New efforts!

[Angel Phoenix77]

https://i985.photobucket.com/albums/ae335/BlueMaxx86/tumblr_lc9ge9dK1R1qzado8o1_500.gif

Yep

ditto.............

[Felderburg]

With your "One day the phoenix will rise again" post signature, I was expecting a flaming Homer to fly out of the bushes.

[darkgob]

SO HOW BOUT THAT CITY OF HEROES

IT SURE IS A CITY WITH HEROES IN IT

[KennonGL]

Ok folks, I think all of these deceased equines have been fully beaten.
Let's move on to stop something else.

So, someone explain Arcana-time and its relation to survival/difficulty to me?

[Aggelakis]

Locking this thread temporarily while I clean this up. Don't create another thread in the meantime, I'll open it back up when I'm done.

[Aggelakis]

https://paragonwiki.com/w/images//0/00/Mod_hat.jpg

Cleaned up a crap ton...went back seven pages, deleted almost six pages worth. If I missed anything inflammatory, please tag the notify-mod button (don't post in the thread and don't PM me about it). If this line of discussion continues here (I cannot and will not control your PMs, but tbh that would be uncool), I am not averse to handing out temp or perma bans, depending on how assholish you are in the continuation. We're all City fans here. We can pretend to be friends while we wait. Back to talking about City, then we can meander off topic again later.

https://forwardwalkingdotcom2.files.wordpress.com/2013/03/be-excellent-to-eachother.jpg

[chuckv3]

Cleaned up a crap ton...went back seven pages, deleted almost six pages worth. If I missed anything inflammatory, please tag the notify-mod button (don't post in the thread and don't PM me about it). If this line of discussion continues here (I cannot and will not control your PMs, but tbh that would be uncool), I am not averse to handing out temp or perma bans, depending on how assholish you are in the continuation. We're all City fans here. We can pretend to be friends while we wait. Back to talking about City, then we can meander off topic again later.

So THAT explains why the page count actually went DOWN since my last visit. I thought I was hallucinating.

[crashpositron]

Hey, let's have some precision here!  Was it a imperial, avoirdupois or a metric crap-ton?    ;D

[blacksly]

Hey, let's have some precision here!  Was it a imperial, avoirdupois or a metric crap-ton?    ;D

By my measurements, it looks approximately equal to a short ton of crap. A short crap-ton, in other words.
Which is fortunate, because it's easier to flush down the internet flusher than a long crap-ton.

[darkgob]

SO HOW BOUT THAT CITY OF HEROES

IT SURE IS A CITY WITH HEROES IN IT

Nice, after the moderation this post becomes my 500th.  Quality posting.

[blacksly]

Ok folks, I think all of these deceased equines have been fully beaten.
Let's move on to stop something else.

So, someone explain Arcana-time and its relation to survival/difficulty to me?

Arcana-time is a more accurate way of measuring power activation time than the usual "how long does it take for a power to animate/execute" number (some detailed definitions to follow).

Its relationship to survival/difficulty is very hard to determine, because some characters really needed to worry about their power output per second (whether Holds, Damage, Heal, etc), while others such as Tanks generally worried more about steady-state damage mitigation via defensive toggles, so the activation time of their powers was not a major issue for their survivability.

Detailed definitions: First, a power may have a set animation time and also may have a different activation time. Usually that shouldn't be the case, but it sometimes happened. We'll merge these into one activation time that is the larger of the two, and that basically is how long the server takes from when it decides that the character is free to activate that power, until the character is done with the power's activation and free to activate a different power. So that is the "official" activation time of a power.

