A key indicator of how far wrong you've gone in your thinking "rubberbanding is good.". Wow, the level of disconnection from reality in that statement is profound. You have so focused your thinking around being set on doing things one way, that it has utterly destroyed your ability to think rationally about the issues surrounding it.
Rubberbanding is way better than swinging on targets that aren't there or getting punched from across the room. In a perfect world we'd have neither, but if I'm playing in reality I know which one I'll pick. I remember days of playing FPSs on terrible internet connections and seeing myself and others undergo tiny rubber-bands constantly. Troublesome yes, but playable.
Indeed, it is a little jarring, but I'd much rather be jarred back into proper positions than continue to not know where enemies really are.
Invited to Beta 2012-03-18 / Supporter since 2012-04-08
as much as i love it some things are just too plain simple to get i can come up with a faceroll cookie cutter totem/spark build within 5 minutes that will have the core of its build finished at a 60 point mark however whenever i try sooo hard to come up with a melee build
that
1
does not rely on RS
2
does not rely on static shock for dps
3
does not need insane amount of hp
4
all the general nodes for melee are far to general i find 2 handed melee nodes to be exactly the same as nodes for swords a simple damage increase or acuracy increase.
5
i dont mind having general damage/acuracy/atk speed increases for 1 handed or 2 handed but the specialization for specific weapons themselves are far to lack luster with the only difference is that
maces can somewhat be a stun build
staffs can combine a 2 handed while retaining a block rate (wich is severely underpowered if you ask me)
and swords having 1 node that allows to bypass block.
i feel one should really build on this if i want to dump all my points in staff specialization that i really can have a very well balanced source of dps/defence
so my question is this. will this change? can we really expect to become sword/mace/axe/staff masters and be viable end game?
can we expect that the weapon choice will change the way we play can we expect keystones dedicated to specific weapon choices?
This proposition was made many times b4, but i say it again:
Why not to implement the roll-system? Like it was in World of Warcraft Calssic (no idea if the system changed meanwhile). In my eyes, it is the best solution.
Here a description of this system for people who dont know it:
1. A unique drops.
2. On all party members screens spawns a little time counter with three otions: Need, Greed, Pass. Players have 30 sec to make a choice.
3. After all players have choosen the desired option the server calculates per RNG a number for all choosers in the "highest" category (Need > Greed > Pass). All numbers will be displayed and the the highest number wins.
4. In WoW the item will be moved automatically to the winners inventory, in PoE the item may be locked for the winner for, lets say, 1 minute.
While I abhor FFA loot, I believe (and have said before) that this kind of system is the next worse thing. Here's why:
1) Screen clutter. I'm trying to play a game here, and I keep getting all these boxes I have to click to get rid of. It blocks the interface (can't click through it), and interrupts gameplay because I have to look at the item and choose an option, then get back to the game. Gets progressively worse as IIR and number of players increase. Options for rarity thresholds to roll can help, but the problem never goes away.
2) Does little to nothing to alleviate loot drama. Many people will roll on everything they can, and who can blame them? If nothing else they need orbs too, but there will still be crying from anyone who would actually use the item.
3) The only reason it is actually better than FFA is that it removes reaction time and proximity to the drop from the equation.
That is speaking to PUGs of course, that's what loot systems are designed for. An organized group can bypass whatever system is in place and allocate loot however they like.
This proposition was made many times b4, but i say it again:
Why not to implement the roll-system? Like it was in World of Warcraft Calssic (no idea if the system changed meanwhile). In my eyes, it is the best solution.
Here a description of this system for people who dont know it:
1. A unique drops.
2. On all party members screens spawns a little time counter with three otions: Need, Greed, Pass. Players have 30 sec to make a choice.
3. After all players have choosen the desired option the server calculates per RNG a number for all choosers in the "highest" category (Need > Greed > Pass). All numbers will be displayed and the the highest number wins.
4. In WoW the item will be moved automatically to the winners inventory, in PoE the item may be locked for the winner for, lets say, 1 minute.
While I abhor FFA loot, I believe (and have said before) that this kind of system is the next worse thing. Here's why:
1) Screen clutter. I'm trying to play a game here, and I keep getting all these boxes I have to click to get rid of. It blocks the interface (can't click through it), and interrupts gameplay because I have to look at the item and choose an option, then get back to the game. Gets progressively worse as IIR and number of players increase. Options for rarity thresholds to roll can help, but the problem never goes away.
2) Does little to nothing to alleviate loot drama. Many people will roll on everything they can, and who can blame them? If nothing else they need orbs too, but there will still be crying from anyone who would actually use the item.
3) The only reason it is actually better than FFA is that it removes reaction time and proximity to the drop from the equation.
That is speaking to PUGs of course, that's what loot systems are designed for. An organized group can bypass whatever system is in place and allocate loot however they like.
Yeah, seriously. His notion is, I think, terrible. 30 seconds for players to click on a buch of shit to indicate if they want something?
No thanks.
Invited to Beta 2012-03-18 / Supporter since 2012-04-08
This proposition was made many times b4, but i say it again:
Why not to implement the roll-system? Like it was in World of Warcraft Calssic (no idea if the system changed meanwhile). In my eyes, it is the best solution.
Here a description of this system for people who dont know it:
1. A unique drops.
2. On all party members screens spawns a little time counter with three otions: Need, Greed, Pass. Players have 30 sec to make a choice.
3. After all players have choosen the desired option the server calculates per RNG a number for all choosers in the "highest" category (Need > Greed > Pass). All numbers will be displayed and the the highest number wins.
