online slower then vs.

Your a dumb fuck for basicly telling me 50hz online for a 60hz game is legit. if they couldn't make online decent they shouldn't have released the game untill it was.

Who is trying to say that online play = offline play? You're basically getting angry that it isn't perfect, as if this was some new amazing discovery.

It's not perfect. But it's a lot closer now than it's ever been, especially when compared to SC4.

I really don't understand why you are attacking this subject like the online vs offline battle has never been fought before. Did some online warrior stomp you and parade around claiming they were the best soul calibur player ever or something?

Basically everyone is at peace with the fact that online is online and offline is offline.

They both have their uses, and good players can switch between the two quite easily.
 
no ones talking about the game, i'm talking about the online mode, which is almost never laggy and always responsive.
I'm talking about VF5 online, not the game. It is not perfect. You lose ch options on block that are in your favor offline and om P+K is buffed etc.

(Though if you'd ever played VF5 offline you'd already know this.)
 
Who is trying to say that online play = offline play? You're basically getting angry that it isn't perfect, as if this was some new amazing discovery.

It's not perfect. But it's a lot closer now than it's ever been, especially when compared to SC4.

I really don't understand why you are attacking this subject like the online vs offline battle has never been fought before. Did some online warrior stomp you and parade around claiming they were the best soul calibur player ever or something?

Basically everyone is at peace with the fact that online is online and offline is offline.

They both have their uses, and good players can switch between the two quite easily.

because clown shoes came in the thread trying to say stuff like "buy better internet" and "thats lagz lol" when clearly it's not, if people just said yeah the games been slowed down by 5% and it makes online run better then i wouldn't argue with them, infact that happened and i left that guy cos he obviously knows what he's talking about.

and theres no way any online warrior could do shit to me, you can pretty much see every low and have tons of time to react to everything.
 
The game is indeed slowed down depending on the connection, if you have a perfectly synced connection with your opponent it seems nearly like offline though. Online I can react to ɑPat's 1A:A and block it. Offline I can't. Weird oddities about online. Nothing is constant.
 
because clown shoes came in the thread trying to say stuff like "buy better internet" and "thats lagz lol" when clearly it's not, if people just said yeah the games been slowed down by 5% and it makes online run better then i wouldn't argue with them, infact that happened and i left that guy cos he obviously knows what he's talking about.

Um, that's also what I said.....

Except.... the game isn't slowed down by 5% consistently. In fact, it isn't fixed or even linear. It's variable in order to improve the online play. This is not new. This is present in nearly all online games, especially more modern ones. Go check out the options in Battlefield 3 that allow you to dynamically change whether you would prefer to have more visual anomalies (frame skips and duplicates) or less.

You will see similar options in GGPO titles that allow you to adjust the maximum frameskip.

inb4 you call me names for having basic, fundamental understandings.
 

ppl's taking it little to serious.

First vid is from good connection with my friend, second is from some random game, not that good connection.
average framerate for first is 57,5 frames/s, for second vid 56,6 f/s.
So it's possible that game runs 2-4 frames slower depending on connection (maybe max 5 frames slower, but not with my connection).

For me it's pretty good, I like that online mostly, better this then nothing, or some shit from sc4.
 
thx for vids, pretty much shows excatly what i was trying to describe.
It looked to be almost 4 seconds by the end of the 60 second match, which is quite alot if you think about it.
 
I did another recording test which basically had the same result.

It's worth noting that on a good connection, the duplicated frames (game freezing for one frame) is relatively rare. For me, it happened once maybe every 2nd/3rd attack. Anyway, I'm very certain this is the result of lag, so the game isn't intentionally slowed down. It's also very minor considering random animations becomes only 16.7ms longer.

Input lag has a much bigger effect on the general feel of online play. Unless you're playing on a LAN, you can assume input is always delayed by at least several frames.
 
News at 11:

Man proves apples fall from trees. Further experiment to prove that all things that go up must come down.
 
