• Welcome to Smashboards, the world's largest Super Smash Brothers community! Over 250,000 Smash Bros. fans from around the world have come to discuss these great games in over 19 million posts!

    You are currently viewing our boards as a visitor. Click here to sign up right now and start on your path in the Smash community!

An issure that probably has been brought up a bunch of times

cmu6eh

Banned via Warnings
Joined
Apr 29, 2012
Messages
356
Location
Parterre
How come online a lot of the older players still use pj64KVE, a lot of the new players use pj64K7E, and almost no one uses pj64k v0.34 (http://pj64k.blogspot.com/) ?

Shouldn't we all agree to just use the latest version of pj64k?

I have found that when playing people on different emulators ( even 2 people using KVE and K7E ) tend to experience higher rates of desynchs. They are probably not compatible because one runs in 32 bit and the other runs in 63 bit...
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
Technically, the latest version IS pj64k v0.13 core 1.4 (created by hotquik) which is the exact same thing as pj64kVE, a port of pj64 1.4 (created by Zilmar and Jabo) to increase compatibility with Vista when it first came out (also works with 7). However, it still uses the 1.4 core and version 0.13. Some people may not have compatibility problems on Vista or 7 with the regular pj64k, so they could use that instead. kVE was just offered as a slightly better alternative to help cope with the crashing (not perfected, though).

pj64k7E
The pj64k7E is, to my knowledge, the same thing as pj64kVE, and the guy just picked it up and started messing with the code in a hex editor to improve a few things, but it still remains as v0.13, core 1.4. So, technically nothing has changed and it should be no different from kVE. You can even look at the changelog, and observant eyes will notice that he really didn't do anything to warrant the new name and touting it as this great new alternative to kVE.

The only thing he fixed (supposedly; unverified), was that random error some people get when loading pj64k that says the plugins can't be located. Bear in mind that this is easily correctable by going to settings and choosing the damn plugins yourself. Takes all of 10 seconds. He might've also changed the way kaillera loads plugins, and will allow you to load different graphics and sound plugins without renaming DLLs. Again, this is something you can easily do yourself (rename any graphics plugin to Jabo_Direct3D7.dll or any sound plugin to AudioHLE.dll and you'll load those plugins when starting netplay). It also looks like he hasn't even included Owna's updated client (ew).

pj64k v0.34
pj64k v0.34 is smasherx's slightly more modest project, in that he didn't opt to rename it completely, and continued with the same version "edition" scheme (the proper lingo eludes me right now), improving from 0.13 to 0.34. Again, the things he improved are basically the same as pj64k7E, so it only improves convenience slightly. He also mucked about with the netplay client, renaming it and calling it differently, however still retaining Owna's client. Not of much concern.

Conclusions and Comparisons
To summarize, all of these seemingly different versions of pj64k are exactly the same. Regardless of what emulator you're using, when you start a game in netplay, the kaillera client lists your emulator as "Project 64k 0.13 (01 Aug 2003)". Right here you can clearly see they're all using the same 0.13 version the original "latest" version of pj64k uses. You wouldn't be able to play with somebody who used a different emulator. Nobody changed anything major, they just made small changes akin to grammar fixes. It is, in essence, a prettied up pj64k 0.13 core 1.4. There are no huge developmental breakthroughs that will decrease your lag or desyncs. They are exactly the same. Any changes you noticed in desyncs were purely coincidental and fall under some psychology term for you wanting to believe there is a difference and seeing it. There is not. Don't let anybody fool you that there is.

32-bit vs 64-bit
All of these emulator versions were designed for x86 systems (again, the version these are all based off is from mid-2003 when x86 systems were largely prevalent). So there will be compatibility problems with x64 machines, but I don't think there'd be much difference aside from a few weird memory errors or crashing. Certainly nothing game-changing like increased lag or desyncs. It doesn't matter if you're on on 64-bit 7 playing somebody on 32-bit XP; the experience will be the same as two people playing on identical machines.

