Petition to get Soul Calibur V on GGPO

This petition is to ensure that the online experience is as best as it can be for us, but this won't happen without your help! You need to sign the petition and tell everyone you know to sign it as well. With your help, this will happen! We need to show them that the players demand this!

SIGN HERE​

In recent years, GGPO has become the gold standard for online netcode in video games. If you do not know anything about the GGPO netcode this is the description, copied and pasted from their website, http://GGPO.net :

What is it?

GGPO is a networking library that game developers can use to add networked gameplay support to arcade style games. GGPO's latency hiding techniques give each player a gameplay experience that is nearly indistinguishable from playing with their friends locally, even against players around the world.

How does it work?

GGPO uses a peer-to-peer topology to run a complete copy of your game for each player, transmitting controller inputs over the network to keep these copies in sync. Each player's inputs are sent to their copy of the game without having to wait for their opponent's to arrive over the network.
If the simulations diverge, GGPO rolls back to the most recent accurate state, corrects the mistake, and jumps back to the current frame, all before the player can notice. This provides the illusion of lag-free gameplay.

How does it perform?

Thousands of gamers are already using GGPO to play classic, old school fighting games against opponents around the world. GGPO's latency hiding techniques maintain the illusion of lag-free gameplay even during intercontinental play.

Can I try it out?

Absolutely! Click the "Try It!" button to get started.

http://www.facebook.com/?ref=home#!/pages/GGPO-for-SoulCaliburV/226542944023734

 

Supported. GGPO can work with 3D fighters and isn't any more taxing that any other netcode. It uses a unique prediction model to eliminate things that cannot happen to streamline the code.

 

That's wonderful news, WCM. I trust that you know what you're talking about. Hopefully this will put all the naysayers at ease. I would post the same thing over at SRK, however I am not able to log into the front page for some reason.

 

Supported.
P.S. Thanks WCMaxi for what you did with Caliburforums a few years ago. If I ever meet you I'm buying you a beer.

 

Tekken netcode is acceptable at 4 bars and good at 5, SC4-like at 3, unplayable at 1-2.

That said, I'd be fine with Tekken's netcode, if the matchmaking is better. VF5 netcode > GGPO for 3D fighters, hoping netcode is at that standard.

GGPO isn't going to happen, systems aren't powerful enough.

 

tekken is even worse than sc4. you guys are crazy.

 

Yes, let's please stay on topic here. This isn't a discussion of which netcode we want. We obviously want the best netcode for our fighting game, and right now that is unquestionably GGPO. If you don't know this in your own experience, please take our word for it.

 

Supported. GGPO can work with 3D fighters and isn't any more taxing that any other netcode. It uses a unique prediction model to eliminate things that cannot happen to streamline the code.

Welcome back!

 

Supported. GGPO can work with 3D fighters and isn't any more taxing that any other netcode. It uses a unique prediction model to eliminate things that cannot happen to streamline the code.

can you, perhaps, briefly explain what was used in SC4 and also VF5, and why/how GGPO could potentially greatly reduce lag from the netcode of SC4 and emulate VF5 online play? i ask because much of what i hear is that VF5 had equal if not superior netcode to GGPO. we all know that SC4 netcode was terrible, but we don't really know why, or what GGPO will do to make it different.

this question is not directed specifically at you, but to anyone who feels they may have answers. i think it is best if we can build the strongest case possible for why this is, without a doubt, something that will benefit SC5. this thread should be a place to not only hype the idea of SC5+GGPO, but to consider the alternatives and make it clear to supporters exactly why they are signing off on this.

to anyone who can answer these difficult questions for me; *pre-emptive high five* you rock.

 

lobo:
Without violating multiple NDAs I cannot. I will say:
1. 3D and 2D both transmit the same rate and volume, being 3D doesn't some how make the volume more, especially since you're mostly sending position + input related data. Visuals and other bullshit is all handled on the client side.
2. GGPO is known to work in a prediction model; thus, if from B only B is possible, then the system would know to expect B or nothing (as a rough example). If a prediction model is used, the game can proceed smoothly even when lagging until the expectation and prediction mis-match, in which case GGPO rolls backward to a point in which they matched. This is mostly packet loss issues.
3. GGPO is used in our arcade DBKai game, which is 3D and also 4 players (thus, double the volume of 1v1).

