A major part of this reply was removed as the data was either obvious or wrong. hopefully one day a full text will be written on all known about walkers but i have no plans for doing that at the moment.

TBI stands for to be investigated

Observations, Conjectures and Hypotheses:

Tile dependent is which direction will be chosen first (NW, NE, SE or SW).

Each walk has the following stages:

stage 1
is explained in the currently last reply (see the edit time and the reply posted just a bit before that), there is some more said on stage 2 and 3 as well

stage 2
here the walker behaves "randomly" he choses his direction at the moment (TBI when exactly). The last thing i tested indicates that if the walker is in the same phase including its number, has walked the same distance, is on the same tile, and is offered the same choice of direction he makes the same decision. It is difficult to create such a situation; i have managed once and the two walkers who were spawned from different buildings and had followed a different path until that point continued in the same way. TBI heavily.
This phase continues exactly as many tiles as the "short" patrol is believed to be.

stage 3
when stage 2 is over the algorithm checks if there is a road connection to the spawning building, if there is the walker becomes a "destination" one. If there isn't or if he encounters an unpassable obstacle, the walker is teleported back to his building.

Destination walkers are always given a route which they will follow even if the road is deleted unless of course an obstacle is placed where they are supposed to pass. They always cut corners, while random ones never do.

[This message has been edited by lemmus (edited 11-17-2004 @ 04:35 PM).]

Hi lemmus, and (since I haven't said it yet) welcome to the forum

I mentioned some of what is known about "random" walkers in reply #2 of a post called OK, need some info not on the site..

I'd guess (based on experience with Pharaoh) that the behavior of "random" walkers depends not only on the roads that they use, but also on other nearby roads. Therefore, the "random" walkers in (for example) a housing block might behave differently if the roads around the block were different.

hi brugle

it turns out that my checking for relevant treads was very careless; checking again i found some good discussions on random walkers, forced walkers and your spawn/return tile research.

still, little is known about the path choosing algorithm for random walkers and i hope some tests will shed some light on it. it does seem that they initially are destination walkers and i haven't seen this mentioned anywhere yet. perhaps more things will come out or perhaps not; i spent hundreds of hours in the course of a few years working on understanding planar graphs to unravel a simple proof of the four color theorem but didn't.

lemmus,
The most extensive reports that I've seen on "random" walker behavior were for Pharaoh, by StephAmon, in these 5 threads. I've only read parts of them.

As far as I have been able to observe the quadrambles that StephAmon describes for Pharaoh pertain to CIII as well. In practice I have found the detailed information too complicated to use for planning road network layout and building placement. To me it suffices to be aware that all random walkers repeat the same routine after four walks, and that the routine depends on the precise placement of that building with respect to the road network in the vicinity of that building, and of course the intersections of the network as such.

Without intersections the behaviour is most predictable, which makes loop and line blocks very popular, but even then a walker may be going in the same direction every time, leading to imperfect coverage. Close inspection using the overlays will reveal such behaviour (now you know what to look for), and if observed it can often be remedied simply by moving the building (delete and rebuild).

The same method works fine for me in more complicated networks, for instance the windmill blocks I used in one of the megacities I designed (Zaanstad). A windmill block consists of a four way intersection with loops at each of the four ends. Infrastructural buildings near the intersections will have their walkers going in all four directions in a steady repetition if placed at the appropriate positions with respect to the network. The blocks aren't particularly dense, but they require only very few infrastructural buildings per house, especially as the city was getting grain from Rome (like in the first tutorial mission).

Thank you for the excellent information. StephAnom's work on intersections and visual domain is very professional and his findings gave me a very good idea in where to focus my own experiments.

I will be updating reply #1 as new things come and will post all updates now and then to indicate that there are some and to save you some time if you have already read the whole reply. I may also edit the first post to provide a better introduction for new readers.

Any comments and criticism are welcome and appreciated.

lemmus,
If you'd like people to be aware that you've edited a post, then add a new reply (maybe something like "I added information on X to reply #1"), since adding a reply puts a thread on top of the non-sticky list.

i know brugle, what i meant was that i will post the updates in a reply and move the tread up only once in a while or when something signifigant has changed in order not to flood it, but still update it regularly as new ideas or discoveries come.

a question: do you already know all of this? judgind from the lack of comments i feel that there isn't anything new said, but i didn't see random walkers evaluating nearby intersections and being given a destination at spawn time mentioned anywhere in caesar 3 sites and the forum or applied in block design, although it is a very useful piece of information. (or i missed something blatantly obvious again, or, in the worse case, am wrong )

lemmus,

I didn't know about the initial "destination" piece of a random walker's walk, although it doesn't surprise me. I've read only part of what StephAmon wrote, and your observation seems to fit in with what I recall.

