Multiple signals on one wire (multiplexing)

This board is to show, discuss and archive useful combinator- and logic-creations.
Smart triggering, counters and sensors, useful circuitry, switching as an art :), computers.
Please provide if possible always a blueprint of your creation.
kann_
Inserter
Inserter
Posts: 44
Joined: Fri Apr 15, 2016 4:36 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by kann_ » Thu Apr 28, 2016 6:45 pm

Hello guys,
thanks for the cool thread. It really got me going with combinators.
I plan a quite big project related to multiplexing. To keep track of all the parts I would like to add them here. I hope you don`t mind. I used the updated multiplexer from the first post.

The first contraption sums up all the signals in a channel range and outputs them as a stable/clean signal. The channel range (negative) is input in the left two Decider Combinators (slight change in the blueprint compared to the image). The big pole is the connection to the multiplexing network and the small pole the output.
Screenshot 2016-04-28 19.17.40.png
Screenshot 2016-04-28 19.17.40.png (408.03 KiB) Viewed 5525 times
Sum Up Channels
The second one is my first build and could use some optimization, but it works. It looks for a defined resource and finds the channel with the highest amount of such resource. It will screw up a little if neighboring channels have the same resource, but it is configured to do three loops. So it will work even if the three highest resource channels are in a row. It also outputs a continues signal, green is the resource amount, yellow the channel, and red just for the memory cell (could be improved).
Screenshot 2016-04-28 19.33.05.png
Screenshot 2016-04-28 19.33.05.png (352.7 KiB) Viewed 5525 times
Find biggest
Feel free to criticize me or improve these solutions. Thanks :)

DragonHeart996
Inserter
Inserter
Posts: 46
Joined: Tue Apr 26, 2016 8:49 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by DragonHeart996 » Fri Apr 29, 2016 12:01 am

I believe that it should be possible to implement a passive multiplexer/ deplexer with no clock and unlimited channels by sending a 'header' signal with channel # followed by the data to be sent, and then a header for the next channel. So:
channel 1 transmitter receives header signal '1' and starts sending data
Ch1 receiver sees signal and starts listening
Ch1 data sent
Ch1 receiver finishes listening after x ticks and sends header signal '2'
Ch2 transmitter starts sending
Ch2 receiver starts listening
...

If this same idea as built by me in minecraft redstone would help (ignore simulated output, repeater locking doesn't work in the simulator, pretend the output gives the sent data, and pretend each wire is a separate signal in factorio) with the ready of the receiver feeding into the start of the next channels transmitter:
http://mordritch.com/mc_rss/5750

User avatar
Dr. Walrus
Long Handed Inserter
Long Handed Inserter
Posts: 95
Joined: Fri Nov 20, 2015 6:30 am
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by Dr. Walrus » Fri Apr 29, 2016 2:56 am

DragonHeart996 wrote:I believe that it should be possible to implement a passive multiplexer/ deplexer with no clock and unlimited channels by sending a 'header' signal with channel # followed by the data to be sent, and then a header for the next channel.
The clock serves the purpose of sending the header signal. The reason why these designs are built with a clock is so that you could send information from all of your outposts in a relatively quick time frame. The designs with a clock can also handle nearly an infinite number of channels, but in some designs each channel takes 3 ticks, so a receiver in a multiplexer with 20 channels takes 1 second to refresh its output. If you wanted 2000 channels, it would take 100 seconds for each receiver to refresh its output. The only limit on number of channels is time.

All of the multiplexers in this topic could be converted to clockless multiplexers just by removing the clock. Instead of a clock, the player would manually send a specific channel signal into the multiplexer line. The transmitter that matched with that channel signal would send back its information. That would work perfectly fine, but it would not allow for automated site management because the transmitters would only transmit by player input and you couldn't implement automated monitoring or anything of that nature. So I think the clock and an 'Active" multiplexer is the way to go.
Visit my profile on mods.factorio to see some unique, off the wall mods: Mod page for madmaster5000 (me)

Older mods:
Pressure Plates

Neotix
Filter Inserter
Filter Inserter
Posts: 597
Joined: Sat Nov 23, 2013 9:56 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by Neotix » Fri Apr 29, 2016 7:00 am

In my factory I'm experimenting with multiplexing and started build some king of hybrid.
Clock control only transmitters (simple sending order) that send two type of signals (address and data). Data can be also multiple signals.
In that way I can send data to specific receiver.
It's not perfect because that system is fragile (one mistake and all data can be messed up) and require ticks table to control signals routing.