I don't know anything about VF5's netcode, can't comment. I know I've played with GGPO working flawless out to 200+ ping. I'd be shocked if the VF code does that. If it did, I'd think you'd have heard more about it or in the least Sega would have put it in their arcade version (since VF5 is dead here).

 

good luck with your petition

 

SC4 works out to 8kb per second of data, I measured it once. Looks like my thread on this was deleted or lost or something. The problem is not bandwidth (would fit easily in a dial-up modem), its purely one of distance and latency. No matter how good your code is, you cannot overcome the latency caused by physical distance and all of the network nodes between you and your opponent.

The thing is, i even played someone who lives 15!! minutes aways from me, and the connection was still shit!! No matter if it was a 5 bar connection, SCIV Online is one of the biggest online MP jokes i´ve ever faced in games to be honest.

 

The resources being taxed when people talk about ggpo and 3d fighters are not online resources, they are cpu resources. Simply put, current gen systems may not be powerful enough to run ggpo alongside a game like Soulcalibur 5. If that were true then things would have to be changed for online play (this already happens in some console games, mostly fps), lower res textures, less animations, less effects. At the very least. It has nothing to do with network resources.

Even ggpo users will have encountered this, specifically with 3rd Strike. If your computer can run 3rd Strike at full fps, at whatever res, offline, no stutters, no slowdown at all, it still may not be powerful enough to run 3rd Strike over ggpo, and as your frame rate drops you will start to chop and lag both yourself and the person you're playing against. In order to play 3rd Strike over ggpo you need a system far and above what is needed to play 3rd Strike offline.

Um, did anybody even read this post? It has alot of good information; Page knows what he's talking about, so listen to him!!

While everybody's on the ggpo hype train, does anybody even know how GGPO works, at least on a broad scale? From what I understand (if I'm right), it works on roll-back technology, meaning, when there is inconsistent connection, the state of the game loop on the systems of both players (kinda like Prince of Persia) rolls back to a slightly earlier time state to regain consistency. In other words, (if I'm right) the game literally goes back in time for a few unnoticable frames when connection falters.

What does that mean (if I'm right)?

That means that your xbox/ps3/whatever has to store(buffer) in memory and redraw several frames worth of graphical data/lighting effects/physics calculations/etc; something that the system themselves, as well as the games created for it, were NOT designed to do. With 2d and older games, this is not much of a problem, but with the latest 3d games, which are designed to push the system hardware to feesible limits in order for you to drool at its pretty lighting effects, there is simply too much graphical/mathematical/other data for ggpo to be properly implemented. Not that implementing ggpo is a 'bad' idea per se, but it's simply not possible without designing the game to do so from before even starting development (and rationalizing what graphical/lighting/physics/special effects to sacrifice to do so). Even then, I sincerely doubt it can be implemented as well as these ROM games on the net (again, might be wrong though).

What about games that use it now? Take a look at how graphically intense the Final Fight remake is (read:not very). Even the new DBZ arcade game in Japan that uses ggpo is graphically sparse compared to other modern (and more traditional) fighters.

Harada has explained this about Tekken (on the avoidingthepuddle podcast with Daishi-san), Ono and S.killian explained this about SF and Marvel; though none of these people went into real detail as to why GGPO is not feasible.

tl;dr? Stop blind support and DO SOME RESEARCH. But hey, at least blind GGPO support is actually a good thing!!

 

I'm all for this. Utah doesn't offer many offline gamers, so online is the only chance I have to play against actually good characters.. I know a lot of you guys want offline focus, and I understand that, but you're not the only ones that will be playing SC5..

 

Black:
1. I don't think you understand how a system works. There's no need to store graphical effects because an engine will render it based on what's happening in the game world.

