Forum Settings
       
Reply To Thread

Driving habitsFollow

#77 Aug 26 2014 at 10:21 AM Rating: Good
*******
50,767 posts
gbaji wrote:
That doesn't mean it's wrong or anything, but by no definition is that a source (at least in this context)
The context being that it wasn't your anecdote, of course.
____________________________
George Carlin wrote:
I think it’s the duty of the comedian to find out where the line is drawn and cross it deliberately.
#78 Aug 26 2014 at 10:35 AM Rating: Good
******
27,272 posts
Sir Xsarus wrote:
You here seemed to think that because I posted I was saying you were wrong about everything.
To be fair, that's what most people say when they reply to Gbaji.
#79 Aug 26 2014 at 10:45 AM Rating: Excellent
Meat Popsicle
*****
13,666 posts
His Excellency Aethien wrote:
Sir Xsarus wrote:
You here seemed to think that because I posted I was saying you were wrong about everything.
To be fair, that's what most people say when they reply to Gbaji.
There's something amusing about watching two people who agree arguing though.

Smiley: popcorn
____________________________
That monster in the mirror, he just might be you. -Grover
#80 Aug 26 2014 at 3:44 PM Rating: Decent
Encyclopedia
******
35,568 posts
Sir Xsarus wrote:
The idea of closing a gap to prevent a light from changing is most likely only going to occur in a situation where a light has quite recently changed on a side street with a fair amount of traffic backed up behind it.


A high percentage of lights where I live are configured to do this. It's rare to see a light that *doesn't* exhibit this behavior. It's not just to deal with one scenario, but to more intelligently manage the lights in suburban areas, where the intersections are "busy" all the time, and traffic patterns may change in ways that aren't easily correlated just to time of day (as they may be in a more grid like downtown configuration).

The point of this type of system is to ensure that the traffic lined up and waiting for the light can go (again, barring someone not noticing the light changing), but that once that first group has gone through, it will immediately change if the following conditions are met:

1. There is traffic on any other sensors waiting at their light (which is going to nearly always be true).

2. There is a gap in traffic on the light that is currently green.

Additionally, once some predetermined amount of maximum-green time has passed, the light will change.

This is how nearly every single light near where I live works. Again, maybe it's different in SF, but I can see this behavior every single day on lights everywhere I drive. The fact that it's so ubiquitous around here is why I'm so surprised by the "that can't be how they work" response I seem to be getting. I assumed that lights everywhere worked like this. Because it's a much smarter way to do things than to just have timers. Why bother with sensors in the road if you don't actually use them to control the lights on the fly?


Quote:
This is the scenario these sensors are looking for and are put in place to ameliorate. From my perspective, this means that most lights won't behave this way...


I think your first assumption leads you to that conclusion. Eliminate the assumption and the conclusion no longer makes sense.

Quote:
...and I still find it dangerous to accelerate in case you can close the gap enough. It's setting you up to not be able to stop in time if the light changes, or if the person in front of you gets a yellow an decides to stop when he may have been able to go through, you're accelerating towards him, and it puts you at greater risk.


If you are close enough to the other person that this is remotely dangerous, then you are close enough that you don't need to speed up in the first place. I'll grant you that if you wait until the last second and then accelerate to a high rate of speed to close that gap, you're being dangerous. But my whole point from the start of this discussion was the idea of looking ahead down the road and making easy minor adjustments to speed in order to avoid problems later on. I'm talking about noticing ahead of time that this condition is going to happen and adjusting your speed to close that future gap when it's completely safe to do so. If you're waiting until the last car in a line has just hit the sensor and then flooring it to get on the sensor before the light trips, then yes, you are being dangerous. But that's not what I'm talking about.

Quote:
why are you so far behind to begin with?