First implementation is Taxi Train. I want to build stations that will send request for train when player stand on pressure plate.

Next I want to build entire system that will send request for trains. In that wat all trains will be universal and when some factory section need items, it will send request and train will pick up them on one station and deliver to another. No more fixed schedules. But it's long distance goal with many problems to solve.

DragonHeart996
Inserter
Inserter
Posts: 46
Joined: Tue Apr 26, 2016 8:49 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by DragonHeart996 » Fri Apr 29, 2016 11:08 am

Dr. Walrus wrote:
DragonHeart996 wrote:I believe that it should be possible to implement a passive multiplexer/ deplexer with no clock and unlimited channels by sending a 'header' signal with channel # followed by the data to be sent, and then a header for the next channel.
The clock serves the purpose of sending the header signal. The reason why these designs are built with a clock is so that you could send information from all of your outposts in a relatively quick time frame. The designs with a clock can also handle nearly an infinite number of channels, but in some designs each channel takes 3 ticks, so a receiver in a multiplexer with 20 channels takes 1 second to refresh its output. If you wanted 2000 channels, it would take 100 seconds for each receiver to refresh its output. The only limit on number of channels is time.

All of the multiplexers in this topic could be converted to clockless multiplexers just by removing the clock. Instead of a clock, the player would manually send a specific channel signal into the multiplexer line. The transmitter that matched with that channel signal would send back its information. That would work perfectly fine, but it would not allow for automated site management because the transmitters would only transmit by player input and you couldn't implement automated monitoring or anything of that nature. So I think the clock and an 'Active" multiplexer is the way to go.
Yeah, I kind of realized that later. Do remind me never to post when I'm tired.

SJCubed
Manual Inserter
Manual Inserter
Posts: 4
Joined: Mon Mar 28, 2016 2:01 am
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by SJCubed » Sat Apr 30, 2016 3:15 am

Neotix wrote:In my factory I'm experimenting with multiplexing and started build some king of hybrid.
Clock control only transmitters (simple sending order) that send two type of signals (address and data). Data can be also multiple signals.
In that way I can send data to specific receiver.
It's not perfect because that system is fragile (one mistake and all data can be messed up) and require ticks table to control signals routing.

First implementation is Taxi Train. I want to build stations that will send request for train when player stand on pressure plate.

Next I want to build entire system that will send request for trains. In that wat all trains will be universal and when some factory section need items, it will send request and train will pick up them on one station and deliver to another. No more fixed schedules. But it's long distance goal with many problems to solve.
I actually made a very similar system to what you want to achieve with the requester train. I have a train dispatch system that takes in information from an outpost and the dispatch station sends out a train waiting there. The train then gets filled up, goes to an unloading station, and goes back to the dispatch system and waits for another request.

https://docs.google.com/document/d/1HiF ... M91F4/edit

Here is a readme that I have made. Just a warning though, there is one bug that the memory cell that holds how many trains are sent doesn't really get subtracted in the blueprint. I have already made a fix but I haven't gotten around to update the blueprint and the readme yet. I could probably send you the new blueprint or you can join my server and see how I've set it up on my map.

User avatar
siggboy
Filter Inserter
Filter Inserter
Posts: 988
Joined: Tue Mar 29, 2016 11:47 am
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by siggboy » Thu May 19, 2016 12:40 am

SJCubed wrote:I actually made a very similar system to what you want to achieve with the requester train.
Apparently around this time, just about everybody is working at a version of this idea.

So now we have kann_: https://www.reddit.com/r/factorio/comme ... _provider/
and Xeteth: https://www.youtube.com/watch?v=rF5tm0HbAGg
and my design: viewtopic.php?f=8&t=25311

and you and Neotix and terror_gnom is working on something as well.