I can't answer for anyone else, of course, but I feel that the techniques I've mastered are sufficient for the city-building I want to do (Pharaoh at the moment). Since I play Pharaoh at Very Hard difficulty and want to always avoid fire, I need for most buildings to be passed by a fire marshal frequently. Therefore, I can't build large blocks, and don't need to know precisely how a walker walks.

Assuming that your posts are somewhat shorter than StephAmon's, I'll probably read them, especially when some of the TBIs are filled in.

By the way, if the definition of visual domain is too long to repeat, would you indicate where StephAmon defines it?

[This message has been edited by Brugle (edited 10-23-2004 @ 02:02 AM).]

brugle,

the visual domain of a building is centered on its northern tile and consists of the surronding tiles which are verified for the presence of a road network when a "random" walker is spawned

StephAnom discusses domains in this tread starting from reply 6. He does not call a domain "visual", that was my addition.

I have updated my example to show how a block can exist without having a road connection to a market
Also,

an interesing observation!
When the "random" walker is on patrol (stage 2) and he passes through an intersection he makes a choice of direction. If the path he has chosen has 3 or less tiles, his next direction as he returns to the intersection is predetermined, if the road is:
1 tile long - he chooses the next clockwise direction
2 tiles long - the opposite direction
3 tiles long - the counter-clockwise direction

Test the following trap:

Legend

If a random walker is patrolling and he comes along the road to the plaza sqares he will remain there until he heads home! A granary example follows (posted before i edited this post)

btw. is it well known that a warehouse's tower can be used as a gatehouse and will not block the road to rome?

[This message has been edited by lemmus (edited 11-22-2004 @ 04:22 AM).]

Very well done Lemmus!

Judging by the amount of effort you've put in in finding out the logic behind random walking, i say you've done a pretty decent job here. Though i haven't read nor bother to understand what most of the posts here are about (hope that doesn't sound rude :P), i have a few comments to make here.

What you mentioned in reply #8 about this post not having a lot of comments from others is incorrect, judging from my perpective. I don't know about others, but for me i never thought of finding out an explanation for the intricate way in which decisions on which path to take in an intersection are based on, so we don't have interest in this topic. Such knowledge can be considered 'extras', since for most of us (for me at least), we build cities as things go, prefering to keep the number of intersections to a minimum instead of setting out on an algorithm to find out the determining factors behind choosing pathways. Cities that consists mainly roads without intersections are aplenty, and most are very well built too.

In building such cities, roads are carefully laid out so as to make sure both sides of it are fully untilised for buildings. Spaces in between are also kept to a minimum to allow for better usage of land area. Where intersections are a must, gatehouses come in, and they also have the advantage of reducing walking distances from workshops to warehouses.

From what Brugle has said, he doesn't need to know precisely how a walker walks, but that doesn't stop him from conjuring up wonderful cities , cities that are both mind-boggling and excellently laid out. Such is proof that knowing precisely how walker walks is redundant in good city-building, unless of course your ultimate aim is to maximise the servicing range of buildings. I'd say this would help a lot in getting a place on Clifford's 50,000 cities list, but other than that, standard knowledge are sufficient to get you through all of the missions easily.

However, if you decide on carrying on this research, you still have my support . I've never been able to create a city even close to having a population of 50,000 without reaching the sprite limit. You might want to attempt that after you're done finding out the randomness of random walkers, and who knows, maybe you could even exceed the current record which by the way is help by Joshofet with a city of 123,011!!

lemmus,
The thread Warehouses as roadblocks? discusses destination walkers crossing the north tile of a warehouse. Reply #4 suggests that some random walkers may cross the north tile of a warehouse (like a garden).

[This message has been edited by Brugle (edited 10-23-2004 @ 12:57 PM).]

