• Welcome to SC4 Devotion Forum Archives.

TESTERS *Needed* - NAM Traffic Simulator

Started by jplumbley, January 14, 2008, 04:57:35 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.


jplumbley

#41
For those who are testing the file, and those who are viewing this wondering what will this do for me when released.  Here is a draft of the readme that is not quite complete.  This will give you an idea of what is going on and maybe explain a few finer points to you.

QuoteThe NEW Traffic Plugins

Simulator "A" written by JPlumbley
Simulator "B" written by Mott

What are NAM_Traffic_Plugins intended to do?

Well, there are two Central "cores" named A and B for lack of a better identifier.  These two Central Cores have both been mathematically designed to re-balance the structure of the Simulators.  With new information discovered by Mott from investigations he preformed the previous Simulators have been found to be harmful to the game in some cases.  Not only were the old NAM Simulators found to be flawed, but also the Vanilla Simulator from MAXIS was found to have flaws in it aswell.  These two Cores have been designed to limit and/or remove these flaws, as well as expand on ideas MAXIS had that just simply were not implemented or not implemented properly.

What is the Difference between Simulator "A" and Simulator "B"?

These Simulators have been designed to do two different things. 

Simulator "A" has been designed by JPlumbley to extend the original travel distances set by the MAXIS Simulators.  For example in the original game the MAXIS Simulator allows the Sims to walk upto 7 tiles, but with Simulator "A" the Sims will walk upto 43 tiles to get to work.  This change in travel distance is due to the calculations used to allow Car Traffic to travel one full distance of a large city tile (512 tiles) on Avenues. 

Simulator "B" has been designed by Mott to work with the original travel distances set by the MAXIS Simulator.  For example in the original game the MAXIS Simulator allows the Sims to walk upto 7 tiles, therefore in Simulator "B" the Sims will walk upto 7 tiles.  Simulator "B" has also been re-calculated to be more balanced and provide better pathfinding overall.

Other modifications to both Simulators include changes to the follow properties with minor variation:

Congestion to Accident Probabiltiy - Revamped the probability curve.
Congestion vs Speed - Made use of this, not used to full potential by MAXIS.  Sims will now look for better routes if the network is over congested.
Trip Length to Minutes Display Multiplier - Used in calculations for time displays on Commute Time Graph.

What do the different difficulties mean?

The only difference between the different difficulties is the Capacity of the networks.  All have been altered from the original MAXIS Simulator.  Road, OneWay Road and Avenue have all been equalized in speed and capacity in preparation for the Network Widening Mod.  The following is a list of the Network Capacities for each difficulty:

Easy Difficulty

Street:                   2250
Road / OWR / Avenue:   5600
Highway / El-Highway:   10800
Rural Highway:      10800
Subway / El-Rail / GLR:   13500
Monorail:                   13500
Rail:                   13500

Medium Difficulty

Street:                   1500
Road / OWR / Avenue:   3750
Highway / El-Highway:   7200
Rural Highway:      7200
Subway / El-Rail / GLR:   9000
Monorail:                   9000
Rail:                   9000

Hard Difficulty

Street:                     1000
Road / OWR / Avenue:   2500
Highway / El-Highway:   4800
Rural Highway:      4800
Subway / El-Rail / GLR:   6000
Monorail:                   6000
Rail:                   6000

What effect will the different difficulties have on my play style?

The Simulators are designed to help your game calculate proper pathfinding situations for your Sims to find thier way to work.  Now, the Network Capacities have an almost equally important role in the way the Simulator works as the Maximum Commute Time and the Speed.  Some people will say that Capacity isnt important and can be set to anything.  Well this is only a half truth, while Capacity "can" be set to any setting it is not wise to set the Capacities too high.  If the Capacities are too high the Sims will never look for another route beyond the shortest path because your networks will never become congested.  If the networks are never congested no speed will be lost and Sims will only take the shortest physcal route.  Now, if the Capacity settings are set appropriately the pathfinding will act more realistic.  In Real Life roads become over crowded and then people look for other routes or new roads are built.  Having lower capacities will force you to design your networks more dynamically, plan ahead and provide more of a challenge.  But, if the Capacities are set too low, it will make the game too hard from a transit standpoint and prevent large cities.  If you have CAM, and a network capacity is set too low, one building may cause over congestion on any network if you are at some of the higher stage levels.

It is advisable that you choose your difficulty based on the size of the city you are planning to build or what stage you are in building your city.  Some general findings during testing find that for existing cities the Difficulties best preform at:

Hard Difficulty for cities upto 250,000 population.
Medium Difficulty for cities 200,000 population to 1,000,000 population
Easy Difficulty for cities 750,000 population and greater.

