Difference between revisions of "User:V453000/PSML"

From #openttdcoop wiki

Jump to: navigation, search
(New stuff! :])
 
 
(One intermediate revision by the same user not shown)
Line 44: Line 44:
  
 
If you want to ask about anything or just kick me another idea, just do so on the IRC, I will be very pleased for sure :)
 
If you want to ask about anything or just kick me another idea, just do so on the IRC, I will be very pleased for sure :)
 +
 +
Idea realized in:
 +
'''ProZone Game 13 - Timed SML'''
 +
 +
http://blog.openttdcoop.org/2010/08/02/prozone-game-13-timed-aka-insane-sml/
 +
 +
http://wiki.openttdcoop.org/ProZone:Archive_-_Games_11_-_20#gameid_13
 +
 +
 +
Ultra mayhem mode in:
 +
'''ProZone Game 2013 - the PZG13 successor'''
 +
 +
http://blog.openttdcoop.org/2013/08/06/dont-try-this-at-home-prozone-game-2013/
 +
 +
http://wiki.openttdcoop.org/ProZone:Archive_-_Games_21_-_30#gameid_2013

Latest revision as of 06:54, 6 November 2013

Packed Shifter MainLine

This article/ text/ thing is all mostly a concept as I have not enough time/ patience / what ever else to test it alone. Please note that some of the information posted might be wrong. I will try to avoid them but at this state of development, anything can happen ;) . Anyways, enjoy your reading.


Here are savegames that I used while developing the stuff... I will not post screenshots but I will just order you to look things up the savegames as there are no screenshots for the things yet because even the savegames contain just a concept :p

Savegame01

Savegame02

The goal

This setup of network is trying to make the ML tracks filled with trains as possible (packed). The mean how to gain this is based on SML concept, enhanced for the: Hybrid Shifters, Fail-safe joins, Heart Timer and some other additions as shown below. Example for TL5: The goal is to have 5tile gaps between each train, not having any empty positions and keeping all gaps the same.

How should it look?

All of the idea is based on releasing all trains from all SLs at a certain time (let's call it Release Period for now), on certain place(s) (Timed SL exit) and of course - after some shifting of the previously joined trains.

The Heart TImer

This is a mechanism controlling all of the SL exits along the map. I think it is best used with LevA Gator with parameter 3500 (my settings) and just make a loop where just a little piece releases the trains. The releasing time must be very short as we need it to be precise as much as possible.

Release Period

This is a thing that connects the whole network together. It is a timer that launches all trains leaving any sideline at the same time onto the ML. This gives every train an exact position on the ML, allowing to shift exactly into packed formations. The Heart Timer does this job.

Timed SL exit

This is the exact place where the train can be released from the SL exit. I have used a very primitive solution that also fails quite a lot as it allows two trains at a time to leave. Needs improvement. I believe some logic mechanism will fix that.

Packing Hybrid Shifters

The thing I already wrote a wiki page about ... just without the word Packing... if we used the original version, the shifter would allow trains to break their set positions and therefore unpack the line they are shifting to. So there is just added the "insurance logic condition" which makes train shift just behind one that already is on the line. Also, the question is if we really need the big Hybrids which take so much space and do not have to improve the network so much in this case. I think the fail-safe shifters should do fine but the Hybrids are just a bit of an extra insurance that more gaps on ML will be packed. On the other hand if we used Fail-Safes only, there could MUCH more SL exits, resulting in more ML shifts, which could also lead to greater packing.

ML Look

As we want to keep an exact position of not just trains on one line but on multiple lines with relation to each other, we need to make ALL ML LINES SYNCED. This means, the whole ML must be straight (as the simpliest solution) or in the curves must be made a synchronizer which would make all lines synced.

Basic Conclusion

This concept seems quite powerful to me, especiall with Lev4 Chimaera I tried it with. I has huge restrictions to the building of the network though. Many things must be exactly for the trains you are using as we are using their acceleration, length and also a desired gap. This must be tested by try-fail method probably. With this is connected a fact that we need only one cargotype on one ML. OR we need to make trains behave the same (== acceleration, == length, == weigth). For example Iron Ore and Grain wagons have about the same weigth, resulting in very similar accelerations. Best is to use just one cargotype though. Next not-so-good point is that the SL exits can not have high capacity as the releasing timer unpacks the SL. This can be fixed with making each SL only for a small area (either play on a streched map so each SL has not much of space OR use the fail-safe shifters in order to make SL exits more often, resulting in less traffic on each). I believe we will reach very nice results with this and I am looking forward to testing :)

If you want to ask about anything or just kick me another idea, just do so on the IRC, I will be very pleased for sure :)

Idea realized in: ProZone Game 13 - Timed SML

http://blog.openttdcoop.org/2010/08/02/prozone-game-13-timed-aka-insane-sml/

http://wiki.openttdcoop.org/ProZone:Archive_-_Games_11_-_20#gameid_13


Ultra mayhem mode in: ProZone Game 2013 - the PZG13 successor

http://blog.openttdcoop.org/2013/08/06/dont-try-this-at-home-prozone-game-2013/

http://wiki.openttdcoop.org/ProZone:Archive_-_Games_21_-_30#gameid_2013

Powered by MediaWiki
  • This page was last modified on 6 November 2013, at 06:54.