Actually, just looked it up. I can confirm pj64k was initially designed for 32-bit systems, so everything discussed here is 32-bit.

---

The only reason we continue to use pj64k and pj64kVE is because they are tried and true and as "official" as you can get with unofficial software. Not sure about this pj64k7E Gaudy character, but smasherx is known to be "malicious" in how he goes about things, so as a standard rule of thumb, we ignore anything he outputs, lest it contain a virus or whatnot. You're welcome to try and use these newer versions, but remember it's at your own risk and we will not help or provide support for anything other than pj64k(VE).

Possibility of Change
The only reason we'd change is if hotquik continued developing pj64k. I'm not sure why they stopped where they did at core 1.4 when 1.5 and 1.6 were released shortly after, but it may have been due to source-code issues with Zilmar and Jabo, or they simply didn't have the time to continue such a project. Who knows, but the only updates we would accept as an improvement would be from the hotquik team, as only they have the source for pj64k 0.13 and only they can improve upon it.

pj64 1.7 incorporates a reliable netplay, so once that is released to the public, I imagine that will be the netplay standard.

Now, there is also Mupen! The latest version is Mupen64++ Beta 0.1.3.12, and this supports netplay and has been empirically proven to reduce desyncs. However, Mupen requires a fair amount more computing power than pj64, and is a little more advanced, so it hasn't quite garnered the attention it deserves. Everybody would have to switch over, and it would be difficult to encourage, people would rebel, be pissy, etc. and it's generally not worth the hassle. However, if you and your friend are looking for a smoother ride and have relatively powerful (by 2005 standards) PCs , try using Mupen.

Phew... that was a lot, but I hope anybody who sees this will realize what's really going on and not let the wool be pulled over their eyes!


TL;DR
Despite what you or others may have you believe, any version of pj64k you find is the same. The only differences are minor convenience fixes, but nothing that will alter your online experience. We all recommend pj64k or pj64kVE found in the index, as they are the most reliable and supported by our ****ty tech support.
 

KnitePhox

Smash Lord
Joined
Oct 17, 2005
Messages
1,838
Location
Chicago, IL
If people play on kaillera with no cheats beyond unlocking everything, they are dumb for not using mupen. no wind/nadoes/music/always dreamland/char select screen animation changes with codes on pj64kve are awesome, can't do this neat and purposeful stuff on mupen...thats why i think its nearly complete ****.
 

Krynxe

I can't pronounce it either
Joined
Aug 1, 2011
Messages
4,903
Location
Lakewood, WA
3DS FC
4511-0472-1729
^ I read this post like 3 times and still don't understand whether or not it supports Mupen.
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
If people play on kaillera with no cheats beyond unlocking everything, they are dumb for not using mupen. no wind/nadoes/music/always dreamland/char select screen animation changes with codes on pj64kve are awesome, can't do this neat and purposeful stuff on mupen...thats why i think its nearly complete ****.
Most people want to play the game it was meant to be played, and Mupen accomplishes just that, as you said. I see no problems here.

It's a minor inconvenience, but I don't think it deserves to be called "nearly complete ****" just because it requires a more roundabout method to implement the more obscure cheats that are easily done in pj64.

The emulator was not designed for cheating, it was designed for, well, emulating, and it does that better than pj64.
 