2. Watching a low rez version of the DBZ game and saying it isn't visually intense is odd. Also conveniently ignores that game is 4P, the insane number of projectiles all being tracked, state of the destructible environment, etc.

3. All good netcode is going to use some measure of resources to work. What this petition is saying is close to that fact that GOOD NETCODE IS IMPORANT. More than which good netcode to use, make sure it is good. This isn't my petition but I'm certainly in favor of it. Good code of any kind requires resources, more people will play SC5 online than off, that's the reality of the situation. If those online people are ever to migrate offline, you need a means for them to. That begins with very robust netcode - no matter who makes it.

4. Harada never said GGPO couldn't work with Tekken. Harada is also not a netcode programmer.

5. Ono's SF4 and MvC3 use a very similar process to GGPO. It is no coincidence that SSF2 HD was out first using GGPO. Ono is also not a netcode programmer.

I know recent trending of game producers as semi-celebrities might make them attractive to quote (especially when nobody is going to re-listen to podcasts to verify your claims), a producer isn't a programmer or an expert on much of anything (when you get down to it), they are the person who makes everything happen in a game's development cycle. Generally they know their product inside-and-out; but something like the finer details of netcode... it would be an odd fetish for any producer to be an expert on this topic.

There's no reason to be against it given what people say about SC4's netcode... aside from just raw trolling. In which case, you got me to write an essay for a troll attempt; good job.

 

i wouldve supported...if i had a facebook account.

hi WCMaxi!

 

What this petition is saying is close to that fact that GOOD NETCODE IS IMPORANT.

This is the reason I signed the petition.

 

I signed because I just want them to understand we want something non-terrible.

 

tekken is even worse than sc4. you guys are crazy.

Dark Resurrection, sure (WORST. EVER.). Tekken 6 was smoother for me than it was in Soul Calibur IV.

 

snip

Good address. The only reason I even mentioned what I did in my post is because I get the feeling, as I read this thread, that GGPO is the savior/end-all-be-all of good network code. I'm simply saying that it is not, and giving reasons why; all simply for the sake of the information and nothing else. I'm definitely in support of good netcode, as it's the main way I get play other people.

1. graphical effects at different times aren't produced out of thin air -- the presence, intensity, scope of a light source, how it effects objects in the game world, the state of the 'camera' in which the gamer views the game world, which subsequently decides what polygons get rendered, textures used, etc -- all of these elements and much much more NOT ONLY need to be stored, but also mathematically calculated, processed and rendered on screen to be displayed, etc. Graphical effects, those that are AND are not on screen need alot of storage and cpu resources and the use of the (unmodified) GGPO technique is not possible without restructuring (and adding restrictions) the graphics engine used

2. i don't think there's a 'low res' or 'high res' version of the dbz game. that's how it looks, and if you compare it to SF4, SC4, VF5, or even tekken, and dbz looks less impressive. 4 player gameplay doesn't help matters much, but every thing you mentioned adds to what i said before

3. agreed

4. this doesn't even make sense. you seriously think that Harada doesn't talk to his programmers? and like I said, if you want to hear the statement for yourself, check out the avoidingthepuddle podcast with harada and daishi

5. SF4 and Marvel DON'T use GGPO. that's all I said. I just re-stated information specifically about GGPO stated by them. Ono also has network coders that I'm preeeeeeetty sure he talks to once and awhile.

I just don't like when people sign up for things without knowledge ANY of what they are or what they can do. I don't like bandwagon mentality at all.

I understand the concerns about netcode, because I share them. But if you think GGPO will magically solve all netcode issues, then you are simply mistaken. GGPO itself is not your answer. Daishi mentioned on his twitter and in other forms of media about how he understands the importance of netcode. Because of all this, I don't see the point of the petition. I see the concern for it, but theres no real point.

i'm not trying to troll, i'll stop posting if it seems like that

 

I think NB tried this out on some their other less profitable titles first to learn the technology without risking big money. By the time SC5 comes out, I think they will be ready for this.