better, in my opinion.

Having the version column would be interesting. Introduces the same distinction we have in L4D2.

In case I missed some other change, I'm okay with said change. As i was reading the thread I agreed with the ideas that you came up with, Master. Just may not remember all of them as of right now.

And about porting the splitter, I just need a copy of the game. The sale will be over soon.


I actually have every campaign as an individual map category under Miscellaneous as its own, and then the campaign's maps as subcategories within that. Since I unfortunately can't have a subcategory in a subcategory, Solo and Co-op distinction will be handled with a column. Otherwise, I would have to create a new category for every individual map and mark it as Miscellaneous, which will then make the misc dropdown ridiculously long, so this is probably the better way.

All of the Miscellaneous categories have been added now, once I move other stuff around and remove the extra players column (because it's applied to everything right now) it will be ready.


Wait a sec, why exactly did you duplicate the IL board into full game?


That is where Miscellaneous is. I am still in the process of setting everything up right now, I've just finished moving the individual maps to their new home in Miscellaneous and the other campaigns have not been moved over to Level Leaderboard yet.


well you can make it a misc il if you wanted?I dont even see a reason to mess with ILs at all really


Hmm. I'm still fairly new to how is laid out so I'm trying to work through it. Do you mean adding a Miscellaneous tab to entries in the Level Leaderboard and placing the individual maps within that? I see how I could do that, which is essentially what I was doing before except within the Level Leaderboard.

So the Level Leaderboard would look like No Mercy, with a Solo and Co-op tab for full runs, then a Miscellaneous tab holding Individual Maps with subcategories for each map in a Solo and Co-op version. Is that what you mean?

Edit: Having a Solo/Co-op column seems to work well with this layout, I'll switch to that method.


Seems I've worked that out just as well and this categorizes it a bit more cleanly. Yay, learning. Thanks for the heads up. I guess I'll need to move the runs over again... a little notification spam never hurt anybody, right?


Alright, there we go. Check out the Level Leaderboard. So much easier to manage now.

I'll work on moving the campaign times into the Level Leaderboard sometime tomorrow.


Reorganization is complete! Solo is now separated for all categories and open for business. Submissions are open again if anybody wants to get to it.

Next to do is rules and figuring out how to time the old runs. I just confirmed quickly so I can comment on it, removing loads in L4D1 does work just the same as L4D2 and the timing method will be changing to that. For now until I can get proper rules written in and everything, refer to the L4D2 load removal method for timing runs here.


"Next to do is rules and figuring out how to time the old runs. I just confirmed quickly so I can comment on it, removing loads in L4D1 does work just the same as L4D2 and the timing method will be changing to that."

So if I were to submit a new Original Campaigns run, per se, I would be able to submit my time without loads? I really want to run that category again, but I'm not running my games on my SSD at the moment so I doubt I'd be able to PB, if not. I already manually removed my loads from my other run and put the time in the description of the submission.

also by the way
on my profile it looks like i have the WRs for the original/all campaigns runs
you should probably rename the original/all campaigns categories to include solo in the name so it doesnt look like im better at this game than i really am :'^(

mariorules64mariorules64 likes this. 

Yup, time without loads as we're doing it in L4D2 will be the standard here, removing only when the loading icon appears up to the frame where HUD reappears. It's really odd how the current rules worked, that loading was included in all campaigns/original campaigns but taken out in the single campaign runs. Pretty inconsistent. This will apply to everything except for the individual maps category that is in each individual campaign on the level leaderboard.

For individual maps, that will be purely control gained/lost, which I currently need to figure out for the rules how to time reliably. Control gain is pretty easy, generally when you see first-person view (some campaigns are a little different, need to investigate), loss is the hard one because the only method I could find to know exactly when it triggers the end level sequence is the tick that the music starts to play. Not a problem if demos are required, just a little cumbersome to always have to examine the ticks to find the timing. Oh well.

Anyway, I've been holding back a bit since I haven't yet tried to work out what the other times are to adjust them yet, and I've also been streaming and all, not to mention wrecked by anxiety over AGDQ. I may be able to do something today with timing the other runs and rules after some real life stuff.

Yeah, I was thinking the same about the categories and solo. L4D2 was actually set up with the same naming scheme, only naming the Co-op category as such, and I was thinking of adding Solo both over there and here.

Oh, I also need to add a version column. There is only Newest and 1.0, right?


Shit, I think I might have timed my loadless time with gain/loss instead of the way L4D2 loads are done now, hold off on updating that then, I'll redo my time manually. While I'm at it, do you want me to retime all of the original campaigns runs, to save you some time?

Yeah, those are the only two versions for right now.

EDIT: My Original Campaigns loadless time is 52:59. I made sure to include menuing time, because I didn't use campaign-changing binds in that run.

