Ok, now that you "updated" the gnutella support. I've done tests. I used Phex so I can connect to 3 Morp420 ultrappers while I'm a leaf. I've done searches for things VERY popular so I can get hits on the first hop (leaves on morpheus UP, so likely morpheus leaves -50% of GnucDNA UPs leaves are GnucDNAs-). NEVER a Morph result popped up! Only bearshares, LWs and shareazas. I'll now tell LW devs to ban your POS again. I see what you do. You reserve Morph bandwidth to Noenet downloads so Morpheus gets others bandwidth + 100% of morph bandwidth. That is call bast*rd leeching!

Die streamcast.

__________________
Liens d'intérêt /Links of interest:
Gnutellaforums en français /The House's rules you have to respect / First search the forum, then create a thread / Free software alternatives! - Logiciels alternatifs gratuits!/

To be fair this might even be a GnucDNA issue as I don't see gnucleus 2.0, iMesh Trustyfiles results either.... But the stoopid Morph devs are responsible of the softwares they **** into their POS, that's it. They are even more guilty as they are the most used GnucDNA client as of now.

__________________
Liens d'intérêt /Links of interest:
Gnutellaforums en français /The House's rules you have to respect / First search the forum, then create a thread / Free software alternatives! - Logiciels alternatifs gratuits!/

Yes indeed--and he's mutilingual in multiplatform too

looking into it

Would be nice if you had some concrete evidence of this and noted what you actually search for.
For sake of argument and addressing this as a real concern, I started up a client and watched outgoing traffic with some files shared.
Connected to a Limewire 4.0.7 Ultrapeer I see outgoing queryhits, pongs, and vendor messages returned by gnucDNA to gnutella. I have attached a screenshot of this.
So what are you on about ??

Furthermore, what you state about 'reserving' all bandwidth for NEOnet is an utter joke. It's funny you think stating something with exclamations at the end of the sentence makes it true.
Morpheus by default opens 8 upload slots for BOTH gnutella and NEOnet and doesn't cap bandwidth to one or the other with any preference. Users can change this if they please.
Nice try, but morpheus isn't a leech.

"I'll now tell LW devs to ban your POS again" , luckily this means squat coming from you. But while you're at it, can you tell Greg or whatever wand you have that makes LW jump to also stop world hunger? thanks in advance.

All in all, we take feedback of any bad behavior or bugs seriously, to be good gnutella community citizens. Whether you believe it or not, using updated gnutella protocol was important to Morpheus and any client using gnucDNA for that matter. John Marshall put a lot of work into it so at the least you may in the future perform some tests that give back real concrete data before spouting off. when there's garbage coming from your mouth it makes it tough to want to respond, but if there are real concerns of bad behavior we ourselves didn't catch I can assure you it's in our interest to address them and so we do. Now that the latest Morpheus is using the latest gnucDNA with the streamlined g1 protocol introducing many more clients using it onto the network, it's not completely unheard of for there to be some potential bugs -- but what you claim surely isn't. Even so..we'll spend some more time looking into it.

by the way, using the word '****' and then calling Morpheus devs "stoopid"... hmmmmm. You should reconsider your position.

Attached Images

query_hits_g1.jpg (27.4 KB, 253 views)

__________________
..got to have kaya

Before you put too much stock into "et voilas" remarks you may want to think about what he actually proved and how?

In any event, as i stated in my previous post it'll be looked into further in case there is some issue.

__________________
..got to have kaya

Hahaha funny, a Morph dev finally trying to defend his app . Okay here is the procedure to reproduce the bug: download Phex at http://phex.kouk.de/, start it. Now get it to connect as a leaf to 3 Morph420 UPs. This may take a while and you have to discard a lot of others UP to get all 3. Search for something very popular in any type ie: outcast, metallica, matrix, harry potter etc etc. Look at the vendor IDs, see there are no Morpheus even with the hundreds of results. This bug can be reproduced at infinite. Why Phex? because LW doesn't connect to Morpheus UPs as a leaf. Why I did the test? Because I never see Morpheus results even though you get 10x the downloads of Shareaza, an other multinetwork client.

Haha again if other Gnet users can't see Morpheus files, who do you think will fill those default 8 slots? Morph IS a leech.

Haha yes I told Greg a magic formula to stop hunger I'm really tired of Morpheus leech and false promises. As stated in an other thread Morpheus can be read as VAPORWARE.