Of course there are for existing city sizes.  If you are starting a city from scratch, it is advisable to start on Hard Difficulty because it will allow you to build a city with a very well optimized traffic system.  If the city becomes too large your Simulator may benefit you moving upto Medium Difficulty.  But if you design your network from the begining to be optimized for Hard Difficulty, you may never need to get to Easy Difficulty.

What is the NAM_Traffic_Data_View Mod?

This is a modification to the Congestion Data View.  The original Congestion Data View is very skewed, what most people may not realize is that the Congestion Data View doesnt report Congestion in the way one would expect.  In fact, you will not notice the on the Congestion map that your networks do not start becoming yellow with congestion until they reach about 175% to 200% and dont actually reach red until they are over 300% capacity.  This is not very accurate a portraying when your networks start suffering consequences from over congestion.  Your networks will still end up having a speed reduction when the network becomes over 100% capacity, but you will never notice it until the network is more congested and begining to become yellow.

Due to this inconsistancy, the Congestion Data View has been edited to show signs of over congestion sooner rather than later.  This will not only help you avoid the consequences of Congestion and give you a more accurate reading into when the network starts to suffer the consequences.

I hope this has been somewhat useful to read.  The current settings of the test file are approximate to the Medium Difficulty Capacities.
"You learn something new everyday."

http://img517.imageshack.us/img517/169/nhpjplumbleykv3.gif
Bringing the new horizons closer to reality.

Berethor ♦ beskhu3epnm ♦ blade2k5 ♦ dmscopio jplumbley ♦ moganite ♦ M4346 ♦ Dedgren ♦ Ennedi Shadow Assassin ♦  Tarkus ♦ wouanagaine
Street Addon Mod - SAM

SC4BOY

#42
Quote from: jplumbley on January 19, 2008, 04:14:30 PM
I will start off by addressing the Commute Time dropping, though your Congestion has grown.  I think this has to do with the Simulator recalculating alot of your paths.  The old Simulator is not very optimal, in fact it likes to use the same route over and over, especially when capacity allows it.  If the simulator starts using more routes as it did, it will allow for the Commute time to drop.

I agree.. Of course this cities small size has not given me the time to "build up" the city transport as it only just started growing higher density lots .. the city is only about 5 years old.. I have spent most of the time on the outter tiles in order to build demands

QuoteNow, about your congestion "issue".  The capacity is set much lower than 10x capacity (apparently what you were using).  Due to this you have built your city to be optimized to this capacity setting, which would actually remove some of the effectiveness of the Simulator such as the simulator spreading out the traffic amoungst all possible routes.  Another thing that seems to add to your congestion issue is first off you said you are using Streets, these are only set at 1000 capacity and not 5000

The setting I use makes the streets 1000. The set you had me install starts to turn orange about 200, not 1000. The original streets are 100 so 1000 is 10x. If you thought it was different, either your plugin is incorrect or I somehow installed it improperly.

Quote(I think thats what the 10x Capacity was set at).  The streets are in a fairly high density area with no "major route", due to this you are going to become more and more congested the further you build away from the bottleneck you have created.  Not only have you used the under capacity streets in high density zoning, you have also not used mass transit.  Rails, subways, monorails etc all have higher capacities, in the test file set at 7500.  If you had any of these in your system, your congestion problems would be virtually nulified becuase it gets Sims off the roads.

You can't tell from those pics but there are only 2 ways into the CBD.. they have to use GLR and the avenue/ highway/ rail which are difficult to see there. Each "spoke" has GLR running into the CBD and the "transport hub" in the center of the city.  There also is bus service, but only inside each suburb and the CBD.. they do not have easy access between. This is a purposeful experiment on my part to both prevent commute loops and to force inter-tile traffic to go through the CBD. I guess the bottom line that I'm pointing out is that there is mass transit, but almost all of it is 5x saturated in those pics above.. I should have posted my transport breakdown. I'll do that later.. sorry didn't think of it.

QuoteIn real life, transit systems are built as a dynamic system where you cannot allow only one type of transportation.  My city has approximately 170k inhabitants.  We have one highway, busses and one main rail line, and we are widely considered a bedroom community (everyone commutes out of town).  At rush hour traffic in our city is almost a standstill because we havent put nearly enough money into our bus system as we should have and our rail system, well, that is a different story all together.  Approximately 60 to 70% of our "Sims" drive to work and that accounts for much of the congestion.  So, I feel that your test accurately portrays a very real situation.

Just wait till I release my mod on the Congestion Data View, that will show you the Actual Congestion in your cities and not the skewed MAXIS Congestion Data View.

Sounds interesting. The primary element of your layout that I like is the "congestion alternate searching" that you have allowed to develop. The rest is pretty much personal taste to me.

