🔥 Yore driving, Deez roasted Nuts! 🔥
-
Early merge / late merge are not inherently wrong.
They are a symptom of the real problem, people not maintaining proper follow distance. This is the real problem and it's consistently a problem in every kind of merge. Any time traffic has to slow down to accommodate a merge, that's creating the traffic.
-
Listen, dude, it's not just me who thinks you're wrong. If you had actually read the thread you'd know that.
Since you obviously don't know that I dare say that you didn't read the thread. qed
Any time traffic has to slow down to accommodate a merge, that's creating the traffic.
And this does not even make sense.
-
@HardwareGeek said:
I'm willing to concede that late merging may, indeed, be the correct way to merge, but 99.984827% of the time it won't actually help the traffic flow. If traffic is heavy enough that Doing It Wrong™ will cause a traffic jam, Doing It Right™ won't keep traffic flowing smoothly, either, and if it's light enough that it flows smoothly through the restriction by Doing It Right™, Doing It Wrong™ probably isn't going to jam it up.
QFFT. That's exactly what I've been trying to say.
one jerk who disrupts the smooth flow can screw it up for the non-jerks.
I mean the morons who ... rush forward and try to merge in front of you
But you only do late merge when the traffic's congested
If I don't start looking until the end of the lane, because "late merging is better", I run a serious risk of not being able to maintain speed as I run out of lane.
all traffic rules should be based on gut instinct and not wanting to piss off unpredictable jerks instead of science and measurement.
Your entire argument of late merging is based on road rage.
Not based upon actually improving traffic flow.
I'm saying late or early merging doesn't matter, but the real problem is people not creating the space to merge.
If the space to merge existed people could early or late merge, in fact simultaneous merge is the ivory tower ideal and that's early and late merging happening at a constant consistent rate due to a slow merge process.
-
And this does not even make sense.
If I want to merge and there's not enough proper following distance in the lane next to me, but enough distance to just fit inbetween two cars.
I merge, and then all the cars have to suddenly and unexpectedly slow down to accommodate the merge, this slows traffic.
-
@Zecc said:
FTFY.Do you ever signal other drivers that you are forfeiting your right of way in the middle of the god damn road when you're going straight and are expected to not stop?
FTFY the answer is no as I live in a fuck-you-got-mine part of the planet, where everyone always drives fast, and there are no traffic jams around intersections, and where a situation where drivers may be give up their right of way to actually improve general traffic fluidity is something I've never seen in my life and am incapable of imagining..
I will now let this subject die. Find another topic to troll me about.
-
Right. And by which magic is this prevented by early merging?
-
early merging?
Early merge / late merge are not inherently wrong.
If you merge when there is already gaps in traffic then you do so without hindering traffic as much.
It's not about merging as soon as you know about a constriction. It's about merging when the is a sufficient gap at normal speed.
This is similar to early merging, but it is not FORCED early merging.
as soon as you have a chance once knowledge of a constriction
The ivory tower ideal is to stretch the zipper merge out as far as you can in order to merge seamlessly at the closest to speed limit as possible.
Not wait until you almost hit the cones and then cut someone off.
-
You have not really driven much, haven't you?
-
buses are required to stop, open their doors, listen for a train, close the doors, and then resume driving.
around here most railroad crossings actually have a barrier. Those that haven't are mostly freight only and will have signal lights and bells. I only encountered those around harbors. No buses on those roads.
-
You have not really driven much, haven't you?
If you mean by, I haven't driven much because I haven't seen how people behave...
Then you are misunderstanding what I'm saying.
Forced late merging only makes sense because people are assholes.
Forced early merging is also people being assholes.
You are creating this false dilemma of forced early vs forced late merging.
-
Those are nowhere near as annoying as the "I'm not going more than the limit in the passing lane" trolls.
Those are called hyper-miling Prius drivers. Oh wait, you said 'limit' - those do less...
-
I prefer the Dutch word: accordeon file, it works even better in English: accordeon jam. It's musical and sweet! Get me some!