KnitePhox

Smash Lord
Joined
Oct 17, 2005
Messages
1,838
Location
Chicago, IL
if there was simple cheat input on mupen similar to, if not better than, pj64k on mupen for online play, i assume most if not all typical users of pj64k would switch; i would. i would also push for such a switch very much so in server chat due to less desyncs.

only thing i'm not really sure on is input and gfx plugins and if mupen compatible ones differ from pj64k ones enough to matter to the general cheatless user of pj64k.

till cheats are implemented, or an unknown significant upgrade in either of these emulators for online play, pj64k 0.13 core 1.4 is king.
 

malva00

Smash Master
Joined
Mar 1, 2002
Messages
3,864
Location
54th and 5th
i would play on mupen if using p2p custom frames didn't make mupen run slower than pj64k (this is what i remember)
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
I do agree, cheats in Mupen are a pain in the ***, but through some mystical power, all of the basic cheats are unlocked by default (something even regular pj64k users have issues with), which would actually make it more intuitive and easier to setup than pj64k. Plus, who uses all those crazy cheats you mentioned anyway, especially when all they're known to do is double the desync rate? People want as few desyncs as possible, and to that end, even avoid using cheats like "auto 5-stock mode" and "items off".

Hell, even firo pushes for Mupen whenever you login to the server or start a game, but I doubt he'll enforce any sort of ban on pj64k because all people do is wine and complain and it's not really that necessary. Not on the same degree as something like switching to LAN.

All graphics plugins are compatible with Mupen (more, actually!), but I believe you have to copy over the pj64 plugins because it's illegal to bundle and distribute them... not exactly difficult, but remember the Project64.RDB file!

Oh wow, it looks like there's been an update to Mupen64K. First since 2007. Anybody wanna guess who's behind this? Yep, that's right: smasherx. Has he talked to any of the authors for these emulators to see if he's allowed to modify and redistribute their work?! Holy ****! I'd stick with 0.7.9. Remember what I said about it being illegal to distribute the Jabo plugins separately? Somebody should really bust his ***.

OR

Use Mupen64++! It was also developed until 2007 until okaygo lost the source code.... so unfortunately this project is finished, but the latest version is stable and works quite nicely (I've had fewer crash problems with this compared to Mupen64K).

It's also worth noting that the fixes in the two pj64k offshoots discussed above are present and perfected in Mupen. There's really no reason people should not be using Mupen for P2P or even server play, unless you absolutely need wonky cheats.

Low system specs are understandable, though, and if Mupen doesn't run well on your computer, it doesn't run and that's that.

Just remember.... far fewer desyncs!


i would play on mupen if using p2p custom frames didn't make mupen run slower than pj64k (this is what i remember)
You sure this isn't just because of a low-powered computer? I really don't know much about this and can't test it since I'm at school most of the time. Also, have you tried Mupen64++ compared to Mupen64K? I'm not entirely sure of the differences myself since okaygo worked on both projects, but ++ seems to be the best choice for netplay.
 

KnitePhox

Smash Lord
Joined
Oct 17, 2005
Messages
1,838
Location
Chicago, IL
Galaxy desyncs pj64k every 18.5minutes after a game has started, regardless of who hosted or how many players. Dunno if it ds's on mupen haven't tested

:phone:
 

King Omega

Smash Journeyman
Joined
Apr 15, 2009
Messages
388
I like 64X+ the best. Unlike Sempiternity's post would lead you to believe, this is an actually different emulator from 0.13 Core 1.4. Proof? Any other emu (except maybe mupen or whatever, idk) lags when dragging the cursor onto the character select for the first time, but X+ doesn't.

It's renamed internally so it can play with other versions, but it desynchs more often because it is different. If everybody started using it combined with Glide64 WX Napalm Extreme Ultimate Sugarcoated Mega Superfluously Named, that would make me happy.

Apparently you all have Pentium II processors and can't run Glide without lag, though, so lol.

Galaxy desyncs pj64k every 18.5minutes after a game has started, regardless of who hosted or how many players. Dunno if it ds's on mupen haven't tested
Erm... on a scale of 1 to absolutely factual, how true is this? Because I recall matches longer than 18 and a half minutes, I think, but if you've got real knowledge about this then I must just be remembering wrong.