Always like to see people exploring though.. ;)

jplumbley

Quote from: SC4BOY on January 19, 2008, 11:04:27 PM

You can't tell from those pics but there are only 2 ways into the CBD.. they have to use GLR and the avenue/ highway/ rail which are difficult to see there. Each "spoke" has GLR running into the CBD and the "transport hub" in the center of the city.  There also is bus service, but only inside each suburb and the CBD.. they do not have easy access between. This is a purposeful experiment on my part to both prevent commute loops and to force inter-tile traffic to go through the CBD. I guess the bottom line that I'm pointing out is that there is mass transit, but almost all of it is 5x saturated in those pics above.. I should have posted my transport breakdown. I'll do that later.. sorry didn't think of it.


I have just double checked the Capacity of Street and it is 1000 in the test file I sent to everyone.

About the way your system is setup.  This is what I can see from the pictures and if what I say is wrong then, please correct me.

- From what it looks like to me you have GLR in Avenue as one of your main connections to your CBD.  The "two" networks cannot be considered separate entities in the reference to Capacity.  Unfortunately, the game cannot determine that there are 2 separate networks on one of these tiles.  Therefore, it cannot determine that there would be 2 separate Capacities.  It does however take the properties of the network it was based off of, in this case Avenue.  So, the capacity of the tile is that equal to an Avenue (if my understanding of the puzzle pieces is correct, and I could very well be wrong about what the capacity is).  But whether the capacity is set different from the normal Avenue, the effects of each traffic type still stack when determining Congestion.  You would be allowed X number of sims in this tile whether they be Cars or Trams.  This will cause a bottleneck, but if you notice this is not where your problem currently resides.  Your problem is currently the streets leading upto this area, Sims must travel upto and beyond 6 or 7 or 8 blocks in some cases to reach the Avenue.  When you have medium to high density buildings along the entire route, you are looking at approximately 600 people per block (minimum) the closer you get to the avenue the more stacking effect you will have from the Sims further away.  You will easily have 4,000 Sims after 7 or 8 blocks of travelling.  Especially in a grid system that is going to the same bottleneck it forces Sims into a limited number of choices.

The MAXIS Congestion Data View also will only show red at approximately 400% Capacity, Yellow at approximately 200% capacity.  Even though the traffic still starts having negative effects at anything greater than 100% Capacity.  So, effectively, your Streets closest to the Avenue are probably sitting around 350% or 3500 Sims on average.

- I could be mistaken about this but it seems as though your Rail and Highways only service the outlying areas, meaning none of the traffic within the city limits are carried by this, or at most very little of it is carried.  The Traffic Data View is what leads me to this conclusion since the only major traffic along the highways is coming into the city and there are no major interchanges until you reach downtown.

This is just the things that I think are causing your problems from where I am standing.
"You learn something new everyday."

http://img517.imageshack.us/img517/169/nhpjplumbleykv3.gif
Bringing the new horizons closer to reality.

Berethor ♦ beskhu3epnm ♦ blade2k5 ♦ dmscopio jplumbley ♦ moganite ♦ M4346 ♦ Dedgren ♦ Ennedi Shadow Assassin ♦  Tarkus ♦ wouanagaine
Street Addon Mod - SAM

SC4BOY

#44
Quote from: jplumbley on January 19, 2008, 11:44:40 PM
I have just double checked the Capacity of Street and it is 1000 in the test file I sent to everyone.

Then I have some error in my install or have left some file in. The red you see in my pics are 200-500 on the streets. This is measured via the "query path" tool, not extrapolated from the "congestion picture". I'll have to examine my files to try to see what in my file is resetting it.

QuoteAbout the way your system is setup.  This is what I can see from the pictures and if what I say is wrong then, please correct me.

- From what it looks like to me you have GLR in Avenue as one of your main connections to your CBD.  The "two" networks cannot be considered separate entities in the reference to Capacity.  Unfortunately, the game cannot determine that there are 2 separate networks on one of these tiles.  Therefore, it cannot determine that there would be 2 separate Capacities.  It does however take the properties of the network it was based off of, in this case Avenue.

Actually no.. this is purely because the stations in avenue that I am using are set to 5000 I think and they are 3x saturated on that avenue. That GLR in avenue was a last minute addition to the city to handle those large HKBAT R$ buildings by the avenue/glr. Originally the avenue simply passes by there with no connections. In fact I made a mistake in the intersection of those stations as I did not understand how the stations worked... or at least didn't give it sufficient thought. Each "housing clump" has a GLR (5 seperate lines + the GLR/Ave) of its own completely seperate from the GLRinAvenue which goes the CBD and certain "unload spots" on the way. They all converge on the central city sunken train station.. unfortunately its capacity is 5000 and its operating well over 25000.. ;(

**NOTE: I was going to attach a transport map so you could see the transport better, but when I exited the city tile to get the region view and switched to "transport view" it looked like THIS (bottom of this message) This started just this load.. I exited and moved around etc and it doesn't go away. I changed regions and it was ok on the other regions. Anyone have an idea of what this is, or seen this effect before? Only change I made was this mod and 2 buildings off the LEX (and re-packed it).