I'll be busy for a while so there will be few or no tests for some time. For now, i'd like to point out two "tricks"

First, you can use zig-zag roads or gardens to nearly double your cartpusher speed when they have a long way north, south, east or west to go. This is probably most useful in providing food and making a raw material-workshop-warehouse-dock chain stretch even longer, but you can make any destination walker go diagonally in a compass direction.

Example. Normally you can only build non-wheat farms at this distance. With the gardens added even a wheat farm will produce at full efficiency.:

Legend

Second
If you have a granary but have not added roads or have only 1-2 tile roads on any of its sides this is what your random walkers will do while patrolling when they turn back after taking a side with
- no tiles added: they will take the next road in a clockwise direction
- one tile added: they will continue straigth on the intersection taking the opposide road
- two tiles added: they will take the next road in counter-clockwise direction

These results while not absolute are very definitive and will work in most if not nearly all cases when the walker is in patrol mode. Knowing this gives you some control over which roads are preferred by random walkers.

Check the following example, which is a very bad idea as 66% of your walkers will be "captured" by the granary. I certainly remember building a few of those "magnetic" granaries but that will never hapen again (the oracles stand for any buildings you have decided to add)

Legend

[This message has been edited by lemmus (edited 10-25-2004 @ 01:57 AM).]

Thanks for experimenting, Lemmus. It does look interesting... quite new to me, in fact.

Where will a random walker head to?

imagine a 13x13 square whose center is the 8th from the spawning building's northern tile in a NE, SE, SW or NW direction. 05 is the square's center, and the walker will head to first road tile it finds scanning the square in the following pattern. N is top left. Extend the pattern to fill the whole 13x13 square. The next clockwise direction is taken for the next walk.

10 11 12 13 14
15 01 02 03 16
17 04 05 06 18
19 07 08 09 20
21 22 23 24 25

The walker will act like a destination walker until he reaches his target but will turn back if he traves his tile limit as specified in this page on the Appian way (the short patrol) unless he is an entertainer, an engineer or a tax collector. If he is one of the latter he will walk the total maximum travel limit of a forced detour walker which is hundreds of tiles, though i have not tested how many exactly, and if he exceeds his patrol limit in reaching his target he will immediately teleport home upon reaching it. This should give the long overdue answer to joshofet's question from this tread. If the walker is supposed to pass a gatehouse the process is terminated when it is reached and he goes to the next phase. It may be interesting to note that this explains why we observe "short" and "long" walks, and that the current public understanding of them is not accurate. Another thing to note is that if the walker cuts a corner at reaching his destination tile from phase 1 he is free to choose a direction and may turn back, while if he is walking in straigth when he reaches his destination this is not possible.

If the walker reaches the target without exceedin his patrol limit, the travelled tile counter is reset and he switches to patrol mode. As soon as the patrol limit is reached the walker heads home as a destination walker and this time he will walk through gatehouses.

There is a great number of implications and tricks connected with this as well as some extra not included here information, but it will tike quite some time and playtesting to write them out.

[This message has been edited by lemmus (edited 11-21-2004 @ 07:17 PM).]

Hi guys,

Firstly, I must say that you all posting here are doing a great job. If I was a CBG developer and had been aware of this forum, I definitely would try to hire some of you. May be you already are in such a position, I do not know...

I had C3 for a couple of years and the first time I buyed the game, I reached I think to Miletus in the Career Game missions and then, for some reason, I quit playing.

A few weeks ago, suddenly I came across the CD again when digging into my archives and thought "Hey man, that was a pretty game, why not play it again" and installed it to start playing. That's when all began...

Since we are, or at least I am, now more familiar to Internet, I started to look around whether I could find some info about the game and that's how I came across here and all of the other valuable sites, like The Appian Way and Grumpus's site. Now all I do in my spare time, is to look around and see whatever info I can get about the game, housing blocks, walker behaviour and else.

The reason I am writing to this topic is that, among all the other questions I now have about the aspects of the game, the most interesting was the walker behaviour in lemmus's city, Golemium I think. How can it be possible for an engineer to travel so much distance? And of course, how can it be possible for maintaining a city with +100k population with only 4 engineering posts? My engineers usually turn back or dissappear after their max. walk distance. As far as I could see, there is no loop in the design (or am I missing one).

Of course I know that this info (and also other main questions raise in my head after reading the topics) are not needed for finishing the game by regular play. I mean I know the missions can be finished without using these tricks and by using proper blocks to maintain the city stable. And I also know that this is an old topic and as far as I could understand, lemmus is not writing to this forum anymore. But "curiosity killed the cat"... Once I have a question in mind, I feel I have to find the answer...