Exciting times :).

I'll look at your google docs, you need to advertise this stuff more openly if you already went to the length of writing it up, so others can learn from it and help you improve the design. (Your system looks rather limited, with a fixed number of outposts.)
Is your railroad worrying you? Doctor T-Junction recommends: Smart, dynamic train deliveries with combinator Magick

kann_
Inserter
Inserter
Posts: 44
Joined: Fri Apr 15, 2016 4:36 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by kann_ » Sun Jun 12, 2016 5:54 pm

Not sure how I should call it, so let me try to explain the function.
All my train stations are connected to the multiplexing network and store how many trains are on the way to that station. I used one memory cell per station to achieve this.

To centralize all these cells I build a multiplexing memory cell that stores a time resolved signal. Every tick the green signal is send and stored. In my network it will store the green signal of all 60 stations in one cell and repeated the signal every 60 ticks.

If there is a better solution please let me know.
An example with 3 channels
Image of the circuit
Blueprint string

kann_
Inserter
Inserter
Posts: 44
Joined: Fri Apr 15, 2016 4:36 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by kann_ » Thu Jun 16, 2016 10:17 pm

I used the multiplexing storage cell from the previous post to make a train counter.
As soon as a train leaves a station it adds +1 to the destination station and -1 to the current station.
I build it as modular as I could, 7 combinators per station. There is up to one second delay (60 stations = 1 second) between train leaving and sending the signal so the values in the gif look a little delayed, but trust me: it works.
The chests represent the first 4 stations used in this test circuit, but on the multiplexing bus it is a green signal with the value representing the train count of each station per tick.
So if the first 4 stations have each 1 approaching train the green signal is this:
Tick - green signal
1 - 1
2 - 1
3 - 1
4 - 1
5 - 0
6 - 0
.
.
MUX train counter.gif
MUX train counter.gif (2.13 MiB) Viewed 5311 times
Blueprint string

User avatar
IngoKnieto
Long Handed Inserter
Long Handed Inserter
Posts: 96
Joined: Mon Oct 03, 2016 9:29 am
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by IngoKnieto » Thu Feb 02, 2017 5:42 pm

I know this thread is quite old, but I think the information is still up to date, so I post here...

I am basically using the Multiplexer example V.2.1 from the first post, and I relized something odd:

The first empty channel (hope that is the correct term here) always holds the data from the last filled channel for 1 tick.
For example:

Channel --- Input signal --- Data on Channel
[-1 to -10]
-11 --- 50 Iron --- 50 Iron
-12 --- 10 Copper --- 10 Copper
-13 --- empty --- 10 Copper then empty
-14 --- empty --- empty
-15 --- empty --- empty
(the clock would cycle from -1 to -15 in this example)

If the clock switches now from -12 to -13, the data on channel -13 will be 10 Copper for (I think) one tick, afterwards it will be empty. Maybe this is not even only happening with the first empty channel, but with every channel, and I just don't realize it, because the falsely stored values are overwritten immediatly by the correct values.
I noticed this behaviour, when I started saving the data on the channels in memory cells for debugging purposes (one MC for each channel. BTW: how do you debug your MP signals?). Memory cell #13 contains the same values as #12, although it should be empty.

It's not really a problem because I can configure the clock only to cycle through used channels, I was just wondering if anybody else experienced this too...

Oh yeah, and I said I am basically using the V2.1 example, because my clock is way slower, but I don't think that has anything to do with it.

kann_
Inserter
Inserter
Posts: 44
Joined: Fri Apr 15, 2016 4:36 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by kann_ » Sat Feb 18, 2017 3:59 pm

The clock in my builds I always had a lot more ticks that the channels that I used so I did not observe what you observe. But if your read the descriptions of V2.1 carefully you will see that the last threee ticks of the clock should not be used as channels.

I assume that is why :)

You can just let the clock run until 100 to fix the problem.

