MorpheusOS 1.8.2.0 Ultrapeers with LimeWire
 

I've been noticing lately that I'm connecting to more Morpheus ultrapeers than anyone else. Also while this is happening I'm also noticing I'm getting fewer search results. As to before to where I was connecting to only LimeWire and Gnucleus ultrapeers while tons of search results. I was wondering if anyone else has noticed a difference in this.

Yup, there seem to be many, many poorly connected Morpheus Ultrapeers. I don't know how they decide how to become an Ultrapeer, but their Algorithm has some flaws. When I force LimeWire to be an Ultrapeer, i've got hardly any leaf nodes (maybe 10 - 20). And almost none of them are Gnucleus or Morpheus.

If this doesn't change, Morpheus could easily make the whole ultrapeer-system useless, regarding the fact, that some 60% of all gnutella users use morpheus.

I think the problem with the number of morphues users is the fact that it USED to be an awesome network, and the users are loyal to that. Unfortunately it is now a second rate program that as Taliban mentioned is flawed. Hopefully one of two things will happen:

the promised 2.0 vesion of morpheus will be debugged to sufficiently merge into the gnutella network

or, the users will figure out that Limewire is much better, as I did

Morpheus ultrapeers
 

In the meantime, should we REMOVE the Morpheus ultrapeers on the Connections screen and try to get better (random) connected ultrapeers?

I don't think it'll help you very much. - Better wait for the new LimeWire 2.5 if the developers come up with a solution or if Streamcast updates its software.

Maybe its the RIAA and MPAA! :x

I am running Limewire 2.4.4 on a Dell 8200...I've noticed that when I am connected to just Limewire ultrapeers there will be over 1000 hosts yet when connected to MorpheusOS or Gnutella Ultrapeers there are 150 or less hosts..
Just wondering if other than watching the connections screen and removing the aforementioned ultrapeers is there any way to block these connections...
Klondike

I think something is already happening. Morpheus Ultrapeers are still blocking any information on how many hosts you can reach at the moment, - however their connectivity is increasing. I guess it's got something to do with the new Morpheus 1.9.1 update.

Umm
 

Freakin MusicCity is the biggest bunch of dorkahontases I have ever seen in my life.

They have screwed up Gnutella far better than any RIAA attempt could ever have:

http://groups.yahoo.com/group/the_gdf/message/7998

The only features I expect in Morpheus 2.0 is more spyware and third rate advertising - beware of this dying company as it flails about in search of capital.

Yeah, let's block Morpheus!

i left a post over in the gnucleus forums under the name of Predator containing 3 links to why morpheus should be blocked.

http://216.74.73.125/~c0re/gnuforum/...d.php?tid=2152

Yea, but when they came over with all those users they really kicked vinnies clustering/gain more marketshare plans to dust.
I say leave them alone as long as it still ****** vinnie off.

and let them **** everyone else off as well? thats a bullsh*t solution. morpheus is dangerous to the network. something needs to be done soon.

Something really needs to be done. The results I am getting are crap. Its like what the Gnutella network was a year and a half ago. Damn you Morpheus (shakes fists)!!!!

Everyone is attacking morphues I have asked a question ov kazaa and limewire several times and i still have not got an answer when limewire or kazaa are running in the background my keyboard gets locked out I am running windows xp this dose not happen with morphues I agree that morphues is not what it was but with each new version they are getting better

Why don't you go Download the source code from Morpheus (Freely availible since it's based on Gnucleus) and fix it yourselves and get MusicCity to put it up.

1. the morpheus source on their site is still the Gnucleus 1.6.0.0 code (from what i hear).
2. the code is broken. you cant compile it without errors.

i've been trying to tell people to stop using morpheus since PE came out (cause im a Gnucleus user ;))

Having used several Gnutella clients, gone over to Morpheus/Kazaa, and then come back to Gnutella when Kazaa shut out Morpheus, I've seen enough of both networks that I might be able to provide a few clues.