4. In WoW the item will be moved automatically to the winners inventory, in PoE the item may be locked for the winner for, lets say, 1 minute.
While I abhor FFA loot, I believe (and have said before) that this kind of system is the next worse thing. Here's why:
1) Screen clutter. I'm trying to play a game here, and I keep getting all these boxes I have to click to get rid of. It blocks the interface (can't click through it), and interrupts gameplay because I have to look at the item and choose an option, then get back to the game. Gets progressively worse as IIR and number of players increase. Options for rarity thresholds to roll can help, but the problem never goes away.
2) Does little to nothing to alleviate loot drama. Many people will roll on everything they can, and who can blame them? If nothing else they need orbs too, but there will still be crying from anyone who would actually use the item.
3) The only reason it is actually better than FFA is that it removes reaction time and proximity to the drop from the equation.
That is speaking to PUGs of course, that's what loot systems are designed for. An organized group can bypass whatever system is in place and allocate loot however they like.
Yeah, seriously. His notion is, I think, terrible. 30 seconds for players to click on a buch of shit to indicate if they want something?
No thanks.
Agreed, but I don't want to rail against the guy so hard. It IS a widespread, familiar system that more or less works in MMOs. It just isn't applicable to the faster pace of ARPGs, even if the two genres are similar in many respects.
:( my heart is broken, I really thought that GGG would do something with the desync, based on the "we are working on it", but apparently the laws of physics just can't handle PoE even when I played a lot of Arpg online and it was never this bad. I'm really frustrated... sorry to vent here, but man... I'm pissed off :/
To start with, I should note that we are currently planning to add several modes for Loot Allocation in parties: Full Free-for-all, something close to the current system, and a mode with longer allocation duration for people who don't want to deal with item tension. This decision was very hard to make because all of our developers are ninja-looters and love extreme item tension, but after talking with a lot of players, we realised that many people would have a lot more fun if we offered these options.
And they fucking caved in. Fuck you casuals and your fucking casual demands.
Apparently it's only a matter of fucking time before this game becomes just another D3. I didn't want to believe it. Fuck! And apparently when there's enough of shitty noobs whining on the forum GGG decides to fuck their core-players and do what the "majority" wants. Fuck this shit. Yes I fucking mad bro.
Wow, it already IS Diablo 3 with it's completely broken "economy" destroying the enjoyment of actually finding good items, in favor of the elitist MMO crowd. Maybe that's what GGG wants, but right now POE is shockingly similar to D3 in that it is an endless grinding slog, that forces you to use in game currency to get good items. Which fucking sucks for the vast majority.
I disagree. The games can be completely parallel and deterministic. The decision the client makes has to be the same decision the server WILL make (for non-cheaters.) The server and client (can) have exactly the same geometry, paths, mobs, everything. If decisions are being influenced by graphics cards, then that is just incorrect design. All that should happen is that the server plays exactly the same game (in effect shadowing the client), a little shifted in time.
Having the same RNG is not the point... Of course a pseudo-random sequence is deterministic. What YOU do isn't. Neither the time, nor the command issued. And since there's a delay between client and server, each will use slightly different assumptions to calculate pathing.
Clearly I am not expressing myself well. I am not claiming that users are deterministic, I am saying their generated click-stream can be reproduced exactly. Let me try again.
First, assume that we can get two or more computers to generate the same output given the same input. If we can't, we are in big trouble.
Next, forget about the real time clock. Think of a "game clock". The game clock is defined to tick upwards from time zero at 1000 ticks per second. All actions are referenced to this clock. We need the game clock so that both client and server can execute 1000 steps per second for the simulation - no more, no less. This keeps pathing (and all other computations, really) *reproducible.*
The client indicates that it wants to enter a level. The server sends the pRNG seed. The level is generated on both the server and the client. The two levels are completely parallel. That is, given the same pRNG seed, the random level generator on both ends will of course generate exactly the same map and mob layout. It has to.
The client has all of the information it needs. It starts the simulation and the game clock. Every mouse press and position and every keystroke is timestamped with the game clock. The client's version of the simulation, which should be identical to the server's or indeed to any repeat of the simulation on any other computer, starts to move mobs, determine paths, check hits, etc. in sync with the game clock. That is, if a swing is supposed to take 300ms and the mouse click that initiates the swing takes place at time x, then barring any interruptions that swing will complete at game clock x + 300.
If anything is going to interrupt that swing, the interrupter must be something that is present in the simulation, and therefore it will interrupt the swing in exactly the same place and the same time in every simulation of the game, no matter when it is executed - 100ms later on the server, or three days later when the devs are analyzing a replay of the game for bugs.
Since I am positing a solo game, there are no other inputs to the system other than the user's click stream. The stream ITSELF is not deterministic, but given the exact same click stream, the simulation must produce the exact same results on the client and server versions of the simulation in a non-cheating game. However, with a cheating client, the server will get out of sync with the cheater, and the server is always right - hence, no cheating.
Furthermore, if a strictly increasing sequence number is included with the input event, then dropped packets can be requested for re-transmittal from the server. That is, events are numbered as well as timestamped. As long as the client holds onto the click stream until the server says it is ok to dump it, then even dropouts should be handled with no desync issues.
Sorry to get so long-winded, but I feel strongly about at least having non-cheating solo players have the best possible game experience. I know that multiplayer can't work the same way exactly, because then each client has to have multiple and necessarily lagged input streams, and rubberbanding will be required. But a solo player is a special (and not uncommon) case.