-
Forced late merging only makes sense because people are assholes.
Newsflash: people are assholes.
-
If you mean by, I haven't driven much because I haven't seen how people behave...
Then you are misunderstanding what I'm saying.
Forced late merging only makes sense because people are assholes.
Forced early merging is also people being assholes.
You are creating this false dilemma of forced early vs forced late merging.
I'm not misunderstanding anything here. Late merging is superior to the other forms for two reasons:
a) Everyone knows the exact spot where merging is to take place and
b) It's making optimum use of the available road, leading to shorter queues
-
Late merging is superior
IFF traffic is completely backed up.
Before you get to that point, merging where possible is better than late or early merging.
Your assumption overlooks the fact that I've been in many late merging scenarios where the non-merging lane wouldn't let people in at all. And then people have to force their way in from a slower lane, thus slowing down traffic EVEN MORE!
Obviously any methodology is going to work better if people are compliant.
-
So? That will happen in your magical-mystery early-merging scenario as well. If those people are assholes enough to not let you merge - why in the hell will they let you merge earlier?
-
So? That will happen in your magical-mystery early-merging scenario as well. If those people are assholes enough to not let you merge - why in the hell will they let you merge earlier?
That's my point.
You are basing your scenario entirely on a worst-case situation where traffic is already backed up, and you want to prevent people from being assholes, when your solution does nothing to prevent people from being assholes.
-
Um, no. But have fun arguing with nobody about this!
-
That's not what you diagrammed.
-
This is forced late merging.
This is what I'm saying you should do. Create suitable follow distance and early/late merging is a non-problem.
-
That's not what you diagrammed.
I see you've tried to change the subject to my inability/laziness to make a perfect diagram.
Good attempt.
-
Perfect? It's not accurate at all if that's what you were trying to describe!
You showed stopping with no traffic in front or behind you to yield to the people who are supposed to be yielding.
-
Your diagram shows that if traffic is moving along everything is ok and if there is a jam then it is a pain in the ass.
Thanks!
-
around here most railroad crossings actually have a barrier. Those that haven't are mostly freight only and will have signal lights and bells. I only encountered those around harbors. No buses on those roads.The crossings I was referencing are tracks I have only ever seen used for freight. Every single one of them is equipped with a barrier, signal lights, and bells. The tracks have been there for ages, but it is an industrial area (probably because of the tracks). The buses are there because it is close to a county detention facility and the aforementioned bus depot (actually, the tracks are in between the detention facility and the depot).
As for why the buses are required to follow this procedure: ¯\_(ツ)_/¯. School buses and some trucking companies follow the same procedure. It's probably so that the drivers don't get in the habit of crossing tracks without checking, which might possibly lead to them getting clobbered at an unmarked crossing.
-
First, you really didn't read the thread. 50% of the people you quoted and thought (I guess) were on your side aren't.
Second, read the studies.
Third please paint pretty pictures WITH THE SAME AMOUNT OF TRAFFIC.
If people are assholes and/or there is a lot of traffic, there will be jams, because the gaps aren't big enough.
If people are compliant and/or there is just a little traffic, it will flow, because the gaps are big enough.In both cases late merging will make better use of the available road space.
Therefore, it's SCIENTIFICALLY PROVEN to be better.I don't know what's so hard about this.
-
With an ending e telling you how to pronounce the vowel in the midde (glade vs. glad) neither of you deserve to complain about how words are spelled.
As opposed to changing the gender of the word/what it represents? (Fiancé/fiancée).
English doesn't have a monopoly on that sort of thing. Indeed the likely reason (which I've quoted before) is aptly explained by James Nicoll
The problem with defending the purity of the English language is that English is about as pure as a cribhouse whore. We don't just borrow words; on occasion, English has pursued other languages down alleyways to beat them unconscious and riffle their pockets for new vocabulary.