EDIT 2: While I was retiming, I remembered that I started my timer on starting the campaign loading screen, not on control gain. (I was bad at running this game even back then lol) If I remove those 3 seconds, my new RTA time is 54:15. I'll change the times on my submission, just re-accept it whenever you get around to it. I'll start retiming the other runs now.

Screenshot 1
Screenshot 2


It's pretty easy to cut the frames up and grab the time, especially since I'm used to doing it, so it's alright. I should be able to get to at least some later today. Thank you for the offer still, I'll let you know if anything changes.

I actually see, while looking at runs that actually have videos that they do include loading screens so that's very good. It'll be easy to update those. I just don't understand how they were previously getting the times... the current first place No Mercy, for example, is submitted as 10:37, but when I time it based on the upper-left perspective (who appears to be hosting) I get 11:15.007. The end sequence from blurring to loading icon takes 8.5 seconds. If you add 8.5 x 4 to their submitted time, you still only get a time of 11:11... maybe I'm just missing something, I'm not sure yet. This is why we never submitted our times in the first place and was asking what to do. Weird. The video doesn't seem to be running slowly either because the end sequence, in my best estimation, is taking 8.5 seconds as it should, or at least extremely close to it based on when end music starts. Oh well, the new methods will take care of this confusion and create a new, easy standard to go by. Now, if I can't figure out a consistent point they took timing from, this will be a problem when I get to the runs that have no proof of their time at all which leads us to...

I am considering rejecting the runs that were uploaded and approved by a mod, bypassing the video requirement (which is every videoless run), and says "demos available upon request!" so that they can see it and upload these demos for others (I'll put up a video based on them personally if I need to). It doesn't seem right to have submitted the runs without any original video or even uploaded demos just because mods have that ability (I tried, I can bypass the video requirement), and because they bypassed this requirement I'm not sure I can adjust the times reliably enough. Any thoughts on this?


I'm fine with the rejections. Freezard's IM videos are on SDA already, so you could probably find an easy way to port them over if you wanted to go that route, as well.
When it comes to the All Campaigns run in 1h 12m 59s, there isn't a video for that one so I'm not sure how they'd show proof. The only video from those four runners I can find is from AGDQ 2013 but with a worse time, so I guess you could put that one in its place for now or something? idk ( (feel free to check out the comments for some cringe)

"I just don't understand how they were previously getting the times... the current first place No Mercy, for example, is submitted as 10:37, but when I time it based on the upper-left perspective (who appears to be hosting) I get 11:15.007."

I tried removing the loads correctly myself, not that I didn't believe you, but 40496 frames at 60 fps is 00:11:14:56, which still doesn't explain how they got their time of 11:38 on the video and their submission time of 10:37. I honestly have no clue. Maybe they meant 11:37, and it was just a typo?


Hmm... problem with Freezard's SDA videos is that they don't embed here properly. The best I can do there is include a link directly to the insane quality (ohhhh how encoding has changed) runs so that can be followed directly to the video.

As long as these mysterious demos exist it can be timed by them as well, although the All Campaigns run does kind of sound like they don't have even demos for that going by their description... well, I guess the AGDQ 2013 run could be used, sure, and hopefully they can produce something eventually to get the current run back up. If not, well... the time is very slow and could be very easily beaten again with a video, I guess. It's crazy, our AGDQ 2017 run is going to be a staggering 20 minutes faster than that one if everything goes well. We've come far.

I checked the No Mercy run again to confirm and came out with 11:15.007 again when based on the first frame being the first time you see first-person view. This is examining the video as downloaded directly from YouTube. If based on when the HUD starts showing like in level transitions the time is 11:14.974, figures it's a single frame from the next second! In this case HUD showing for time start looks accurate so I'll go with that, which would make this run 11:14. Avidemux doesn't show total frame count, but their video is 29.971fps so I'm not sure what you mean by 60fps. To time this, I am using the exact method shown in the load removal tutorial I made for L4D2, which I should probably copy to the guides here too. Maybe you can spot our differences?

My timing on the run was based purely off of the perspective of the first-loading player that's in the upper-left, removing a couple of extra frames where the loading icon is showing and black frames until the next level shows with a HUD. Even if it were a typo, 11:37 is even more out of whack... if I eventually manage to figure out how they did it I'll report back, but I have a feeling it may remain a mystery.

We can talk about it more on stream sometime if you'd like, where I could show you how I am examining the frames in real-time as well. It looks the same as what is shown in the video in that guide, though.

Seeing this, I timed your Original Campaigns run to see what I'd come out with and I have 53:00.100, so we have a little bit of variance. So close to the next second once again! I'll update your run to that time for now and we'll see where our differences are coming in.

By the way, I will probably be leaving the time as a single column, displaying only the loadless times. I actually plan to change L4D2 to this format as well but have been waiting until all of the runs there are updated first so it's easy to track what still needs to be done in the meantime considering there are far more runs to change than there are here. It'll look much cleaner to just have one column and then ensure all future runs are always loadless.