Um... Because some of us live in a world where every car on the road isn't gridlocked with the others. You turn onto a main street from a smaller side street. A line of traffic 5 cars deep has just started moving at a green light 300 (or 200, or whatever distance might create a gap) feet ahead. You're free to pick any speed to travel towards that light, but it makes sense to pick the one that puts you right behind the last car in that line at the correct distance so that you form into that line. What I'm talking about is simply being aware of setting your speed properly to avoid leaving a gap. I see lots of people who seem oblivious to the fact that the light will change if they don't close that gap and they'll slowly accelerate down the road and get stuck (or get the person behind them stuck). Just accelerating a bit faster would make all the difference (and is completely safe)

I'm not talking about nutty speeds here. I'm talking about the difference between a leisurely stroll and a brisk walk. Being aware of what's going on ahead of you and adjusting your speed just a little bit can make a big difference.


Quote:
Do you have to speed to close the gap? Given that you were most likely stopped, there shouldn't really be a gap.


You've never turned onto another street? You've never gotten off an intersection and onto a surface street? There are lots of ways you'll find yourself looking at the tail end of a line of cars that are some distance ahead of you. Happens all the time.

Quote:
The scenario is where you're coming up to an intersection after most of the traffic has cleared and you accelerate hoping to get through before it changes. That's increasing your risk.


If you wait until the last second, sure. But my whole point is that drivers should recognize this is going to happen ahead of time and adjust their speed/distance well ahead of time when it's still safe. Just like with the scenario I mentioned earlier with the friend who only needed to be aware that the car that just changed lanes might do it again and that it would be prudent to adjust her speed so that if that driver did so, it wouldn't be right when she was next to it. Same deal here. Pay attention. Look down the road. Recognize common driving patterns and their outcomes and make minor adjustments to what you are doing so as to minimize the bad outcomes and maximize the good ones.

I'm not talking about rocket science here. Just a bit of observation and forward thinking.

Quote:
I don't know if asking the person who knows exactly what the system does and has helped design it is an anecdote.


The information you get from an expert is a source for you. It's an anecdote when you repeat that information to a third party. Because I don't know the source and didn't get the information from said source. My "source" is *you*. And you're just telling me something you heard from someone else. A someone else that I have no means to get information from directly myself.

Quote:
If you asked me how something I wrote worked, I'd think I'd be the authoritative source.


Yes. Because you wrote it and thus are the expert on it. But if I asked you about something you wrote and then told someone else what you said or meant, that would no longer be a source. It only becomes a source if I point them to you, and they are able to read what you wrote directly.

Quote:
I mean you don't have to believe me, but it's not like I asked some random person, or I mentioned talking with a friend who had read an article or something. Take it or leave it, it is still true.


Sure. But from my perspective you did ask some random person. I don't know this expert. You do. Surely you can see how I have no reason to put any more weight in what he said than what you say, because from my perspective you are the one saying it. Point me at this experts blog, or an article he wrote on traffic signals, complete with his credentials as an expert in the field, and now you are providing me with a source. Just saying "I know a guy who said..." doesn't count.


Let me point out, again, that I'm not discounting at all what your friend has to say. I'm just countering that while maybe such lights are uncommon in SF, they are very common in SD. As I've said several times, I see this behavior nearly every single day. It's surprising to me that so many people either haven't seen this, or haven't paid enough attention to notice it. Maybe I'm just more observant than most (actually, I know that I am), but I honestly thought this was something more people would have noticed.

Edited, Aug 26th 2014 2:50pm by gbaji
____________________________
King Nobby wrote:
More words please
#81 Aug 26 2014 at 6:39 PM Rating: Excellent
*****
10,601 posts
Holy **** you repeated yourself a lot there. Wow.

It's very likely that most of the lights you're observing don't actually work that way, while some do, and you're noticing patterns that aren't there. This is very common by the way, we try to pattern everything, and it's increased if we think there is a pattern.

Edited, Aug 26th 2014 7:40pm by Xsarus
____________________________
01001001 00100000 01001100 01001001 01001011 01000101 00100000 01000011 01000001 01001011 01000101
You'll always be stupid, you'll just be stupid with more information in your brain
Forum FAQ
#82 Aug 26 2014 at 6:54 PM Rating: Good
Worst. Title. Ever!
*****
17,302 posts
Most of the lights in my area run on timers. But they are sequenced together to control the flow of traffic.