-
b) It's making optimum use of the available road, leading to shorter queues
The nice thing about the shorter queues is that they are less likely to back up past intersections. when congestion backs up past an intersection, then you have assholes stopping in the intersection, leading to traffic on multiple streets being obstructed.
-
WITH THE SAME AMOUNT OF TRAFFIC
It's not the number of cars that differ, it's the amount of space they leave between them. Xaade's point is the same as mine: people not leaving enough space to merge are what causes jams, not early vs late. Early vs late is, at best, making the best of a bad situation; it can't address the real problem, which is assholes not letting you in ever.
-
This is what I'm saying you should do. Create suitable follow distance and early/late merging is a non-problem.
That second one isn't "early/late merging". It's just proper merging.
-
You showed stopping with no traffic in front or behind you to yield to the people who are supposed to be yielding.
I did not draw the whole city, no. Sorry about that. I've amended my post, HTH, HAND, FOAD.
-
Really, it's sort of a prisoner's dilemma: if I late merge, and the assholes in the lane next to me leave space for me to late merge, then we both go on with our day with no trouble. If I late merge, but the assholes don't let me in, I can't get in unless someone stops to let me go by, because I've run out of room in my lane and still can't find an opening. If I early merge, and they do let me in, science has shown that traffic ends up slower, but I barely feel an effect. If I early merge, and they don't let me in, odds are higher that I'll have room to get over.
So my local maximum is early merging, while the global maximum is late merging + people letting me in. So do I trust local drivers to let me in? Nope.
Paint diagram, where my expected utility is in red and green and theirs is purple and blue:
-
In both cases late merging will make better use of the available road space.
Actually, the Minnesota study stated that early merging is preferable in the case that traffic is light enough that it is moving at or near normal speed, which is exactly the case @xaade's diagram shows. However, in the case of heavy traffic, @xaade's late merge diagram applies whether the merge is early or late; the only difference is the diagram doesn't show the empty, unused lane that exists beyond the merge point in the case of early merge. Early vs. late doesn't really change the traffic flow, just the distance the backup extends prior to the actual lane restriction.
-
-
That second one isn't "early/late merging". It's just proper merging.
That's what I've been saying.
Rhywden seems stuck on stuffing some strawman argument for early-merging that I didn't make.
Early vs late is, at best, making the best of a bad situation; it can't address the real problem, which is assholes not letting you in ever.
Thank you.
Actually, the Minnesota study stated that early merging is preferable in the case that traffic is light enough that it is moving at or near normal speed, which is exactly the case @xaade's diagram shows. However, in the case of heavy traffic, @xaade's late merge diagram applies whether the merge is early or late; the only difference is the diagram doesn't show the empty, unused lane that exists beyond the merge point in the case of early merge. Early vs. late doesn't really change the traffic flow, just the distance the backup extends prior to the actual lane restriction.
Yes.
Find a highway without much traffic? GENIOUS
If people were to drive properly, the highway traffic isn't the problem.
The problem is that my suggestion doesn't solve the traffic leaving the highway.
You can put all those cars on the highway as a dispensary to the surrounding streets, or you can put all that traffic through at a higher rate with proper gaps. Same amount of cars make it through the highway, speed is the difference.
If you're finding that traffic is backing up on your highways always even with proper drivers, then your congestion doesn't come from the highway, it comes from bad flow off the highway.
In cases where there is more cars than road, nothing you try will help.
-
It's not the number of cars that differ, it's the amount of space they leave between them.
But the number of cars, the space between them and the traffic flow are all related.
Xaade's point is the same as mine: people not leaving enough space to merge are what causes jams, not early vs late.
There's no way around that when there are enough cars. And when there aren't it probably doesn't matter very much.
-
So do I trust local drivers to let me in? Nope.
The Minnesota study showed that late-merge worked well, at least in part, specifically because they were training people on how to do it correctly by means of signs (in some cases dynamically changing the merge point based on traffic speed) and a PR campaign. Without the USE BOTH LANES and TAKE TURNS signs, it would probably work less well. As other people have stated, (re)training drivers is an important part of making late-merge work optimally.