QuoteSo, the capacity of the tile is that equal to an Avenue (if my understanding of the puzzle pieces is correct, and I could very well be wrong about what the capacity is).  But whether the capacity is set different from the normal Avenue, the effects of each traffic type still stack when determining Congestion.  You would be allowed X number of sims in this tile whether they be Cars or Trams.  This will cause a bottleneck, but if you notice this is not where your problem currently resides.

It is interesting to hear though that the "GLR/ Avenue" are really a single path.. I hadn't thought about that before. An interesting observation I happened to make is that wheree I crossed the rail, I split out the GLR then ran it back into the AVE after crossing as there are no AVE/GLR puzzle pieces for "crossing over" cases. When I did that the GLR over RAIL (using the on-slope pieces) are bright red while the GLR/Avenue pieces are not... I suppose a peculiarity of TE lots?

The congestion numbers (for streets) are definitely in the 200-500 range. Stations are bright red in almost all cases, so the streets, though congested, are not all of the problem. And the "multipath" of yours is what is allowing it to work as well as it does. Without it it would be totally jammed with the usual "one red stripe" symptom.. :) If I had more or better stations, it would probably show even more streets being nice and red.. hehe

QuoteYour problem is currently the streets leading upto this area, Sims must travel upto and beyond 6 or 7 or 8 blocks in some cases to reach the Avenue.  When you have medium to high density buildings along the entire route, you are looking at approximately 600 people per block (minimum) the closer you get to the avenue the more stacking effect you will have from the Sims further away.  You will easily have 4,000 Sims after 7 or 8 blocks of travelling.  Especially in a grid system that is going to the same bottleneck it forces Sims into a limited number of choices.

This basic idea is correct, whether or not the details are.. ;)

QuoteThe MAXIS Congestion Data View also will only show red at approximately 400% Capacity, Yellow at approximately 200% capacity.  Even though the traffic still starts having negative effects at anything greater than 100% Capacity.  So, effectively, your Streets closest to the Avenue are probably sitting around 350% or 3500 Sims on average.

The numbers I have been giving you were read by the query tool, not by inferring anything from the map color.

I did go back as I mentioned above to take a screenie of the traffic volumes. Its good as they show all traffic types seem to increase. I don't know if that is just "the city was ready to grow" or the effect of your patch. I expect the latter... would suggest that it's doing a nice job.


Traffic Volumes by Type

Quote- I could be mistaken about this but it seems as though your Rail and Highways only service the outlying areas, meaning none of the traffic within the city limits are carried by this, or at most very little of it is carried.  The Traffic Data View is what leads me to this conclusion since the only major traffic along the highways is coming into the city and there are no major interchanges until you reach downtown.

This is generally correct. Each suburb can use the "major arteries" only to go through downtown and to their destination. No vehicles may "jump the suburb" to bypass the CBD. This was a conscious decision.

QuoteThis is just the things that I think are causing your problems from where I am standing.

hehe.. there is no "problem" at all..other than some of the things we're gleaning from these discussions.  I've made no effort to "fix" the things shown in the  pics/graphs.. I had neither time nor objective, to readily attack a longer term test with a reasonable "A/B" test. It only takes a "snapshot" of the status from a standard to a "JB" situation. I think there are things to learn about it. The issue in my mind is not "what do I do to fix my city".. those are probably rather conceptially simple.. perhaps pedantic..   But rather, how does this act vs the "10/ 10/ better" NAM approach. The best observation in my mind is the nice feature of the "congestion driven multipath selection" shown.

I believe for that to work well, it almost will have to be "customized" or "scaled"  to larger cities for capacities. The range of traffic pathing AND station capacities just offer a challenge for the player.. as it should of course. For the "small town" I doubt it will much matter at all, but cities that range from say 100k to the 3 mil or so will have rather different problems. At least that's where it sits in my mind for now. I'll have to continue to wait and see what your/Motts, etc experiments reveal as applied to people's cities.

If they are set too low, congestion will QUICKLY build up even in a rather low-density city such as mine, and if too high, it will be unnecessary in any but the very largest city.. and still of course the capacity of stations will be severely tested.. hehe.. I'm gonna have to learn to mod those I can see right now..a situation which doesn't sit easily with me. I don't normally play with dense cities. I'm running this as my CAM test city and just playing with some ideas.