My route home as three lights.

If I hit the first light on a red, the next light can be driven thru before it turns red as long as you get up to the speed limit relatively quickly. If you do that you can get about 2-3 cars through the second light before it turns yellow. Then you'll reach the third light just as the left hand turn (my turn) turns green. So if everything goes right, I can make it through all the lights without hitting a red (or only hitting the first red).

Today I got up behind a person in line at the first light. It turns green, and the person is driving 25-30 mph. Speed limit is 45. So I get caught on the second light turning red. Then the third light's left hand lane turns green, I'm still stuck at the second light waiting. Then it goes back to red. Still Stuck at the second light. Then my light finally turns green, but the third light turns green again as I'm approaching, still stuck behind the slow driver. The third light turns red just as I approach.

So because of that slow driver I missed two light cycles on my usual turn...
____________________________
Can't sleep, clown will eat me.
#83 Aug 26 2014 at 7:10 PM Rating: Excellent
Soulless Internet Tiger
******
35,474 posts
The horror!!! Smiley: mad
____________________________
Donate. One day it could be your family.


An invasion of armies can be resisted, but not an idea whose time has come. Victor Hugo

#84 Aug 26 2014 at 7:18 PM Rating: Good
Worst. Title. Ever!
*****
17,302 posts
Uglysasquatch wrote:
The horror!!! Smiley: mad


Needless traffic backups at an already potentially busy area. Nothing like backing up rush hour traffic because you are driving 15-20 miles under the speed limit.

Luckily most of the summer vacationers have left. Usually around the time I get out of work those three lights have traffic backed up all the way through them and into the 4th intersection behind, because it's also the main drag leaving the beach. In those cases you usually sit through one or two traffic light cycles unable to move because there is no where to go.
____________________________
Can't sleep, clown will eat me.
#85 Aug 26 2014 at 8:31 PM Rating: Decent
Encyclopedia
******
35,568 posts
Sir Xsarus wrote:
Holy **** you repeated yourself a lot there. Wow.

It's very likely that most of the lights you're observing don't actually work that way, while some do, and you're noticing patterns that aren't there. This is very common by the way, we try to pattern everything, and it's increased if we think there is a pattern.


Sure. I'm sure every light doesn't work this way. But if it does, then my advice is good advice. And if it doesn't, then it doesn't matter. So it's a good idea to follow my advice and not leave large gaps between you and the car in front of you. There's no harm at all to do this, and potential harm to not do it. So...?


I'll also point out that the degree to which lights around here are programmed is absurd (and really obvious if you pay any attention at all). Where it's really obvious is left hand signals. Most of our signal controlled intersections have protected left turn lanes. And they tend to have some interesting logic to them. They will change behavior based on a combination of factors.

Consider a standard 4 way intersection:

If all the lanes in both directions on an axis are occupied (sensors tripped), then when the light cycles, both left hand lanes will turn green first, then both directions going straight will, then the light will cycle to the other axis. If one direction has folks in the left turn lane, then the light will first turn green for that left turn lane and the straight going the same direction. After a while the left turn light will turn red, the straight will stay green, and the straight going the other direction will go green (so both straights are green). If a car lands on the left turn sensor for the initially empty side (the one that turned green for the straight second) early enough in this cycle, then the straight that turned green first will turn red while the second straight will stay green and the left turn for that direction will turn green. Then, finally, the light will cycle to the other direction.


It's trivially easy to see that the light behavior changes on the fly based on which lanes sensors are tripped and when. If you don't get on that left turn lane's sensor early enough, you wont get a green light at all during that cycle. If you can get on it early enough (typically before the cross traffic light turns yellow, then your turn light will activate. I'll also point out that you can clearly see the "measuring gap" logic here as well. In the case where both left turn lights are tripped and turn green (so both straights stay red), if one left turn lanes traffic leaves a gap, that light will turn red, allowing the straight in the other direction to turn green while still keeping the left turn light going in the other direction green. That light will stay green for a much longer period of time (until there's a gap, or some maximum time limit is reached) and only then will it turn red and allow the straight traffic going the other way to proceed.