And probably I will be around to ask many questions, if it doesn't bother you of course...

adigordon writes:

How can it be possible for an engineer to travel so much distance?

Your question is actually answered in this thread (especially reply #15), although without some background knowledge it might be hard to understand. I'll try to summarize the basics.

A "random" walker's walk is composed of 3 parts: first it walks directly to a particular tile (which can be different on different walks), then it wanders around in a somewhat random way, and finally it returns to its building or disappears. In most cases, a "random" walker can go only a fairly short distance in the first or second part, but when returning to its building it can go quite far (hundreds of tiles). One of these "random" walkers can be made to go a very long distance by having only a long route from the road tile where the walker starts to the road tile where the walker finishes.

However, engineers (and a few other "random" walkers) behave differently. An engineer cannot be made to go a very long distance in the way described above (at least, he cannot do that without rearranging the roads during his walk), but he can go a long distance during the first part of his walk. I don't think I've looked at the city you mentioned, but I'm sure that lemmus designed the roads so that the engineers went a long distance on the first part of some or all of their walks. (I verified that this would work a while ago, but have never used it in a "real" city.)

I hope that that satisfies your curiosity.

adigordon writes:

If I was a CBG developer and had been aware of this forum, I definitely would try to hire some of you.

too bad that you are not because i am in fact looking for an (interesting) job right now

welcome to the forum. i am very glad to see someone who is interested in the mechanics and is willing to do some searching, so shoot all the questions that interest you, and i'll do my best to answer. speaking of which - it is past 2am here, so this reply will certainly not be of the highest quality while there have been periods of several months during which i did not visit the forum, at present i am checking it regularly. i'll probably be posting as trurl however, trurl=lemmus.

last month i started to write an article on walker bahavior that may be interesting to you. i'll try to finish it faster now that there is someone to make use of the information right away but there is plenty to do here these days, so it may still take some time.

before we move on to engineers, a warning about an incorrect myth that may get you confused if you read about walkers on the forum, the appian way, or anywhere else: there are no "short" and no "long" patrols! the algorithm used for generating all four patrols is the same, and they are of varying length which depends on the nearby road network.

in golemium there are 8 posts in fact, 4 for each of the mega-blocks. If you take a closer look you will see 4 plaza tiles about 8 squares away from the posts in each direction - these are the tiles that the engineers and the actors head to when they are spawned - and this is the first part of the patrol that brugle speaks about and which is described in reply 15. both engineers and actors can travel for hundreds of tiles in this first part of the walk, as opposed to most other walkers who will turn back after they exceed their patrol limit (which is the "short" one on the appian way). the four winding roads that start from the posts are and end at the plaza tiles pass around every building in the block.

i timed the building of the posts so that the engineers would be out of phase, with maximum distance between them. (surpizingly they desyncronize after time even though the desing was such that they shouldn't; we discussed the reason for this with brugle in another thread). note that in this way it may be possible to achieve coverage with just three posts per block, but while i havent tried it i suspect that the roads in golemuim are a little longer that the maximum that can be covered with 3.

something else on the topic: golemium may soon be surpassed in size. i did some thinking on how to achieve the absolute maximum recently and designed a few blocks and road connections that could be used. however the question of what the biggest possible city would look like is not trivial: lots of creativity will be needed to achive the optimum balance between cutting the road length (which can translate in a couple less service buildings) and a increasing the small casa/fountain ratio (which can mean about 2 fountains and a reservoir less). while i do not yet have a clear picture of the city plan, there will be 8 fountains per reservoir for sure, and if possible, i'll try to fit everything in a single block (in order to use only 1 theater and possibly 1 less engineer posts).

[This message has been edited by lemmus (edited 04-19-2007 @ 07:43 PM).]

Hi guys. Remember me?

I never knew there're 3 phases to a random walkers's path; I thought there're only 2!