I will say that relative to the original NAM setup I had this is some improvement. If I run the city on the "old" settings, I start almost immediately to have a "demand sag" of abandonments due to "low demand". It doesn't seem to be a MAJOR shakeout as it shows some clearing, but I can't say long term as I have only run it out for a couple of years. The "JP patch" also gets this "sag", but it SEEMS to be not as bad. Time will tell.

Keep up the good work.


TRANSPORT MAP PROBLEM:

Starmanw402007

I'm not sure why I'm not be allowed to load photobucket to upload my newest photos of Steven's Point traffic Simulator test. But for the past week I've been unsuccessfull at getting that website to load fast. I've been wanting to upload these photos now. Can someone tell me what's up? If Photobucket is down or just annoyingly slow? And btw, this is my 200th  post.
Your Friend;
Mayor Of Steven's Point & Maxiston
(Proud To Be Cities Of Sim Nation!)

beskhu3epnm

#46
Quote from: Starmanw402007 on January 20, 2008, 07:40:06 AM
I'm not sure why I'm not be allowed to load photobucket to upload my newest photos of Steven's Point traffic Simulator test. But for the past week I've been unsuccessfull at getting that website to load fast. I've been wanting to upload these photos now. Can someone tell me what's up? If Photobucket is down or just annoyingly slow? And btw, this is my 200th  post.
As a quick alternative, I traditionally go here:
Imageshack
I click the browse button, select the file and hit upload. The site generates a link (the bottom link - direct link) that you can use here at SC4D. I've never had any problems with this, and it does not require a login at all. If I have more than one image, I load as many tabs as images I have, and upload them individually, generating the links I need.

Just a suggestion if you're in a bind... sorry for threadjack! (Thread jacking when you are helping is definately appreciated - JPlumbley)

I shall enter round 2 of testing now... I have some ideas based on what ideas I've seen thus far...



http://sc4devotion.com/forums/index.php?board=79.0
New Horizons Productions Berethor ♦ beskhu3epnm ♦ blade2k5 ♦ dmscopio dedgren ♦ Ennedi ♦ Heblem ♦ jplumbley ♦ moganite M4346 ♦ Shadow Assassin ♦ Tarkus ♦ wouanagaine

Starmanw402007

#47
After trying to get into photobucket for 1 week. My new host will be imageshack ok. Now, I can go ahead and update my Nam Traffic Simalutor test of Steven's Point traffic. So without further due Here's the information you requested JP. I hope this is Correct. Can someone adjust the codes for me to make it so that the codes dont show. Thanks In advance.  (fixed codes) - JPlumbley









Are these the photos you want for nam traffic simulator test?
Your Friend;
Mayor Of Steven's Point & Maxiston
(Proud To Be Cities Of Sim Nation!)

Palpatine001

First Solaria



A interconnected maze of city tiles housing 3 million sims

http://s124.photobucket.com/albums/p5/Palpatine001/?action=view&current=Solaria-TransitwithRHW.jpg

This is the transit network although parts of the RHW can not be seen.




First quick shake down test.

All cities minus the oldest three (Imperial Command Centre, Rangitoto Town and Te Puke) across all tile sizes and development classes have seen a remarkable improvement in transit conditions, I have noticed congestion decreasing a little bit as better commute times are found and the transit options available (I use all in an intergrated interlink system) being more efficiently used. Traffic volumes on the highways are up for commuters going that longer distance, thus I am theorising my highways where available are being better utilised than they were pre test. All in all, better utilisation of the highways as they were designed to do when I placed them, and more efficient useage of the interlinked mass transit system has caused a noticable increase in demand again in RCI although my higher capacity stations are filling to capacity faster than pre test. No matter as the later built cities are designed to take on higher capacity stations and extensions to transit lines. I have noticed jplumbly has looking at
QuoteCongestion to Accident Probabiltiy - Revamped the probability curve.
Congestion vs Speed - Made use of this, not used to full potential by MAXIS.  Sims will now look for better routes if the network is over congested.
Trip Length to Minutes Display Multiplier - Used in calculations for time displays on Commute Time Graph.

and this is most likely the reason for the more efficient commuting and a slight reduction in commute times (around 10% across all cities)

In my oldest three cities the highways maxed out real fast as well as the rail network in the respective tile causing some what of a nightmare to try and correct my end. However with these tiles having being originally developed pre Rush Hour then being constantly modified over the years, coupled with these 3 first cities of Solaria hold 50% of the region's population I am not surprised traffic congestion and volumes have skyrocketed. I believe this is due to the test file giving a more accurate reflection of the mass amount of traffic moving between the 3 cities (IPC being the largest and oldest city in Solaria at 900k and the all powerful RCI hub, Rangitoto Town at 400,000 in a medium tile, just as old as IPC and a Residential Hub, with Te Puke being an industrial hub). Further more indepth testing has to be done on these older cities and will be given in my next report at the end of the week.