-
If people were to drive properly, the highway traffic isn't the problem.
I guess that's true if you live in a place where there's more road than cars.If you're finding that traffic is backing up on your highways always even with proper drivers, then your congestion doesn't come from the highway, it comes from bad flow off the highway.
Yes, too many cars for fast driving. Duh.
In cases where there is more cars than road, nothing you try will help
However, there are still things you can do that will make stuff worse, like merging early.
-
However, there are still things you can do that will make stuff worse, like merging early.
Do you merge at the last possible second on an onramp?
You are more than likely going to find yourself needing to stop, back up the onramp traffic, and have to wait for traffic on the highway to slow down or cut someone off.
You come to speed with the merging lane and merge where there is a gap.
If I'm in the constriction lane, and I'm following a gap to the lane accepting the merge, why would I wait until I almost hit the cones to merge in?
In what scenario will waiting until the cones change anything about the positions of the cars to the front and back of me in the lane accepting the merge?
At best I can even be a proper driver and create the gap necessary for the car in front of me to merge in safely.
-
Do you merge at the last possible second on an onramp?
Depends. If the traffic I'm merging into is slow / stopped, then yes, of course. If not, then no, of course.
You are more than likely going to find yourself needing to stop, back up the onramp traffic, and have to wait for traffic on the highway to slow down or cut someone off.
Fuck. Stop being so stupid. Nonononono.
You come to speed with the merging lane and merge where there is a gap.
No shit. If there's sufficient speed then there's no problem. Why do you keep bringing this up?
If I'm in the constriction lane, and I'm following a gap to the lane accepting the merge, why would I wait until I almost hit the cones to merge in?
USE THE ENTIRE ROAD THAT'S THERE. DON'T MAKE THE BACKUP WORSE.
-
USE THE ENTIRE ROAD THAT'S THERE. DON'T MAKE THE BACKUP WORSE.
Only if there are no gaps.
You keep missing that there is an intelligent decision to make here, neither hard-fast rule is superior.
It's stupid to drive parallel to a gap all the way to the cones.
If you cannot find a gap, you drive all the way to the cones and wait for one.
-
-
Are you suggesting that someone drive parallel to a gap all the way to the cones?
-
No, I never suggested that.
-
Ok, then I misunderstood.
-
Cool. Perfect zippering in both cases!
-
I missed a few posts once the discussion between early vs late merging was brought up and I skipped ahead. Since that's still going on, has anyone mentioned already that what causes the problems isn't late vs early merging, but the fact that there's an obstruction limiting the road's capacity in the first place? Merging strategy isn't going to magically alter the fact that a road expected to handle x cars/hour is now only able to handle x/2 cars/hour.
Simple thought experiment before diving deeper into it: Early merging is equivalent to a longer constriction with late merging, as it pushes the merge point back. However, the length of the constriction is irrelevant to the flow of traffic. The constriction could be 100m or 1km long and the effect on the traffic coming in is the same: The flow of two lanes has to squeeze through a single lane. Thus, early vs late merging is a retarded discussion.
Before you flame, consider the following: At full capacity, a lane would have cars going at the posted speed limit, separated by a safe breaking distance. The faster the cars are moving, the bigger the safety gap needs to be.
Now, consider a fully loaded lane that a car wants to merge into. In order for the car to merge, it needs to have space. There are two ways that can happen: Every car ahead of the merge point has to speed up for a few seconds above the speed limit, or every car behind the merge point has to slow down for a few seconds.
The first can be discarded outright. It would require people to react to something happening several cars behind them, and for cars to enter the safety gap ahead of them, which is inherently dangerous. So it's a magical and dangerous option.
The alternative is for the car behind the merge point to slow down a bit, increasing the gap until there's enough room for the car merging in. Since he slowed down, the car behind him entered the safety gap, and will have to slow down himself until the car ahead has resumed normal speed long enough to recreate the safety gap. And so on down the lane.