You can literally sit at the same intersection and watch this behavior every single light cycle. BTW, the same thing happens on a major street with a less busy side street. Say traffic is green going in both directions (north/south) and cars are driving by. A car pulls into the left turn lane southbound. Another car pulls into the left turn lane northbound. It does not matter which one pulls into their lane first. If a gap appears in the southbound traffic first, then the southbound straight will turn red and allow the northbound left to go. The northbound straight will only turn red once there's a gap (or again, some maximum time limit is reached which can be quite a bit of time for some intersections). It's really hard to not notice that the second a larger then normal gap appears, the light changes and allows the southbound left to go.


I've actually seen a cycle within a cycle happen in this case. The northbound straight will have a gap and the southbound left will get a green, while the northbound left is still waiting. Traffic will appear in the northbound straight, and a gap appears in the southbound left. It'll turn that light red and the northbound straight green again. All while the northbound left is still waiting. Then, the second a gap appears in the southbound straight traffic, that lane will finally turn red and the northbound left will get a green. Obviously, there are still max timers at work, but they can be several minutes in length. Plenty of time to witness a couple other lights change while waiting.

I'm not saying this because it's something I've seen once or twice. I see this behavior constantly. The behavior of many (most?) of the lights around where I live cannot be explained by anything other than sensors and programming acting the way I describe. It's not random. It's very very predictable.
____________________________
King Nobby wrote:
More words please
#86 Aug 26 2014 at 9:51 PM Rating: Excellent
*****
10,601 posts
Smiley: dubious
____________________________
01001001 00100000 01001100 01001001 01001011 01000101 00100000 01000011 01000001 01001011 01000101
You'll always be stupid, you'll just be stupid with more information in your brain
Forum FAQ
#87 Aug 27 2014 at 2:08 PM Rating: Decent
Encyclopedia
******
35,568 posts
Sir Xsarus wrote:
Smiley: dubious


Wait. What? You're not even going to attempt to argue with me? But... I demand an argument! This is just unacceptable!
____________________________
King Nobby wrote:
More words please
#88 Aug 27 2014 at 3:15 PM Rating: Excellent
*****
10,601 posts
Tell you what, if I ever have a chance to talk to one of the people who works with the light system in San Diego, I'll find out everything and let you know.

Smiley: nod
____________________________
01001001 00100000 01001100 01001001 01001011 01000101 00100000 01000011 01000001 01001011 01000101
You'll always be stupid, you'll just be stupid with more information in your brain
Forum FAQ
#89 Aug 27 2014 at 3:30 PM Rating: Good
GBATE!! Never saw it coming
Avatar
****
9,957 posts
This thread is NIXNOT.
____________________________
remorajunbao wrote:
One day I'm going to fly to Canada and open the curtains in your office.

#90 Aug 27 2014 at 3:54 PM Rating: Decent
Encyclopedia
******
35,568 posts
Sir Xsarus wrote:
Tell you what, if I ever have a chance to talk to one of the people who works with the light system in San Diego, I'll find out everything and let you know.

Smiley: nod


Sure, and in the meantime, I'll just continue driving around making basic observations and adjusting my behavior accordingly.

Funny fact: Today while driving to work, I made a right hand turn onto the main street, went through one light, then approached the next where I needed to make a left turn. There were a couple cars already waiting there about a block ahead of me. The left turn light went green, and those cars went through (as well as the left turn going the other direction). I was still roughly a block away and watched as the left turn arrow immediately turned yellow once the two cars at the light went through and no others were behind them. As I approached the now red left turn lane and parked, I noted that the left turn going the other direction was still green and allowing the long line of cars through. Meanwhile the light going straight in the other direction was now green and letting cars through as well. Then I noted that just a couple seconds after that the left turn lane going the other direction cleared of cars, the light going straight in my direction (which I was now parked next to) turned green. And then I thought of you. Smiley: smile