A thing I noticed was all commute times in the commute chart going up by x10, a bug maybe as all buildings are reporting short and medium commute times as before the tests started? Ill get a pic of it for the next report.

Next report with graph samples out Friday NZT
Now Idle
Please leave a message after the beep and I'll get to you as soon as I can bother

bigdope404

Here is my testing results.

I tested large city tile with a small population (1760) that is primarily a rural farming community.

The simulator I used before was the 10XSpeed, Commute, and Capacity. CAM was installed.
-  Abandonment.

I noticed no abandonment due to the Traffic simulator

- Another is do you think the Sims have become smarter in the way they choose paths.

The Sim have become noticeably smarter in choosing paths.
For Example:


Before Freight Trucks took the long way around to a street instead of using the Highway that was right there.

And After

The Freight Trucks use the Highway.
Data Views

Before






After 1 year







After 25 years






All areas stayed green because of the limited commuters in the city.

I hope this data has been useful.

jplumbley

Quote from: Palpatine001 on January 20, 2008, 09:07:36 PM
First Solaria

First quick shake down test.

All cities minus the oldest three (Imperial Command Centre, Rangitoto Town and Te Puke) across all tile sizes and development classes have seen a remarkable improvement in transit conditions, I have noticed congestion decreasing a little bit as better commute times are found and the transit options available (I use all in an intergrated interlink system) being more efficiently used. Traffic volumes on the highways are up for commuters going that longer distance, thus I am theorising my highways where available are being better utilised than they were pre test. All in all, better utilisation of the highways as they were designed to do when I placed them, and more efficient useage of the interlinked mass transit system has caused a noticable increase in demand again in RCI although my higher capacity stations are filling to capacity faster than pre test. No matter as the later built cities are designed to take on higher capacity stations and extensions to transit lines. I have noticed jplumbly has looking at
and this is most likely the reason for the more efficient commuting and a slight reduction in commute times (around 10% across all cities)

In my oldest three cities the highways maxed out real fast as well as the rail network in the respective tile causing some what of a nightmare to try and correct my end. However with these tiles having being originally developed pre Rush Hour then being constantly modified over the years, coupled with these 3 first cities of Solaria hold 50% of the region's population I am not surprised traffic congestion and volumes have skyrocketed. I believe this is due to the test file giving a more accurate reflection of the mass amount of traffic moving between the 3 cities (IPC being the largest and oldest city in Solaria at 900k and the all powerful RCI hub, Rangitoto Town at 400,000 in a medium tile, just as old as IPC and a Residential Hub, with Te Puke being an industrial hub). Further more indepth testing has to be done on these older cities and will be given in my next report at the end of the week.

A thing I noticed was all commute times in the commute chart going up by x10, a bug maybe as all buildings are reporting short and medium commute times as before the tests started? Ill get a pic of it for the next report.

Next report with graph samples out Friday NZT

First thing I have to say... Pictures dude.  Anyways, thanks for the description of the way the Simulator has preformed.

I will comment first on the "increase" in the Commute Time Graph.  You cannot look at this compared to the Time from the previous Simulator you had installed.  Every Simulator is different, in fact, if you had ANY thing that effected Travel Speed or Maximum Commute Time then the graph would have been effected.  In most of the old NAM Simulators both or one of these two functions was edited, by doing this and not re-calculating the Trip Length to Minutes Display Multiplier the Simulator would definately distort the graph and make you see something different than the previous Simulator.

Now, the actual Commute Time is not actually measured in time, its measured in "hop counts" essentially and each tile has a specific "hop count value" to it.  This is unique to the network that is used, for example the cars on MAXIS Road had a network speed of 31.  This means that the car can travel 31 tiles per unit of time or one tile = 1/31 units of time.  Basically, if you multiplied the speed by 10 each tile would have a "hop count" of 1/310 units of time.  Maximum Commute Time defines the longest there and back trip a Sim is allowed to go.  In the original MAXIS Simulator it had a value of 6, so in a there and back trip you would be allowed to travel 6*31 or 186 tiles.  Of course you cant travel that far to get to work, because you need to get back.. so the REAL distance a Sim could travel was 93 tiles to get to work and 93 tiles home.  So, now we have the distance a Sim can travel... What I did was assume that a Sim could travel 512 tiles on an Avenue and calculated the Commute time if the Speed = 60 tiles/unit of game time.  Then assumed that the travel time was approximately 1 hour and calculated the Trip Length to Minutes Display Multiplier.  So, if your average commute time is approximately 30 mins on the display graph your average Sim Commute is about 256 tiles by Avenue, closer to 350 tiles by highway, and even further by the mass transit options.  This is more or less the way for you to read this graph, its by far not "real time".