The easiest debugging tool would be to read all the channel in a central locations with the reader from the first post.
If all channels have the same data you could also have a look into my multiplexing memory cell. It basically reads one singal from all channels into one memory cell and saves it as different signals (channel1=yellow belt, cahnnel2=red belt...) Let me know if the blueprint string is out of date and I try to find the save game.

If you keep the input and output of the multiplexing cable as described in the first post. There should be absolutely no problems in my experiance.

User avatar
IngoKnieto
Long Handed Inserter
Long Handed Inserter
Posts: 96
Joined: Mon Oct 03, 2016 9:29 am
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by IngoKnieto » Sat Mar 04, 2017 4:48 pm

kann_ wrote:The clock in my builds I always had a lot more ticks that the channels that I used so I did not observe what you observe. But if your read the descriptions of V2.1 carefully you will see that the last threee ticks of the clock should not be used as channels.

I assume that is why :)
That could be it, I missed this in the description, thanks.

Regarding the channel debugging I ended up building one memory cell for each channel. That is a little space consuming, but I can check the signals in any channel at any time. And I can use the stored signals to do further analysis things like compare 'global' demand and supply of one resource in the multiplex network.

But I still have some bugs I can't solve. Like multiple trains going to one station, although the station is definitely locked by the first train. But what would a computer network without bugs be... 8-)

wolletd
Burner Inserter
Burner Inserter
Posts: 14
Joined: Wed Mar 15, 2017 8:04 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by wolletd » Fri Mar 17, 2017 12:15 am

Hello folks,

I wanted to share my shot on multiplexing with you. This is roughly based on some of the many ideas in this thread, but also contains a storage cell I created earlier that day:

Image

The storage cells require an input signal for at least 3 ticks to not brick themselves so the bus clock is 4 ticks. On the other hand, the storage cells provide stable, non-flickering output signals just like chests. I haven't found anything like this so I had to build it. I would like to have it more compact and/or work with 1-tick inputs, though. Main thing for me were those chesty output signals.
Detailed Images
Please let me know what you think about this. I haven't used this outside the test build yet, but it seems to work pretty well within it's limitations and my expectations.
Demo Blueprint
Transmitter Blueprint
Receiver Blueprint
Storage Cell

User avatar
DaveMcW
Smart Inserter
Smart Inserter
Posts: 2797
Joined: Tue May 13, 2014 11:06 am
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by DaveMcW » Fri Mar 17, 2017 1:13 am

Here is a different way to make a storage cell. The only limitation is it cannot handle input on 2 consecutive ticks, so make sure the channel increments every tick.

Set the white signal to the channel number in the constant combinator.
storage.jpg
storage.jpg (194.75 KiB) Viewed 4198 times

Ishakaru
Burner Inserter
Burner Inserter
Posts: 6
Joined: Tue Mar 14, 2017 8:10 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by Ishakaru » Wed Mar 22, 2017 5:20 pm

I've been thinking a lot on the subject before finding this thread. I needed a multiplexing solution, and every implementation I came up with required that I either configure each point myself (can't get further from acceptable), or a system that had the possibility of conflicts (new node requesting it's own "IP").

Both systems have the issue that I just realized reading this thread: The more nodes in the system, the longer it takes for each node to do something. This completely ignores the fact that each node doesn't need to send data every time it's queried.


The solution I'll be working on is what I call "priority multiplexing". I'll work it off a clock just like the OP: over a 60 tick period, 2 pulses: one for action one for received. Each node will have a previous and next node that signals are routed through. If it needs to send data, it will block the data from the previous node, and send it's own down the line. It will continue to send the data until it gets a received message. I guess this could work in a passive multiplexing mode, but I would be worried about signal noise.

I'll post pictures when I get it going.

Ishakaru
Burner Inserter
Burner Inserter
Posts: 6
Joined: Tue Mar 14, 2017 8:10 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by Ishakaru » Fri Mar 24, 2017 6:55 pm

Priority multiplexing: The basics still apply. A pulse is generated and actions are executed upon those pulses. The previous implementations used a round robin implementation. This had the weakness that a node had to be individually configured shortly after placement.