Let's all migrate to a dead server. There are like a thousand of them and they're all filled with hilarious bugs.
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
Galaxy desyncs pj64k every 18.5minutes after a game has started, regardless of who hosted or how many players. Dunno if it ds's on mupen haven't tested
c'mon bro, you're smarter than that. this isn't true at all. it depends entirely on your connection and packet transfer, which is different every single game, so desync occurs more or less randomly.

I like 64X+ the best. Unlike Sempiternity's post would lead you to believe, this is an actually different emulator from 0.13 Core 1.4. Proof? Any other emu (except maybe mupen or whatever, idk) lags when dragging the cursor onto the character select for the first time, but X+ doesn't.

It's renamed internally so it can play with other versions, but it desynchs more often because it is different.
um, what is 64X+?

Also, I thought you might be interested to know that the only way to change the "internal name" is to alter the source code, which nobody has except for the now-defunct hotquik team.

Regarding your proof, is that the best you can come up with?
 

King Omega

Smash Journeyman
Joined
Apr 15, 2009
Messages
388
um, what is 64X+?
Do you want a link? I don't believe the creator's site exists anymore, if there ever was one. All I know is it's on my hard drive and was touted as being a "better Project64k" wherever I downloaded it from.

Also, I thought you might be interested to know that the only way to change the "internal name" is to alter the source code, which nobody has except for the now-defunct hotquik team.
In that case, it wasn't renamed to 0.13 Core 1.4, but rather the name was simply not changed from 0.13 Core 1.4 to begin with. That's not too hard to figure out.

Regarding your proof, is that the best you can come up with?
Explain to me how a repackaged version of the same emulator could possibly lag less than the allegedly identical original version, and then call my proof weak.

It lags noticeably less (that is, not at all) in many places where I experience offline lag when using VE.

EDIT: lol, I did my research, and it looks like X and X+ are just another pair of edits of 0.13 by Smasher. From his posts on random forums about them, I can't actually tell what significant changes he made to it, but whatever it was, it worked. Less slowdown (lag is probably not the best word), period.

I doubt there are any good downloads of it available, so I uploaded it for anyone interested.
http://www.mediafire.com/download.php?u6sva2vcqin7ilu
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
Point is, if it wasn't renamed, it wasn't changed.

I have never heard of this "64X+", nor does any documentation appear to exist anywhere on the internet (I assume the real name is something different).

So yes, I'd love a link.
 

King Omega

Smash Journeyman
Joined
Apr 15, 2009
Messages
388
Edited my post. I found out Smasher made it, etc., etc.

I bet you know more than I do about this, but I can tell you one thing for certain. The emulator I linked to above slows down less during normal operation than do 64k 0.13, VE, and 7E.
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
Hm, this is interesting. I haven't seen emulators touted with a different core build. I'm skeptical, though. It seems the "Help" menu was conveniently removed, so I can't tell what is really going on here (not decompiling that ****, either).

This also doesn't appear to be smasher's work. A bit of digging around will reveal that smasher has been stealing the work of people who edit these various emulators. It appears that Gaudy and Henryx are the original authors of the emulator on his website, and from the ReadMe file in the link you posted, some Pixelsmasher guy looks to be the man behind this emulator. Smasher has never used those contacts before, but you'll also notice that little ©Smasherx74 on the very bottom which tells me he just slapped that on there to claim ownership and passed it off as such.

****ing scumbag.

The ReadMe also doesn't appear to be his style; it's too well-written and informative, aside from the starred part at the bottom that calls people "zombies" and has run-ons, which appears to be written by him.

I have no idea where you found it! I found one forum post from 2010 by Smasherx, but there wasn't much of substance there, and nobody responded. Everything else seems to be Xbox emulators.

Beats the hell out of me, but if it truly is based off of the 1.6 core, then maybe it is better. I refuse to believe it, though, because the source code for 1.6 is behind lock and key, which means the guy just fiddled with it in a hex editor (meaning he's unable to actually change anything core to the emulator), or rewrote the entire source code (if he did this, why wouldn't he name it something unique?).
 