Bet you never thought I could write 3 paragraphs about one little comment... Now you know better LOL.
__________________________________________________________________________________________

I would like to make a patch of the most widely used Transit Stations, but that would take quite a bit of time and there would essentially need to be 2 patches for each station.  One for my Simulator and one for Mott's.  What needs to be changed is the Traffic Switch Cost Property.  Most lotters do not understand how to use this property properly.  Most will either set it to 1 or 0, which unfortunately is not the best thing.  What needs to be done is a minor calculation, based on the Network it is designed for.  So, if it is a bus stop you are going to want to calculate it based on the speed of the slowest Street Network.  So, if the lot is a 1x1 Bus Stop, then you are going to want to set the Traffic Switch Cost equal to:

1 (number of tiles) x [1+3]/25 (speed of the Street network)  The reason why I have that +3 is because you want to give the bus stop a little bit of a disadvantage compared to the Road it is next to.  If you dont add something, the simulator may actually use the Bus Stop as a "short cut".  
__________________________________________________________________________________________

Anyways, thanks for the tests there guys!  Keep em coming.
"You learn something new everyday."

http://img517.imageshack.us/img517/169/nhpjplumbleykv3.gif
Bringing the new horizons closer to reality.

Berethor ♦ beskhu3epnm ♦ blade2k5 ♦ dmscopio jplumbley ♦ moganite ♦ M4346 ♦ Dedgren ♦ Ennedi Shadow Assassin ♦  Tarkus ♦ wouanagaine
Street Addon Mod - SAM

Tarkus

#51
Well, here's the results of my testing on a city of roughly 43,000.  I'm using jplumbley's Traffic DataView modifications in all of these pics.  The first simulator used is mott's a03, whereas the second is jplumbley's "Hard" plugin.

If you need larger pics, let me know.  I'm probably going to do a bit more testing on this as well, with different cities, and I may even compare it to the various existing NAM Traffic Simulators, out of curiosity.

Test Part 1--mott's simulator

Congestion--appears entirely green through the entire city:



Volume (Road Traffic)--Traffic appears fairly evenly distributed, mostly a light blue-grey color, though there are some points that are bluer than others.  The most volume reported is on the neighbor connection on the south side of the grid, in the center (the grid is facing north).



Volume (Pedestrian):  Usage appears fairly slight, scattered through various points in city, with probably the most coverage around the downtown area (near the center of the grid)



Volume (Car):  Similar to Road Traffic view.  The only two road-based modes of transportation used in the city are Cars and Freight Trucks.  (There are no buses, or any sort of rail or other mass transit in this tile).



Commute Time:


Overall Traffic Volume:



Test Part 2--jplumbley's Simulator

These pics are after about 6 months of running the new simulator, enough time for it to "kick in".  Very little change in the congestion/volume figures.  It does seem, however, that the pedestrian range has increased slightly, creeping farther west.   I also measured the most heavily-traveled section of road (the aformentioned southern neighbor connection), and it was peaking at just shy of 1,100 cars.









The Commute Time has dropped slightly.  The point where the new simulator kicks in is just about exactly half way through the graph.



The overall Traffic Volume supports the findings of increased Pedestrian traffic, with a slight increase in Pedestrians, and a slight decrease in Cars.



Other Notes

There were no other unusual happenings (abandonment, etc.) in the running of this experiment, with either simulator.  The population in the city did increase by 176 sims over the course of the test, though this is negligible.

-Alex (Tarkus)

SC4BOY

Just a followon note about the simulator. The report I submitted covered a range of about 4 months where it seemed to "level out". I have found that over a period of about 5 years, I believe the "model" continues to level and smooth traffic even further.. This is good as it continues to level traffic over the entire grid.. even rather complex moves.

JP- is there a "boarding penalty" for all transit switches? I read that R$$$ sims are assessed 40 min "boarding penalty" for mass transit and was wondering if the act of boarding or switching transit modes is always assessed a penalty?

jplumbley

Quote from: SC4BOY on January 21, 2008, 06:55:44 AM
Just a followon note about the simulator. The report I submitted covered a range of about 4 months where it seemed to "level out". I have found that over a period of about 5 years, I believe the "model" continues to level and smooth traffic even further.. This is good as it continues to level traffic over the entire grid.. even rather complex moves.