This is the "shockwave". It is not a phenomenon caused by poor drivers, but simply the result of requiring more space on the lane than is currently available. How far back can the shockwave go? Until it meets a gap bigger than the minimum safety distance. That is, until it reaches a point where the road is at less than full capacity. The car at that point will see that his distance to the next car diminished, but as long as the distance doesn't fall below the minimum safe distance, he won't need to slow down himself. At that point, the shockwave finally dissipates.
That is with just one car merging into a fully loaded lane. Each car merging will require this same dynamic of the cars behind slowing down, and the time they have to slow down for increases linearly with the number of cars that need to merge. The more general rule is that as long as the constriction doesn't limit the road capacity below what is required at that point, there won't be congestion. In the case of a two lane road being limited to a one lane road, that means the road has to be at most at half capacity. Meaning every car should be, on average, two minimum safe distances behind the car ahead.
-
Meaning every car should have be, on average, two minimum safe distances behind the car ahead.
In a real world situation, increasing the safe distance to a proper follow distance would help accommodate that, as the constriction may not be long enough to impact cars for too long.
But it's a pointless conversation as well because people don't even follow minimum safety distance either, which is why most of these constrictions end up congesting traffic.
So, the real reason we're having this late/early discussion is because people are bad drivers, which was the whole thing I was going for.
Early/late isn't a meaningful discussion.
-
Early merging is equivalent to a longer constriction with late merging, as it pushes the merge point back.
IF everybody magically agrees on when/where to merge, yes.
With late merging there is a defined place where merging should occur.
Neither will work with 100% assholes in the continuing lane.However, the length of the constriction is irrelevant to the flow of traffic. The constriction could be 100m or 1km long and the effect on the traffic coming in is the same
My gut feeling says no. Longer single-lane area = more opportunities for shockwaves. And early merging still leaves road unused, thus lowering the road's total capacity.The flow of two lanes has to squeeze through a single lane. Thus, early vs late merging is a retarded discussion.
IF everybody predictably behaves the same way, yes. How often does that happen in your world?
-
But it's a pointless conversation as well because people don't even follow minimum safety distance either, which is why most of these constrictions end up congesting traffic.
Minimum safety distance won't help, because if someone (at whatever point) merges into your safety distance, you have to slow down to build it back up. Ergo: Shockwave.
If you double your safety distance so you won't have to eventually brake for someone to merge in, you lower the road's capacity. That really doesn't scale well.So what you "discovered" is that congestion won't occur if there isn't enough traffic. Genius!
Again: If traffic is light, late merging gives you a defined place where to merge, although it probably doesn't matter. The traffic will flow anyway.
If traffic is heavy, congestion will occur anyway. Late merging STILL gives you a defined place where to merge AND uses all the available road space.
-
In a real world situation, increasing the safe distance to a proper follow distance would help accommodate that, as the constriction may not be long enough to impact cars for too long.
That's a consequence of road over-utilization. An interesting fact I realized while writing is that a road's capacity is actually a constant, independent of the posted speed limit. The safe distance between cars is determined by human reflexes and how comfortable they feel with cars being close at different speeds. As the speed increases, the distance between cars has to increase proportionally.
Increasing the speed limit lets you cover longer distances, but when it comes to capacity (cars/hour), you need to add lanes to increase it.
My gut feeling says no.
Gut feelings are a poor substitute for reasoned arguments.Longer single-lane area = more opportunities for shockwaves.
Once the cars are inside the constriction, other cars won't be merging in. They can go at the speed limit and they don't need to slow down any more. Even if the constriction never ends, they're going to be cruising comfortably.If traffic is heavy, congestion will occur anyway. Late merging STILL gives you a defined place where to merge AND uses all the available road space.
All that matters is how many cars are coming out of the constriction. If more cars are trying to go in than can come out, no merging strategy is going to make a meaningful impact.