I see this literally every single day. It's how the lights work. You can wait to talk to an expert. I'll just look around me and arrive at my own conclusions.
____________________________
King Nobby wrote:
More words please
#91 Aug 27 2014 at 8:55 PM Rating: Excellent
Avatar
******
29,919 posts
if you guys ever want to play around with intersection modeling and light timing, if you can find a trial copy of PTV Vissim, thats what most of the DOT's on the left coast anyways use http://vision-traffic.ptvgroup.com/en-us/products/ptv-vissim/ Those signal modeling people who know how to use that software effectivly are able to bill some seriously huge rates on consulting ($130 / hour plus). The hilarious thing is that most of the ones that were hired for a certain river crossing bridge project that was cursed and then died a horrible death had no clue how to use the program in the first place and were reliant on much lower paid IT staff to show them how to use it. If I had the traffic engineering degree to go with my knowledge of that particular program I could probably make alot more money. Sadly my degree is in professional sneakynessism instead.
____________________________
Arch Duke Kaolian Drachensborn, lvl 95 Ranger, Unrest Server
Tech support forum | FAQ (Support) | Mobile Zam: http://m.zam.com (Premium only)
Forum Rules
#92 Aug 27 2014 at 9:06 PM Rating: Excellent
*****
10,601 posts
$130 an hour? What are they peasants?
____________________________
01001001 00100000 01001100 01001001 01001011 01000101 00100000 01000011 01000001 01001011 01000101
You'll always be stupid, you'll just be stupid with more information in your brain
Forum FAQ
#93 Aug 27 2014 at 9:58 PM Rating: Good
Skelly Poker Since 2008
*****
16,781 posts
After gbaji traffic light logic, you should all agree that traffic circles are the way to go.
____________________________
Alma wrote:
I lost my post
#94 Aug 27 2014 at 11:09 PM Rating: Excellent
Avatar
******
29,919 posts
Did you know the average 4 way traffic installation costs between $250,000 and $1,000,000? and will consume even with LED lights a couple thousand dollars of electricity every year, and another 10 hours of employee labor on average in maintenance? Did you know the average traffic circle sign costs $50 and averages $0.00 electricity every year and half an hour of maintenace? There's a reason the DOT's of the world keep pushing traffic circles even thoughe everyone hates them...
____________________________
Arch Duke Kaolian Drachensborn, lvl 95 Ranger, Unrest Server
Tech support forum | FAQ (Support) | Mobile Zam: http://m.zam.com (Premium only)
Forum Rules
#95 Aug 27 2014 at 11:20 PM Rating: Good
Citizen's Arrest!
******
29,527 posts
Dread Lörd Kaolian wrote:
Did you know the average 4 way traffic installation costs between $250,000 and $1,000,000?
Maybe they should stop making the parts out of gold-plated rubies because that's the only way my brain can figure out a justification for that cost.


Edited, Aug 27th 2014 11:29pm by Poldaran
#96 Aug 28 2014 at 6:08 PM Rating: Good
****
7,861 posts
The One and Only Poldaran wrote:
Dread Lörd Kaolian wrote:
Did you know the average 4 way traffic installation costs between $250,000 and $1,000,000?
Maybe they should stop making the parts out of gold-plated rubies because that's the only way my brain can figure out a justification for that cost.

Union labor would be my guess.
____________________________
People don't like to be meddled with. We tell them what to do, what to think, don't run, don't walk. We're in their homes and in their heads and we haven't the right. We're meddlesome. ~River Tam

Sedao
#97 Aug 28 2014 at 7:25 PM Rating: Decent
Encyclopedia
******
35,568 posts
Dread Lörd Kaolian wrote:
Did you know the average 4 way traffic installation costs between $250,000 and $1,000,000? and will consume even with LED lights a couple thousand dollars of electricity every year, and another 10 hours of employee labor on average in maintenance? Did you know the average traffic circle sign costs $50 and averages $0.00 electricity every year and half an hour of maintenace? There's a reason the DOT's of the world keep pushing traffic circles even thoughe everyone hates them...