SuPeRbOoM

Smash Master
Joined
Oct 27, 2005
Messages
4,509
Location
Edmonton, Alberta
Galaxy desyncs pj64k every 18.5minutes after a game has started, regardless of who hosted or how many players. Dunno if it ds's on mupen haven't tested

:phone:
c'mon bro, you're smarter than that. this isn't true at all. it depends entirely on your connection and packet transfer, which is different every single game, so desync occurs more or less randomly.
Go and LAN and test for yourself. You're only going to get 18.5 minutes of gameplay until you get a "lag out"

These lag outs don't happen on old kaillera servers, so it must be an emulinker problem.
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
hahaha this is true?

Well I've never lasted more than two or three games without something going horribly wrong, so I guess I've never made it to the 18.5 minute mark. That's still pretty weird, though. Any idea what causes this?
 

KnitePhox

Smash Lord
Joined
Oct 17, 2005
Messages
1,838
Location
Chicago, IL
I've used a decent amount of codes to minimal to none and 18.5 mins has been max time. I am smarter than that ROFL, get good enough internet to not lag out before then and you'll see. Wireless won't cut it lol

No clue what causes the ds but seems to be an emulinker problem.

:phone:
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
I've used a decent amount of codes to minimal to none and 18.5 mins has been max time. I am smarter than that ROFL, get good enough internet to not lag out before then and you'll see. Wireless won't cut it lol

No clue what causes the ds but seems to be an emulinker problem.
Sorry, bro, apologies. I'd just never heard of that before. Now that you mention it, I do get the "lag out" sometimes, but since I DS so frequently in between (blame the kaillera scrubs), it doesn't happen often. I have thought it was weird that it would sometimes happen when not ingame. It all makes sense now!

SO

I feel a little stupid now, but after doing some research, I've realized some of the 1.4 source are open to the public (y nobody tell me???) and are modifiable. However, since smasherx just thieves Gaudy/Henryx's work, I looked at the PJ64K7E a bit more, thinking they had modified the code, but it looks they have no C experience, so they haven't fiddled much with the source. They have, however, updated some of the ROM info to current 1.7 standards, tinkered with the NRage plugins to completely remove the RAWDATA option, leaving it permanently unchecked, fixed some plugin loading errors, incorporated pj64 1.7 Jabo plugins, and some DS stuff. Doesn't appear to be game-changing, but it's certainly worth a look if kVE is giving you problems.
 

firo

Smash Ace
Joined
Jul 27, 2008
Messages
600
Location
Champaign, Illinois
So yeah, I highly considered making the push to mupen along with the push to LAN, but there have always been people with computers that don't have the resources to handle mupen, and realistically the only way make the change was to ban PJ64 for a while (like the ban on exc in order to force people to use LAN). So I didn't think this was right. I still have that message "Try playing with mupen!" whenever you log into the server though.

About this 18.5 min lag out, I've noticed it, and it has been around for a while. My question is, for the older dudes, was this problem always there, even on Emularena grounds? I know that not many played on LAN during that time (hell, good was used only up until the last year or two of the server), but it is something I would like to know. I compared the net code for the version running galaxy and the emulinker 1.0.2 source and they seem identical. There's no obvious point in the code where the bug would occur. If there was, it probably would have been fixed in the initial Emulinker releases, so the bug might be pretty involved. But I'm not 100% sure. Does it happen with all the different kaillera clients?

What would help in at least diagnosing the problem is if I can get the contents of the packets being sent and received right at the desync point. I don't know much about packet inspection, but I need to do some testing and look into it.
 