With my system there is 0 changes that need to be made per node. Once the blueprints are made, you can place as many as you want. The weakness with my system is that nodes need to be placed in serial. Parallel lines will cause feed back loops and generally cause chaos. The solution to this is to use branching nodes.
Basic node point setup.
Advanced node point setup.
Branch setup.

Chocki
Burner Inserter
Burner Inserter
Posts: 7
Joined: Mon Apr 11, 2016 1:28 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by Chocki » Mon May 15, 2017 8:49 pm

I have been trying to recreate the design by Dr Walrus found in the quote below.
Dr. Walrus wrote:Also I found a bug in the version I posted on here earlier which is caused when you input blue equal to the transmitter's channel. I fixed it and redesigned the transmitter and receiver both with 2 less combinators less.
*Edit* I changed around the receiver a bit and it now has 2 less decider combinators and 1 more constant combinator.
Picture and Blueprint String of New Design
For those that are unaware, it was ruined when combinators where changed to sending the output count of the selected signal, rather then the old way of sending the condition count out as the selected signal. In short the decider that checked if clock 1 (blue) was less then max count (yellow) outputting the green count (clock 2) as blue. If I lost you.. the clock wouldn't count up.

Well, I kind of fixed it. I added an arithmetic to allow for the count signal change. But I have a feeling this messed with the tick count. As I introduced an error. The decoder output now has a leak of the clock 2 signal, always equal to channel number +1. If anyone has any idea on how to solve the problem, please let me know.

Changes:
Changed signals to remove the need for alphabet mod. Blue=A, Green=B, Yellow=Z
Changed main wire color to green (Personal choice)
Added AC to clock to allow signal count pass.

Code: Select all

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

User avatar
AkariAkaori
Manual Inserter
Manual Inserter
Posts: 1
Joined: Wed Feb 17, 2016 11:41 pm
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by AkariAkaori » Sun May 28, 2017 5:28 am

Here are some multiplexer circuits I made in November 2016, which seems to have survived the Decider update when I loaded up an old save in 0.15.x

Videos of it working in 0.15.x:
http://plays.tv/video/592a5cf725b66990f ... circuits-1
https://plays.tv/video/592a5d8803e03275 ... circuits-2

Tutorial: https://imgur.com/a/elUs1

Blueprint string:

