Hi Slav2,
Sorry to beat a dead horse, but I had an idea on how fleet members could communicate with each other.

I've been experimenting and learning AHK and I had an idea on how fleet members could "signal" each each other though the fleet broadcast window.

During "First run" you would need to open the Fleet Window > Broadcast Settings. Then, for example, set "Broadcast: At Location" to the color green, "Broadcast: Need Backup" as red, and so on. This would help with the pixel color detection. Also, you would have to find a spot to put the fleet window so that the client would be able to constantly check the "History" tab in fleet.

Once a message is detected by color the client would store it in temp memory, click the "Clear History" button, so that it can continue to monitor for new messages and also react accordingly.

An example scenario that I can give you is;

Client 1 (C1) = Fleet Commander/Booster
Client 2 (C2) = Miner
Client 3 (C3) = Looter

C1 warps to asteroid belt > waits for warp to finish > detects warp has finished > looks for asteroids > if yes "Broadcast: At Location" > waits for fleet members > detects fleet members in ranger (I don't know of a good way to do this) > starts fleet boosts

Meanwhile.....

C2 sits in station monitoring Fleet History tab > detects the green bar from "Broadcast: At Location" > undocks > right click event in Fleet History > warp to 0 > click "Clear History" in Fleet History tab > waits for warp to finish > detects warp has finished > looks for asteroids > starts to mine > when ore hold full jettison > clicks "Broadcast: Need Backup" to signal looter

Meanwhile......

C3 sits in station monitoring Fleet History > detects green bar from "Broadcast: At Location" > ignores event > clear Fleet History > continues to monitor > detects red bar from "Broadcast: Need Backup" > undocks > right click event in Fleet History > warp to 0 > click "Clear History" in Fleet History tab > waits for warp to finish > checks "Containers" tab for cargo containers > continues to loot cargo......

I still very new so it takes forever for me to code something, but I am working on it. I still don't have a good understanding of how the program stores temporary memory. This is directly related to the looter portion. Looter would receive multiple broadcast events and I don't know how to save it so that it knows what it has looted and what it still needs to loot.

My goal would be able to run a single client on multiple VMs. In theory this would reduce the lag and down time from switching to different EVE clients running, since each VM would only have to monitor its self.

Slav2 (or anyone else for that matter), I'd like to know your thoughts on my idea, if it is possible, and if I can help in anyway.

EDIT: Of this is assuming CCP doesn't change the current in game GUI too much so that we can work off of what we currently have. If CCP changes everything dramatically go ahead and shit on this idea and light it on fire.

Just a idea. If you are in an Orca or an Rorqual you have an excelente looter. You can have it doing the job of a looter using Tractor Been in the extras hi slots. If I remember well, orca 150,000 m3 and rorqual 300,000, more the fleet hangar and others. So it can hold a lot of ore and the belt will not last forever if you have then using mining drones with 5 skiffs with implants and max out. So I think it can have the initial looter job and then if needed you can add one that get the ore in the hangar hold, so your chance to lost cans will be a minor problem.

Honestly, what you're trying to do is duplicate the Tank features across VMs. Frankly, I think there is a better way to do that. Eve Pilot doesn't mess with the Eve interface because it is unnecessary. Just reverse engineer Eve Pilot's Tank mode to make the algorithm use a shared folder INI file to coordinate the clients.

Just my 2 cents.

The .ini file is just a simple text file that the program writes to save and recall data to future use. Since VMs are technically their own computer, its almost like one computer a file to another over the internet. Unless Slav2 writes a whole new messenger program or uses one that is currently available, and has a fast enough server to host it, I don't feel like that would be very time worthy cause for him.

Maybe if you allow remote access then one VM could control another VM to tell it where it is. But that is a whole other monster to tackle. Not to that it would leave your computer slightly vulnerable because you are allowing another computer to access it, even if you select which computers are allowed to and which aren't. I myself wouldn't be very comfortable leaving it open.

Edit: Maybe a program like this would be enough for the VMs to "talk" to each other, stackoverflow.com/questions/14432801/how...-for-an-instant-mes, because building and sending packets goes way over my head.

Look in my signature for an example of what I mean.

With your way, with the host PC being the same, yes you can designate the drive and folder that all the VMs share and use the same .ini file to locate where each other is an any time if the EP has it built into its code to recognize any given signal. I was talking about two completely different machines on a completely different network, have them use the already build game UI to communicate with each other.