King Omega

Smash Journeyman
Joined
Apr 15, 2009
Messages
388
Hm, this is interesting. I haven't seen emulators touted with a different core build. I'm skeptical, though. It seems the "Help" menu was conveniently removed, so I can't tell what is really going on here (not decompiling that ****, either)

This also doesn't appear to be smasher's work. [...] which appears to be written by him.
Okay, that explains a lot and also makes total sense. I was feeling kind of iffy about reporting back that Smasher made it, since I felt like that flushed my credibility down the drain... Glad to know he didn't.
Beats the hell out of me, but if it truly is based off of the 1.6 core, then maybe it is better. I refuse to believe it, though, because the source code for 1.6 is behind lock and key, which means the guy just fiddled with it in a hex editor (meaning he's unable to actually change anything core to the emulator), or rewrote the entire source code (if he did this, why wouldn't he name it something unique?).
I really have no idea. I wish I could remember how I came by it—I know it wasn't from Smasher. I think the guy actually had a site (a subdomain, anyway), but it's gone if it ever existed.

It's interesting, though. In the readme, it calls itself a "hack" and talks about some plugin changes, but it neglects to mention anything else. Like, nothing in that readme seems to support my claims that it's faster or more efficient. I suppose "hacking" the way it handles plugins might provide a speed-up in certain areas?

Also, this one being "X+" instead of just "X" (this comes form downloading it originally, since I'm not sure the + is ever even mentioned in the files) implies it's changed even further than whatever hex edit/other thing was done to create "X."
****ing scumbag.
Couldn't agree more.

About the 18.5 minute desynch thing, that's pretty awful. Of course, most people I play desynch waaaay before 18.5 minutes, so I don't have to worry about that. That is, if stability issues are fixed to the point where 18.5-minute matches are reasonably possible, THEN it should become a concern.
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
It's interesting, though. In the readme, it calls itself a "hack" and talks about some plugin changes, but it neglects to mention anything else. Like, nothing in that readme seems to support my claims that it's faster or more efficient. I suppose "hacking" the way it handles plugins might provide a speed-up in certain areas?

Also, this one being "X+" instead of just "X" (this comes form downloading it originally, since I'm not sure the + is ever even mentioned in the files) implies it's changed even further than whatever hex edit/other thing was done to create "X."
The only "hack" part I can think of is that the guy somehow edited the 1.4 source for the interpreter/compiler/application to make it more like 1.6, but nobody knows what 1.6 is actually like so it'd be largely guesswork or somebody with a lot of coding and emulation experience fixing stuff. Even then, you have no way of knowing how similar it is to 1.6, so why make the bold claim it's based off the 1.6 core? That, combined with the lack of documentation makes me very suspicious.

I also came across a post of smasherx asking for the 1.6 source code, and one of the project leaders told him there are no plans for releasing it, so I know smasherx doesn't have it.

I have no idea, and there's not enough info to make a sound judgment.
 

SuPeRbOoM

Smash Master
Joined
Oct 27, 2005
Messages
4,509
Location
Edmonton, Alberta
About this 18.5 min lag out, I've noticed it, and it has been around for a while. My question is, for the older dudes, was this problem always there, even on Emularena grounds? I know that not many played on LAN during that time (hell, good was used only up until the last year or two of the server), but it is something I would like to know. I compared the net code for the version running galaxy and the emulinker 1.0.2 source and they seem identical. There's no obvious point in the code where the bug would occur. If there was, it probably would have been fixed in the initial Emulinker releases, so the bug might be pretty involved. But I'm not 100% sure. Does it happen with all the different kaillera clients?
The 18.5 minute lag out timer only occurs on LAN connection. If you switch to exc, the time will be increased to around 35-40(not sure exactly) minutes. So lag outs on egx were never really noticed before because they were so infrequent to what's happening now.
 

The Star King

Smash Hero
Joined
Nov 6, 2007
Messages
9,681
^I noticed that double 18.5 minutes, 37 minutes, falls within "35-40 minutes", and LAN sends double the packets per second that Exc does. So I guess it has to do with that? Maybe this means Good connection lags out every 55.5 minutes?