Anyway, I'm wondering something. When Brugle mentioned " A "random" walker's walk is composed of 3 parts: first it walks directly to a particular tile (which can be different on different walks)", does it mean that the walker will walk to that tile regardless of the actual road distance between the spawning tile and that tile? My experiences tell me otherwise, and that's why I'm hoping you guys could shed some light on this area. Thanks a million.

hi

the light has already been shed in reply 15:

lemmus writes:

The walker will act like a destination walker until he reaches his target but will turn back if he traves his tile limit as specified in this page on the Appian way (the short patrol) unless he is an entertainer, an engineer or a tax collector. If he is one of the latter he will walk the total maximum travel limit of a forced detour walker which is hundreds of tiles, though i have not tested how many exactly, and if he exceeds his patrol limit in reaching his target he will immediately teleport home upon reaching it. ... If the walker is supposed to pass a gatehouse the process is terminated when it is reached and he goes to the next phase.

Further to recent discussions both here and in Pharoah Heaven, I have been running a few tests. Many of these, alas, were wasteful because I was investigating matters which, unknown to me, had already been reported elsewhere, but at least I could corroborate much of the information provided by lemmus/Trurl. In fact, I can now rationalize his observations, as well as explain why his theories, while fairly reliable, occasionally do not work. Most of the credit must go to him.

I have chosen this thread because it contains most of the history behind Trurl's theories. Readers should also refer to Puzzled by random walkers?

On visiting Pecunia's site I came across the implication that a walker's choice at an intersection might be influenced by the same random number 'grid' (stored in the .map file or, in a compressed form, in the .sav file) that is used to determine, among other things, whether four tents arranged as 2x2 will merge into a single building. I modified the CCK 'Caesarea' map to fill the entire grid with zeros. I also zeroed the grid which controls some of the terrain 'randomness' to remove any effect that might have (if in fact there is any).

Not surprisingly, all locations become potential merging sites. All terrain tiles display the same graphic, immigrants appear on the map two families at a time at exact 50-tick intervals (suggesting that their precise timing in normal play is slightly randomized to break up obvious patterns) and all initial labor seekers execute a NE quadramble first.

Walkers behave in accordance with StephAmon's predictions for the outbound portion of the walk. On conversion to random mode they begin with a preference to go NE. One tile into the random part of their walk they switch their preference to SE. Every two tiles thereafter the preference rotates clockwise.

When they encounter an intersection they will take their preferred direction at that point in time if it is available to them, i.e. if there is a road going that way and it is not the road they have entered the intersection from (no reversing). If the preferred direction is not available they will take the first available direction clockwise, but their preference at the next intersection will be the same as it would have been if they had been able to go their preferred direction in the first place. In other words, the cycling of preferred directions is not disrupted by being unable to go that way.

I have yet to investigate the effect of grid numbers other than zero (I wanted to get this preliminary information out) but I am expecting that two or more bits from those numbers are used to break up the cyclical pattern. Trurl's observations relating to a walker returning to the same intersection reveal a pattern because the tile on which the intersection sits is 'seeded' by the same random number. When all randomness is removed, we no longer need to return to the same intersection to see the patterns he reports.

Thus we can rationalize Trurl's observations:
The direction they will take at an intersection depends on two things - the tile itself and the number of tiles the walker has passed since he began the "random" part of the walk.

This is accurate. 'The tile itself' is seeded by a random number. If that number is zero, the preference will be NE + (tiles/2) rounded up. By that I mean a three tile walk /2 rounded up = 2. 2 steps clockwise from NE is SW.

If the walker comes back to the intersection after he has travelled one tile forth and back, the algorithm will output the next clockwise direction to the one it previously output.

That is correct. If, for example, he turns SE in accordance with his preference he will travel 2 tiles (1 each way) and choose SW. If he arrives from NE and his preference is to go NE, he cannot do so and he will go SE instead. When he returns to the intersection his preference is now SE - the way he's just come - so he goes SW. Same result, different way of arriving at it.
Edit - I garbled this the first time - hope it makes more sense now

Similarly, suppose he approaches a 4-way intersection from NE and turns into a 2-tile stub to SE. Trurl's prediction would be that on returning to the intersection he will go NW (two turns clockwise from SE) and that is usually the case, but if he was actually wanting to go NE when he arrived (and unable to because of the no-reversing rule) he will return to the intersection in SW 'mode'.

Many of us will have noticed that a walker who arrives (in destination mode) at a walk target on a diagonal path (such as on cutting a corner) will often seem to 'reverse'. With all randomness removed we see that he will always go NE in that situation, regardless of the direction he arrived from. If NE is unavailable, he will go SE instead.