That's the problem, John does the job and you are sitting there making cash by spywares on others people work. And you don't care about gnutella or you would have stop using it since 2002 Two years and a half that you leech. You first wrote Xolox that was also destructive to the Gnet with overquerying, so don't lie here. You don't care about Gnutella, period.

No. You **** others people software that you bundle in a GUI. The bundling part is your dev work, the raping is the stoopid part.

In conlcusion, what I did is reproducable, everybody is encouraged to do it (phex and all) and tell you that you don't know anything at all. Your screenshot doesn't help at all, it is only the default traffic analyser from GnucDNA. Did you ever wrote something working by yourself? Sorry to sound harsh but you are obviously lying and coming from a shitty company like that, it makes me angry

__________________
Liens d'intérêt /Links of interest:
Gnutellaforums en français /The House's rules you have to respect / First search the forum, then create a thread / Free software alternatives! - Logiciels alternatifs gratuits!/

So I think about it.... You connected as a leaf to a LW UP (why?!!) Can't you see the problems seem to be coming the *****ing Moprh UPs?? Or even it could only be GnucDNA corruption of queryhits just as GnucDNAs (1.1.0.5 at least) are sending corrupted chunks of download wwhen you are able to download from them (they are referred by the download mesh, not by query hits).

Don't say I'm not helping you debugging...

Edit: here is the link of the thread about gnucDNA corruption LimeWire 4.1.4 Beta
I see John has fixed a THEX bug days ago so maybe 1.1.0.7 morph is using is exempt of that particular bug.

__________________
Liens d'intérêt /Links of interest:
Gnutellaforums en français /The House's rules you have to respect / First search the forum, then create a thread / Free software alternatives! - Logiciels alternatifs gratuits!/

Your posts sound great because they're filled with false statements.


First off, you weren't promised anything and surely not by me. So you are owed _nothing_. Sure, it's true Morpheus has focus in other areas not just Gnutella ( i.e. Neo Network), so calling it vaporware makes me laugh.

John naturally does most the coding for gnucDNA, it IS his project. I don't want to digress about other things Morpheus is working on since this is a gnutella forum, but if you think morpheus developers do nothing... just keep watchign as Neo Network grows ;-)

Anyway, there's no trying to convince you of anything -- obviously the past history of Streamcast whether i was part of it or not has made you biased enough to not even give current attempts to improve the software a chance.
That's ok though, we'll have the last laugh since Morpheus surely isn't going to die and your posts as much as you think they'll make Morpheus die... won't. ;-).


I'll respond to your remarks in the future if they're actually constructive, otherwise there's much better threads to read on Slashdot.

Also, i won't be using phex.. i showed that a gnucDNA leaf returned a queryhit to a limewire client. Why do i care about phex when LW is the dominate client on gnutella ( no disrespect to phex developer, but i won't be wasting my time testing other people's software).

Lastly.... i do agree there seems to be a lack of Morpheus ( gnucDNA) returned queries overall in the past and so we'll spend some time looking into if this is a problem or just because of the nature of Limewire's domination in the current topology. What I do know is we do return queryhits.

__________________
..got to have kaya

Now that was a false statement. I'm direct but hate telling false things.

You are selling a product. You are making money of my bandwidth. You are not giving back to what a P2P is named for: Peer to Peer I give, you give.

Fair. Neonet might become great. Now, you proved my point: get out of gnutella already!!! You have Neo. GnucDNA is non commercial software quality. Stop abusing our and my bandwidth!!Multinetwork clients sucks.

Well that's funny cause users are gonna decide. Even with the millions of newbies, you never know. At least, I know I wouldn't want to work for such a loser company.

Anyway you can't do anything about GnucDNA because you don't understand it, right I'm sure Swabby would like help from people making cash of his baby.

Afraid of showing my evidence?

Cool, look hard, it's your paid job. Truth is that GTK-gnutella only on linux returns 100x more results than you.... (well how do you say 100:0 in maths? )

Kayaman here is my advice: don't try to protect Morpheus reputation with lies as Morph' reputation is nulll. Better try to be receptive, helpful and wise to gain some reputation that you can be proud of.