If this is true it would suggest that after a certain number of packets sent it always lags out. If 60 packets are sent per second on LAN, than 18.5 minutes = 66600 packets sent (oh em jee Satan is causing us to lag out!). Just over 2^16, so maybe after it exceeds the 16-bit limit it lags out?

This is all speculation, sorry if I'm dumb lol
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
^Goddamn you're observant!

Somebody should test this and record the exact time it takes to lag out on each of the different connections.

(not it)
 

The Star King

Smash Hero
Joined
Nov 6, 2007
Messages
9,681
If everything I just said isn't complete bull**** then the exact times would be

LAN: 18 minutes and 12.2666... seconds
Exc: 36 minutes and 24.5333... seconds
Good: 54 minutes and 36.8 seconds
Average: 1 hour, 12 minutes, and 49.0666... seconds
Low: 1 hour, 31 minutes, and 1.333... seconds
Bad: 1 hour, 49 minutes, and 13.6 seconds

Assuming that the number of packets sent per second is exactly what the Connection type scroll bar says it is.

P.S. There should really be an easy way to type vinculums for repeating decimals. They're so much prettier.
 

clubbadubba

Smash Master
Joined
Apr 27, 2011
Messages
4,086
damn i was about to say exactly what star king said. never noticed this 18.5 min lagout either, weird.
 

blaze3927

Smash Ace
Joined
Aug 12, 2009
Messages
832
Location
Melbourne
by desynch, does you mean it just desyncs in game or you get the warning and the game drops?

im running emulinker SF 0.72 on the aus server I havent noticed any exact timings for DS
 

King Omega

Smash Journeyman
Joined
Apr 15, 2009
Messages
388
I still don't see why would we should care. I know it would good to fix it, as if something had been accomplished, but considering most lagless, two-player, well-connected matches desynch well before 18.5 minutes, it really doesn't matter.

I have 0% jitter and a great all-around connection, and I still desynch very frequently for no reason. Of course, oftentimes I'll have series of matches where I don't desynch in hours, but then somebody else will desynch, so lol.

Then again, I've never really trusted the DS message when it comes to accuracy. When I'm playing somebody with dozens of lag spikes and then it tells me I desynched, I get a little suspicious.

...Isn't it bull**** to say any particular player desynched in a two-player game? You only have two participants, so either they're synched or they aren't. If one is desynched, so's the other, right? In that case, it's always the fault of the laggier party, I'd imagine.
 

Fish641

Smash Apprentice
Joined
Dec 16, 2010
Messages
102
I seriously doubt people sit and complain that Player X dsynched. No one's putting fault on someone...
 

KnitePhox

Smash Lord
Joined
Oct 17, 2005
Messages
1,838
Location
Chicago, IL
for the 18.5minute thing the game just freezes and server posts that the host DS'd. your connection obviously isn't good if you FREQUENTLY desync BEFORE 18.5 minutes, lol, good one.
 

King Omega

Smash Journeyman
Joined
Apr 15, 2009
Messages
388
I seriously doubt people sit and complain that Player X dsynched. No one's putting fault on someone...
I'm not talking about X. My point is that the server displays messages like "Desynch detected! Player51487649, you are lagged from the other people in the game!" but I don't think that can possibly be accurate in a two-player game. Whichever player desynchs, the server should only be able to tell that they're desynched, not which one initiated the desynch.

But I often get desynch messages against very laggy players where it tells me I've desynched, and I get suspicious. It should always be the laggier player's fault, because their game can't keep up.

The server does not have its own game of Smash 64 running somewhere that it measures all the other players against. It only measures the players' synch against each other.
 

Sempiternity

Smash Lord
Joined
Sep 11, 2010
Messages
1,695
Location
Connecticut
Thanks, bro, but it doesn't support kaillera so it's not of much use here.

Plus you need to use it through the command line which turns a lot of people off. I don't understand; they put so much effort into creating the best emulator around, but don't bother to code in a UI? That alone is enough to make me not want to use their product.
 
Top Bottom