Try going into training mode and turning on the delay lag option and recording the same set of moves with it on and with it off. Say record a standard BB twice in a row (so BBBB) and find the time that it takes for the 3rd B to hit after the 2nd B (since you're mashing, it'll come out as fast as possible, and the *mid* hit spark thing on each hit will provide a common starting/ending point between the 2. This training mode option definitely feels a lot more sluggish, it totally throws off my just frame input timings.

You could do the same thing with online/offline and time when the health bar starts to change for the 2nd and 3rd hits.
 
I'm pretty sure that toggle only adds input delay (you can test this by tapping guard and then seeing when the character starts the guard animation), so that will make moves come out slower but for a different reason.
 
Rude person is rude.

I can't help it. I have no idea how much harder I need to explain this concept that everyone seems so fascinated with. I've even made a really simple analogy.

People can continue to record the same matches and run them frame by frame and you'll discover the same thing I've already said - frame skip and frame dupe are normal in online games. Whether they be intentional to prevent visual artifacts or a result of the network implementation.... it isn't new to SCV. :\ Technically, this isn't even specific to online games....
 
Well, I was giving the guy the benefit of the doubt, I wouldn't have been shocked to discover it had been intentionally slowed down. It's interesting to know specifically how games deal with latency anyway, some games use very different techniques. Ie, shooters normally go for a host/client system where the host handles everything with the clients only predicting actions while waiting for the real updates from the host. While some games uses a mix of host handling stuff and the clients doing some stuff on their own (Capcom likes use that system for their shooters).

I'm honestly not sure how SC5 handles everything in detail. I've seen no cases of rollback or characters changing animations when they shouldn't, so the basics must still be a classic peer-to-peer netcode where each client waits for the opponent's input before updating the game. I find it puzzling that despite using that as base, the input lag is still drastically smaller than SC4. Either SC5 does something magical, or the developers did some serious mistakes with SC4.
 
The logic flaw in the game being slowed down by a fixed rate is that.... if it can't run at rate X due to delivery failures, latency, etc, then why would it run at any other speed any more reliably? Even one additional second flatly applied over the 60 seconds of a match wouldn't generally be enough to cause significant improvements. If it did, why wouldn't they just make the game run at that speed offline, too? The OPs logic was flawed with his "SC stretched time" illustration. The offset tick illustration was even more incorrect.

Most modern games by design don't flatly use peer-to-peer or peer-to-server. Instead, they use a combination with the server generally acting as a governor/sentinel/whatever you want to call it similar to the shooter method you mentioned. I would be pretty confident that anything on XBL uses something more similar to that same method with input frames (just like locally) captured, relayed to the server and then sent to the peers. No idea on PS3 since they don't have the restrictions that XBL has. Regardless, either configuration is affected by latency and drops.

SCIV online was a monstrosity and I'm sure they've improved the code itself, but you should also remember that it was 4 years ago. FttH, FttC and other high speed providers are more common now while I think people were still using DSL over antiquated phone lines and other crap in 2008. Not to mention the server infrastructure has surely been upgraded since. Transport upgrades are probably half of the "magic".

I've seen characters stand up in the middle of guaranteed combos, I've stepped horizontals, not-usually-whiffy verticals go right through people and I've seen frame duplication on my side while my friend is still running at full speed. Plenty of people report combos or JFs that require no visual cue (ie Yoshimitsu's ear slicer is always the same timing while Poe's stab has slightly different timing depending on distance, character hitbox, etc) turn into completely different inputs than usual (Yoshimitsu example a:B+K becoming A+B,K or A+K or the infamous 236236A+B+K inputs turning into GrimSlide/NS). It's not often, but it's enough to lose a match.....

This is still 10,000X better than SCIV where crouching was flippin' impossible and AA gave guaranteed throws. It certainly beats KoF12 and SCV doesn't cause nearly as much of a temper tantrum when my combos fizzle like KoF13. I don't even want to mention T6......
 
Back
Top Bottom