That's kinda the highball cost range for intersections though, and includes a lot of costs related to traffic engineer/study time rather than just the physical cost to construct (and you'd have to do the same sort of studies for traffic circles). It's also an unfair comparison, because the correct comparison to a traffic circle is a set of stop signs, not traffic lights. Traffic circles do not provide any form of traffic control (just as stop signs do not). The need to spend the extra money for traffic lights is based on assessment of accident reduction and gridlock prevention. And that extra cost is generally more than worth it.
____________________________
King Nobby wrote:
More words please
#98 Aug 28 2014 at 9:14 PM Rating: Excellent
Avatar
******
29,919 posts
gbaji wrote:


That's kinda the highball cost range for intersections though, and includes a lot of costs related to traffic engineer/study time rather than just the physical cost to construct (and you'd have to do the same sort of studies for traffic circles). It's also an unfair comparison, because the correct comparison to a traffic circle is a set of stop signs, not traffic lights. Traffic circles do not provide any form of traffic control (just as stop signs do not). The need to spend the extra money for traffic lights is based on assessment of accident reduction and gridlock prevention. And that extra cost is generally more than worth it.


Signal cabinet with pedestal: $15,000. Primary signal controller $15,000 for a Naztec unit, Secondary failover controller $15,000. Simple power relay and basic battery backup unit with deep cycle marine batteries and transformer, $10,000 Signal arms for 4 lanes with pedistal $15,000 each *4 = $60000. Signal head units $950 each * 8 = $7600, Induction loop units *8 installed before pavement $950 each = $7600, 4x Infrared emergancy vehicle signal actuators $950 ea * 4 = $3800, $2,000 for lane striping, non skid reflective stop bars, crosswalks, etc. $1,000 * 4 for the walk / don't walk indicators and buttons, another $20,000 for the 4 corner crosswalk islands with ADA approved bumpy plastic inserts and retrofit curbs which after sales tax is about $174,524 for the parts alone. Add another $50k if you are paving the intersection. It usually takes a 10 man crew 3 8 hour days to do the installation for another $9k in labor, and the contractor usually makes about 16% profit on the job. So thats around $181,000 without the paving, $230,000 with, $250,000 if you add in the equipemnt and markings for a left turn pocket. Thats for an intersection that meets current federal highways at-grade intersection design specifications and ADA standards. it also includes none of the costs related to the traffic engineer study/time, design time, public outreach meeting costs. etc. which you would have to do for a traffic circle as well, which is why i left them off. And traffic circles do provide passive flow controll by slowing down inbound traffic and allowing more flow in the direction of whichever lane has higher traffic entering and leaving the circle just because of yield rightaway. The argument, and I feel it is a stupid but slightly valid one, is that traffic circles allow more traffic flow in very underused rural areas when compared with stop signs and will reduce accidents because traffic remains in motion. The reality is that people are too stupid to drive around traffic circles correctly most of the time, and don't know what a yeild sign is because it has too many letters to read.

So yeah, No. I kind of know a little bit about this stuff for various reasons...
____________________________
Arch Duke Kaolian Drachensborn, lvl 95 Ranger, Unrest Server
Tech support forum | FAQ (Support) | Mobile Zam: http://m.zam.com (Premium only)
Forum Rules
#99 Sep 02 2014 at 2:56 PM Rating: Decent
Encyclopedia
******
35,568 posts
I was talking about the top end of your range. Most of the sources I've seen set a range of $200-500k, which is right about where your itemized list puts us. I was more questioning the "up to 1 million" bit. While I don't doubt that some very complex intersections might push that level of cost, it's unlikely they could be replaced with a basic traffic circle.