JP- is there a "boarding penalty" for all transit switches? I read that R$$$ sims are assessed 40 min "boarding penalty" for mass transit and was wondering if the act of boarding or switching transit modes is always assessed a penalty?


Part 1 - That should be the cases.  Sims should level out and be smarter in their choice of work and path.

Part 2 - The Transit Switch Cost in many custom station lots does not get used.  It should to make the Simulator work and should be calculated appropriately, setting it to 1 is not the appropriate measure (which some people do).  So, if I had my way, then yes they would be affected by some sort of "boarding penalty".
"You learn something new everyday."

http://img517.imageshack.us/img517/169/nhpjplumbleykv3.gif
Bringing the new horizons closer to reality.

Berethor ♦ beskhu3epnm ♦ blade2k5 ♦ dmscopio jplumbley ♦ moganite ♦ M4346 ♦ Dedgren ♦ Ennedi Shadow Assassin ♦  Tarkus ♦ wouanagaine
Street Addon Mod - SAM

SC4BOY

hmmm.. maybe I should have used the word "transfer" rather than switch... I understand and agree about the "transit switch" issue.. What I am referring to (and I don't have TRUE VERIFIED INFO that it is right) is that specifically R$$$ sims get a 40 min (how they decided it was "40 minutes" given the vagaries of the specific meaning of SC4 variables and its arbitrary units, I don't know) added to their commute time when they board mass transit (I don't know if that is the FIRST BOARDING only or for every transfer thereafter). The "transit switch penalty" you refer to is not wealth dependent is it?

jplumbley

Quote from: SC4BOY on January 21, 2008, 07:56:58 AM
hmmm.. maybe I should have used the word "transfer" rather than switch... I understand and agree about the "transit switch" issue.. What I am referring to (and I don't have TRUE VERIFIED INFO that it is right) is that specifically R$$$ sims get a 40 min (how they decided it was "40 minutes" given the vagaries of the specific meaning of SC4 variables and its arbitrary units, I don't know) added to their commute time when they board mass transit (I don't know if that is the FIRST BOARDING only or for every transfer thereafter). The "transit switch penalty" you refer to is not wealth dependent is it?


To my immediate knowledge there is no wealth dependant cost or maximum time for transit stations or transit commute time.  I can look again as I know there are wealth dependant properties within the Traffic Simulator that give the "chance" that each wealth will look at Transit before Cars.  I will investigate further.
"You learn something new everyday."

http://img517.imageshack.us/img517/169/nhpjplumbleykv3.gif
Bringing the new horizons closer to reality.

Berethor ♦ beskhu3epnm ♦ blade2k5 ♦ dmscopio jplumbley ♦ moganite ♦ M4346 ♦ Dedgren ♦ Ennedi Shadow Assassin ♦  Tarkus ♦ wouanagaine
Street Addon Mod - SAM

Diggis

What I remember of previous discussions was that if a sim was forced to use a non prefered form of transport there was a penalty.  So if you have a city which forces sims to use mass transit and the $$$ sims have an 10% chance of choosing mass transit then 90% of them will get penalised for it. (I think)

rickmastfan67

Starmanw402007, off topic here for a sec, but you shouldn't use PNG's to post your images from SC4.  Saving them in JPG format would have been a much better idea.  I just converted one of your PNG images into JPG format and it uploaded it to the end of this post.  Your original PNG was 1.23MB, but when I converted it to JPG, it became 278KB at Compression level 13 with really no image quality loss. 

Original Image in PNG format:


JPG Compressed Image:


You can find more info on how to upload your images properly by following the link in my Sig to the JPG Compression tutorial. ;)  I know every here would appreciate it. ;)  Now back to your regularly scheduled thread. :P
-- James Mast, aka: rickmastfan67
Painter/Public Relations at: Masgrafx Racing.

Check out the 9/11/01 Car Set Checklist.

Take a look @ the JPG Compression tutorial.  A must read to help cut away unnecessary file size from 500k+ images to help dial-up users out.

Starmanw402007

Thanks for letting me know rick, I didnt know that.
Your Friend;
Mayor Of Steven's Point & Maxiston
(Proud To Be Cities Of Sim Nation!)

flame1396

Thought I took more pictures but I can explain everything...

First, commute times doubled, I think due to capacities being set back to defaults. I'll need to tweak that.

Second... rail, bus, and pedestrian volume increased and car volume decreased slightly

No real effect on much else.



I really can't say if pathfinding improved or whatnot because the capacity reset really doesn't work with my play style.
The most astounding and unique aspect of the human race is our fervent application of our ingenuity to kill each other, thus completely defying the near-universally proven fact that the ultimate goal of a member of a species is to ensure the survival of the species.