Ciao

__________________
Liens d'intérêt /Links of interest:
Gnutellaforums en français /The House's rules you have to respect / First search the forum, then create a thread / Free software alternatives! - Logiciels alternatifs gratuits!/

Don't you get why i just showed that case, most clients ARE leaves so showing that it returned a queryhit is showing it works.
I didn't mention I also ran as a Ultrapeer and see tons of traffic passed on. I also ran my client returning queries and uploaded dozens of files since this last post as an ULTRAPEER only connected to gnutella. How on earth could clients have found them if it wasn't returning query hits ( alt-locations aside, but i some of these are unique just for testing)

I'll repeat it again though... will look at this in more depth and so will John. Until then there's no gain responding any further.

_If_ there is deeper problem in gnucDNA, we'll fix it , update it, and thank you for noticing ;-) If this is some problem with Phex to GnucDNA or vice versa, i can't say it has a priority at all.

If it makes you angry when you THINK im lying then you need to sort that out with yourself. I believe you're making some false statements mostly founded on misinformation or assumptions, but I'm not telling any lies. I've only responded with things i checked myself first-hand.
Regarding streamcast having bundles and blah blah blah which ****** you off so much and so suddenly we're the devil and anyone associated with it too ( which BTW, Limewire also did so your angelic ideas don't move me) , explain to me how Morpheus didn't help Gnutella again? as fyi I don't care for the bundles either but users can purchase an ad-free version just like LW and Bearshare do... yet Morpheus is evil! !! !! !! ( i think this makes it more convincing, right?)

Hmmmm .... something comes to mind called fighting for years in the U.S courts for the right of software developers to continue to create technology rather then just settling and closing shop like every other p2p client that was sued did before and after Napster .... oh but wait let me stop right there since it's so conveniently never something anyone here mentions.

/done

edit:
You make me laugh, exactly what i did was be receptive to an issue you raised on a late saturday night while im late to be somewhere..... tested briefly to at least respond and left it open about the fact we'd look deeper into this. Have a good weekend :P

__________________
..got to have kaya

et voilà, you are wasting your breath. After seeing some of the results of Zlatin's UDP crawler (and creating some statistics with my own crawler) I'm sure that there are not enough Morpheus hosts in the LimeWire sub-net to cause any problems.

If LimeWire switches to a new network structure in the future, with 100+ UP-2-UP connections and a TTL of 2, you will not have a single Morpheus peer left in your search horizon, so why bother?

Also, since Morpheus does not support the new UDP transfer protocol, they won't be able to download from many LimeWire hosts anyway.

Kayaman:
LOL The popularity you enjoyed in 2001 comes at a cost. From a technology borrowed to Kazaa. Any P2P company would have won the cases, so don't act as if you saved the world.

Trap_Jaw:
Trap_Jaw, I think the crawler doesn't count all GnudDNAs (ever been in that Gnucleus 1.8.4 chinese cluster??? It is HUGE.) And my uploads of big files tells me there are lots of Morpheus. This is confirmed by using Phex that has a hard time to find LWs but finds a lot of Morpheus peers. They ARE slowing Gnet. Of course with F2F transfers, their behavior will have less effect

KathW:
Allô Kath! My tone is done on purpose as I want Morph devs to think about what they are doing and maybe get a cat out of the bag

Ciao

__________________
Liens d'intérêt /Links of interest:
Gnutellaforums en français /The House's rules you have to respect / First search the forum, then create a thread / Free software alternatives! - Logiciels alternatifs gratuits!/

Trying the latest Gnucleus from CVS, there is a huge bug: Gnucleus (GnucDNA??) doesnt remember hashes of files accross launches... You know that when you want to test sharing 41 movies and 50 songs, it is a major pain (doesnt help to share content on Gnet either). I did send queryhits in statistics as a leaf to 2 morph420 UPs (1 per minute as an average) and a Shareaza seem to have found one of my shared file (upload). Finding gnucdna results was a real pain. I did found few old Morpheus 4.0.53 results out of hundreds results. This is not normal, I should have seen much Morph results and queryhits. Normally I would get swamped by uploads in seconds. I took 10 minutes to get one.

__________________
Liens d'intérêt /Links of interest:
Gnutellaforums en français /The House's rules you have to respect / First search the forum, then create a thread / Free software alternatives! - Logiciels alternatifs gratuits!/

That Gnucleus 1.8.4 cluster has maybe 500 ultrapeers at most vs. 70,000+ LimeWire ultrapeers? It may seem large when using Phex, because Phex (and most non-LW clients for that matter) cannot really connect to LimeWire ultrapeers. Not even 10% of the nodes you can reach with LimeWire (and that can reach you) are not running LimeWire. And most of those are either BearShare or Shareaza.