'Default' walkers always start with a NE preference, so if the walk starts in an intersection it will go NE if possible. I have not studied 'false starts' (which, in my experience, can occur in dead ends and intersections if the quadramble is defaulted due to absence of road tiles, but not if the default is due to disconnected road tiles). I believe that they occur only when NE is not available, but I may be wrong. What I find odd is that StephAmon observes, in reply #10 of his Pharoah Predicting Roaming Walks thread that 'dead end' false starts seem only to occur when NE is unavailable even with randomization.

Unlike Pharoah, Caesar 3 seems to handle a road network differently where it comprises a mesh of interconnected road tiles, such as on double width roads or large plaza squares. Here, a walker approaching a target diagonally may continue on that diagonal line (even though theoretically now in 'random' mode) until it reaches the edge of the mesh, whereupon it will go NE. Not all intersections within the mesh appear to be regarded as such and I haven't made much sense of walker patterns.

[This message has been edited by Trium3 (edited 06-23-2008 @ 04:53 PM).]

Trium3,

Interesting observations. (I started to comment on those which surprised me (a little), but desisted.)

a walker who arrives (in destination mode) at a walk target on a diagonal path (such as on cutting a corner) will often seem to 'reverse'. With all randomness removed we see that he will always go NE in that situation ... If NE is unavailable, he will go SE instead.

The qualifier "With all randomness removed" suggests that with a normal map (that has "randomness"), a walker arriving at a walk target on a diagonal path may prefer a direction other than NE. My experience is that such a walker will always prefer NE (with SE second choice). I've examined this situation (with "grounded walks") in several (perhaps 5) locations, so I'd be surprised (although not extremely surprised) if it is affected by "randomness".

[This message has been edited by Brugle (edited 06-23-2008 @ 12:16 PM).]

lol, this topic has been active for four years. Haha!

Brugle,

I was not particularly suggesting that randomness would directly affect that particular property of a walker, merely emphasising that in my tests any randomness there may be has been removed.

I am picturing in my mind my recent encounters with diagonal 'reversers' and I cannot remember one that contradicts what you say. I have a few in my current 'Valentia' project and they all 'reverse' to NE or SE (with one bizarre exception - see below). If your observation is correct (and I don't doubt for a moment that it is) that would suggest that NE is the preferred direction for any walker's first tile of travel whatever random seed is used (I don't see why a walker undergoing a mode conversion on a diagonal path would be a special case). I expect that to be established one way or another after tonight's scheduled experiments (which this time, hopefully, will not take me until 5.30 am).

My reference to StephAmon's comments on 'false start' walkers and their behaviour "even with randomization" further supports the idea that a walker's first tile of travel is simply fixed to a NE preference (and clockwise 'next bests') without modification. Indeed, the standard 'default' walk will begin towards NE if possible.

As for that exception - I mentioned the point about diagonal walkers on a 'mesh' passing through the target tile and continuing diagonally without changing direction until they reach the edge of the grid. I have seen examples of walkers grounding to a tile on the NE edge of the grid (travelling east) where you would expect them to turn SE, but they turn south! One would not have expected that choice to be available to them.

I don't fully understand, but I've said a few times that the game doesn't seem to treat a dense mesh of intersections as separate junctions requiring choice. However, these are rare in normal gameplay. I merely point out, since we discussed some tests of yours elsewhere, that (in Caesar 3 at least) mesh-like networks do not produce 'normal' behaviour and tests on them are likely to be misleading.

It is very satisfying, after all this time, to sit and watch walkers twist and turn through perhaps a dozen intersections and to be able to accurately predict every step from spawning to return. Of course, I can only do that because I've removed an element of variation in the algorithm. It's not of much practical use except, as Trurl says, in the case of walkers making multiple passes through the same intersection and, as he notes, the common feature in Caesar 3 which would cause fewer problems with the application of knowledge is the granary. If Pharoah works the same way we might ponder just how long should those stub roads be where walkers can turn into booths, bandstands and pavilions.

Note - attention is drawn to a rewrite of one paragraph of utter rubbish in my previous post which I screwed up first time

Edit: Since I am not yet qualified to post back-to-back and I have more information I am breaking my own rule and starting a new thread. I suppose this one is over-long now anyway - it takes some time to load over my connection.

[This message has been edited by Trium3 (edited 06-24-2008 @ 10:19 PM).]