I'll repeat the point that traffic circles provide the same amount of traffic control as stop signs (ie: zero), so the better comparison would be between those two.
____________________________
King Nobby wrote:
More words please
#100 Sep 04 2014 at 8:15 AM Rating: Excellent
Avatar
******
29,919 posts
A stop sign is by definition a form of traffic control. It's also a form of traffic flow management. it's not a form of Traffic Demand Management such as tolling or HOV lanes where use is restricted by law, but it's very purpose is to control the flow of traffic so you don't hit the car going across your lane. A stop sign is not a form of Active traffic control, but rather Passive traffic control, such as a Traffic Circle, but saying it isn't a type of control at all is probably not exactly accurate. 100% is way more than zero.

____________________________
Arch Duke Kaolian Drachensborn, lvl 95 Ranger, Unrest Server
Tech support forum | FAQ (Support) | Mobile Zam: http://m.zam.com (Premium only)
Forum Rules
#101 Sep 04 2014 at 4:08 PM Rating: Decent
Encyclopedia
******
35,568 posts
I don't work in the industry, so I have no clue what the proper terminology is, but I'm talking about the ability to affect the volume of traffic that passes in various directions through an intersection at a time and the creation of openings in busy traffic to allow for entry onto a street from parking lots, alleyways, etc. Whatever that's called, that's what I'm talking about.

The problem with stop signs (and roundabouts) is that they do not do this. The rate at which cars exit the intersection in any given direction is dependent solely on the rate at which they enter another side and cross (ie: no "control" over that rate). A signal light forces specific directions to stop, not just one at a time, but whole sections of cars. They can also be programmed to stay green in certain directions for longer than others, even based on time of day, allowing traffic engineers to design a network of them so as to reduce gridlock and congestion. You cannot do this with stop signs nor with a roundabout.

Roundabouts are arguably *worse* than stop signs at this. The classic example is someone trying to enter a road from a parking lot or driveway a short distance from an intersection. With a signal light, large gaps in traffic will occur when the light cycles (and often during various phases of each cycle) allowing for that side traffic to enter the road easily. With a stop sign, each car stops and lets cross traffic go. This can create a small but usable gap between one car and the next allowing that side traffic to enter the road. With a roundabout? Each car follows the next, likely not spaced out enough to allow for side traffic to enter *ever*.

That's why they are terrible for "traffic control" (or whatever the term is). They are terrible to use in any location with both high volume traffic and cars entering exiting into parking lots/garages/etc along said road. They are a nice pretty amusement to place within an existing traffic grid with other forms of intersection controls around them, but not ever as the sole control. Not unless you're in some quaint village somewhere or something. Their use in any high traffic area requires the application of yet other methods of allowing traffic to enter/exit those roadways utilizing a merge methodology (which is usually space intensive). Plus the roundabouts themselves take up more space.

Using roundabouts in most US cities would require far far more work than just replacing existing intersections with them. You'd have to redesign the entire roadway system (and possibly a large portion of your city itself) to make use of them. Hard to argue that's worth the effort. Now, if you were designing a whole new city from scratch, you could create large throughways with entry/exit lanes and roundabouts to manage intersections between them (essentially acting like US highways), with smaller grids of surface streets utilizing stop signs or something, and possibly make it work ok. You'd still have the issue of no ability to influence traffic load in different areas of your city at different times of the day, like you can with stop lights, but if we're designing a city from scratch so as to make use of roundabouts, I suppose we could design it such that traffic needs are more evenly distributed (but probably not given that the same people who love roundabouts tend to also love uber centralized metro designs for cities, which is the worst combination possible).

From a traffic safety perspective (once people get used to them), roundabouts are ideal. For everything else that intersection control systems can do? They're terrible. Again, simple stop signs provide more utility than roundabouts, and arguably cost less as well. The downside is that traffic does have to stop. But that's the point. You need traffic to stop periodically (and preferably in a controlled manner) to have a traffic system that works under any kind of heavy load.
____________________________
King Nobby wrote:
More words please
Reply To Thread

Colors Smileys Quote OriginalQuote Checked Help

 

Recent Visitors: 298 All times are in CST
Anonymous Guests (298)