While Music City talked a good game about all the great new things that were coming to their client, there was absolutely no indication they even had a programmer on staff. All the code came from Kazaa (or whatever its parent company was called). Music City produced nothing on their own. They even stopped participating in their own support forum, the only real support coming from other users and one Kazaa programmer who stopped in once or twice a week.

The point being that when Kazaa pulled the rug out, Morpheus was totally unprepared and most likely had no one readily available to get things going again. That they were able to field an even slightly modified version of Gnucleus in the time they did is a minor miracle. So I really wouldn't expect a whole lot out of Morpheus as far as client updates go. I heard a talk given by the head guy at Music City and it was clear he cared about the users only as a means to put money in his pocket. He is one who will spend not one penny more than absolutely necessary to keep the peasants from storming the castle gates. That's a great attitude if you own stock in his company; not so great if you're using his software.

Also, Morpheus/Kazaa users were not particularly well known for being generous with their files. As I'm sure has occurred in the Gnutella world, there were long, frequently strident discussions in the Morpheus forums about how best to kill the leeches, and whether they should be tortured before being disposed of. The Morpheus/Kazaa network was largely sustained by a small percentage of users who shared massive numbers of files. The remainder of the users were split pretty evenly between those who shared nothing and those who shared a small number of files, probably what they'd downloaded in the last week or so. The observation that connecting to Morpheus ultrapeers results in fewer search results may well be less about problems with the Morpheus client than it is with the Morpheus users. The client can only report on what the users are sharing. If the vast majority of them are sharing just a few files, that could well explain the sparse search returns.

Actually Morpheus is not completely innocent. They even broke Gnucleus (giving every morpheus client the same servent identifier) before releasing it.

The servent identifier has to be unique, for example to send a push request to the correct node. At the moment other servents have to believe that thousands of Morpheus servents using the IP 192.168.0.1 are the same gnutella node.

I was wondering about that. I tried the Morpheus/Gnucleus software right after Kazaa dumped Morpheus and I noticed all those local IPs. Mine originally went out the same way. The thing that surprised me was most everything seemed to work regardless of all the 192.168.0.x IPs. I still haven't figured that out.

http://mitglied.lycos.de/mdsgeist/LimeWire.zip

LimeWire version with the added ability to block gnutella connections to certain vendors.

If you don't want to connect to Morpheus anymore, this will solve your problems, because it has an option to disallow certain vendors from creating to you by allowing you to specify strings that will be matched against any servents user agent.

Thank you fo the fix. I am getting tired of manually removing Morpheus connections, but with LW 2.5.3 it seems that is the only way to get a good number of hosts. Unfortunately for me, I suspect this .zip file contains a LimeWire.exe, and I need an OS X version. Please try to get this into the general distribution soon!

Do you have a patch for that modification? I really would love to put that into my build.

The patch....

oh and I believe you will also need the MessagesBundle.properties.

The file should actually work on Windows and Linux without any changes, since it's a regular zip package containing java packages and the scripts to start LimeWire on Windows and Linux. - Unfortunately I'm not familiar with MacOS X at all, but overwriting an existing LimeWire installation with the files in the packages should allow you to use my version on MacOS X.

Thanks! I have got it and I am looking at it right now...

is there any way to guage a well-connected ultrapeer by messages in/out or bandwidth?

no, not really, - LimeWire ultrapeers for example impose a bandwidth limit of 15k/s on outgoing messaging traffic, so you won't be able to tell any difference between most ultrapeers. The rest of the ultrapeers are likely to be dropped automatically.

Over the next few months results will become better, since the major clients are going to implement upload queues and partial file sharing (in a much more sophisticated way then eDonkey does). The upload queues will probably reduce the amount of requeries sent substantially (indirectly increasing your search horizon) and the partial file sharing is going to increase the availability, especially of large files.

Sadly, we will have to wait at least three or four months until those new features will be available throught the gnutella network. It is true that Shareaza already implemented those two and Gnucleus is already testing partial file sharing, but as far as LimeWire, BearShare and Morpheus are concerned (those clients make up for the vast majority of nodes) they are not yet working on that.