Code: Select all

0eNrtXVtv4jgU/iuVH3fDbuyEq3b3YVtpVW1nRmIuKl1VKIDbWoUEmdBdNOK/rwMzUy6B+PgSKPgFKSQ5cc7Jufo79lfUG07pmLM47faS5Bm1vr7+M0Gtf1YOs3Osn8TLvyfsMY6G2X/pbExRC7GUjpCH4miUHQ1onw0or/STUY/FUZpwNPcQiwf0P9TCc6+QQMRZ+jSiKevn0yDzew/ROGUpo8sRLQ5m3Xg66lEuHlJAykPjZCLuTuJsDIKi76EZalX8X6riKeI9U54Muz36FL0wcbm45pVOV5weLO6dZCceGJ+k3a0XemE8nYp/fgxkeUXlDi0fMEmjjKsVMdJkTHm0HAv6SdyQTNPxFEZyviQa0/6PceHs55FTGq8yiA0E98S1jPenLF0eZgJZOR2sn8bze0GcZNQ4HWzSCjZp3Wdj2ZIH2fdtbAkjY8p+aXyjYlQUODsYjSO+GFQL/a4miozKeCZGNhVq9cCTUZfFggZqPUTDCc0RFNkpKLwtiTzmBiDmHoS3/gZv/1DgLX2hfJY+sfhxD5NTPqUgZcCQrx3nfe3CFC2sV2vFWnpoGPWoeCv06erdxZ88iQb9aJJSftG+uq7cXLU/fxDXiBeaLARTC0NSrVYbPq6/Gjk/E7WsAd7FtMtCy7vrzps1ewu58++VOwPQnWTlztC4jf9uVnDZRp6s2/hfzdp4E1YZxsHSveSGk/z5EAyU9oFmzLRfwGkrHtDftNO/GfaB+8xzDmcJRAxbltzbRwtL2u7LYdJ/viCVT6z/XGiyMcRkg2JmaVsoFV8FqzHAgPGlKITNPf1oS/JT2/x6wp1hQm5UIWkSgGExOYDY/IzHE5rRAN6kLV7tgE/GSOyzKKGcGw00kpvDKB8xYOZtSCfYn5lWt+ThSQX1BXQwzLuv06rLaXoIDbWAX8nbLE/oKigByzlPNlWFwOzN+04b6hvK6megpYD5UqxBNSw4Uw2rQ5hfl3OBdZAKuTjGtqesQWRcU6tq3Vy3i4tYxFpGVOosAiSrwgf0DGHzbaVV1f2nmzvz//KzLuysVTlZV9Vm1oVd1gX1JaGhrCs0mHXVoFkXILLHLrKX0MkSI/uqYu6MXe6snjuH0KwLEPg7V2rNWNtN7mCZ9dnpn17WtS5ITOQUsKGSo1ieVuyoqNzdwVVul6SacpJoqqQVliVxu64SzSM1ag2ted58cWBfLWwoux54a95q3cKiBhgECss5DYwVIgNXEdTIsf39KVOeQ9kjZh/mnnZ+BkQteChbCzvmtbCjU5Xf4rcBJS0u6b6jo4TPLi7pcFhY2Q0c1sVhXRzWxWFdHNbFYV0c1sVhXRzWxWFdHNbFYV1OEevSpn3KxKvJgV5CE51bbeXOrUvlzq0vyp1bGauVerdc1dpS1ZpYScuC44NRHDqbxsB2atn4PDeaCG30mjmIk/FWNSMQp9BBnGxF615BVOApLZ1Q06i4FGAfHThGMxUgWstaKMHPag5+VjL8LNRSQAd/0dKwJoT5DSDy2yuY7CxCx7jZR9DsI2xu2AQgxsWgtqMcG8WwFZSNg3mUIWOQpjY0q11fgsJaV9U1eJXa4HVkxY8ACCXwJHF/MtkXNrrYnEubdeccy5mjJkbmqF2B7BTQKKErflhgfnA4pIErfmggDYiWGTWLNMAOaQBJrgpKv2rFj7rLv48qXNIHKryh9q/jsauafURNV+AoVYlqNkqR8GawMy0Eg7jf1Cwvde4Ky0s1E1Cqj8pQqvfKUKorZSjVMrYrYyHsU9vtwMgqBwD0DbHQ6nIWKCo5oHTRMjT6PcKBVqdCiYrj6y+A3tHbJcTIVHqo3n1wVjtVwFc3D2CV9IKCrOJa6R+p8Fj84q/siTJzR/XTmDtyWwy5LYbeZiOwntVw+xEd3X5EJs34zYf2daEJb5xffma31eUgyUBw3AsUkCNcoMB3Zqu05M1TiQcCVxWRdzMSngPQmW5jWYCTSbZBs/PyMChpty6RlzXzqFUWW8S+knxP038T/pzZEQ9F4iVfaPf7zoU7HjD/Hz30sVU=
Since the intended use case was loading correct items into trains:
- The memory cell is included in the receivers to prevent flicker when other channels are transmitting
- A 2 tick minimum for each channel is also required to prevent flicker that messes with requester chests when it updates the memory cell

Image

Image

User avatar
GalactusX31
Inserter
Inserter
Posts: 28
Joined: Thu Jul 20, 2017 11:27 am
Contact:

Re: Multiple signals on one wire (multiplexing)

Post by GalactusX31 » Mon Jul 24, 2017 7:21 pm

I do not know if this post in a dead post, but in my experience with the mechanic logic of other games, you do not need a clock to make a multiplexor-demultiplexor work.

The purpose of it is to know only the state of the incoming signals of 1 channel Or another, and be able to switch to another channel and know the signals of the same, not all channels at once.

I have designed a system to be able to send, using only a wire, multiple channels, and the signals that those channels contain, and no clock is needed.

Post Reply

Return to “Combinator Creations”

Who is online

Users browsing this forum: No registered users