Second, this activation time of a power is server-side. However, we do not play on the server, we play on a separate client. So for a power to be activated, your client sends a message to the server saying "My character would like to activate Power X". Assuming nothing unusually bad happens, after some time, the client receives a message saying "Power X has been activated, what is your next command, O Master?". However, the time between the outgoing "start" and the incoming "finished" messages is NOT the same thing as the activation time of a power, because you have both lag and server ticks to worry about. Lag is different to each client and thus cannot be assigned to a power (we cannot determine how long a power will take to activate counting lag, unless we set lag to be a particular randomly-chosen reasonable number). However, the effect of server ticks can be assigned to a power. Once a power is finished activating, the "Finished activating Power X" message that gets sent back to the client can only go out at a server activity tick. So if the server's activity ticks are at 1.0 and 1.2 seconds after the power started activating, and the power is officially listed as taking 1.01 seconds to activate, then the message that it has finished activating can only go out at 1.2 seconds after the start. So your client will see a 1.2 second break between starting the power and receiving the notification that the power is finished. For practical gaming purposes, this 1.2 second is how long it takes for the power to activate and for the activation result to be received so that the client can then send an activation request for the next power. So the 1.01 second power really can be treated as a 1.2 second activation.

This latter activation time, that rounds up the official activation time to the next server tick increment, is the "Arcanatime" definition of a power activation time. More generally, Arcanatime states that events on the server that require communication with the client can only occur in intervals of server ticks, forcing most durations to be rounded up to the next tick interval. But it's usually only used for power activation time.

[Vee]

I can't help but consider my ridiculous anagram and spelling out of what KABLAM stands for getting roped in with the inflammatory stuff to be a major internet tragedy. That's several minutes of concerted activity now lost to the aether. Maybe we can get an in-game memorial in one of the successors?

[Codewalker]

This one?

Kablam! Earth's powerfulliest person! With the size of Krakoa, the sanity of Azrael, the old timey expressions of Beast, the patience of Lobo, the statutory rapeyness of Angel, and the shortpants of Magnus!

[Arcana]

Arcana-time is a more accurate way of measuring power activation time than the usual "how long does it take for a power to animate/execute" number (some detailed definitions to follow).

Its relationship to survival/difficulty is very hard to determine, because some characters really needed to worry about their power output per second (whether Holds, Damage, Heal, etc), while others such as Tanks generally worried more about steady-state damage mitigation via defensive toggles, so the activation time of their powers was not a major issue for their survivability.

Detailed definitions: First, a power may have a set animation time and also may have a different activation time. Usually that shouldn't be the case, but it sometimes happened. We'll merge these into one activation time that is the larger of the two, and that basically is how long the server takes from when it decides that the character is free to activate that power, until the character is done with the power's activation and free to activate a different power. So that is the "official" activation time of a power.

Second, this activation time of a power is server-side. However, we do not play on the server, we play on a separate client. So for a power to be activated, your client sends a message to the server saying "My character would like to activate Power X". Assuming nothing unusually bad happens, after some time, the client receives a message saying "Power X has been activated, what is your next command, O Master?". However, the time between the outgoing "start" and the incoming "finished" messages is NOT the same thing as the activation time of a power, because you have both lag and server ticks to worry about. Lag is different to each client and thus cannot be assigned to a power (we cannot determine how long a power will take to activate counting lag, unless we set lag to be a particular randomly-chosen reasonable number). However, the effect of server ticks can be assigned to a power. Once a power is finished activating, the "Finished activating Power X" message that gets sent back to the client can only go out at a server activity tick. So if the server's activity ticks are at 1.0 and 1.2 seconds after the power started activating, and the power is officially listed as taking 1.01 seconds to activate, then the message that it has finished activating can only go out at 1.2 seconds after the start. So your client will see a 1.2 second break between starting the power and receiving the notification that the power is finished. For practical gaming purposes, this 1.2 second is how long it takes for the power to activate and for the activation result to be received so that the client can then send an activation request for the next power. So the 1.01 second power really can be treated as a 1.2 second activation.

This latter activation time, that rounds up the official activation time to the next server tick increment, is the "Arcanatime" definition of a power activation time. More generally, Arcanatime states that events on the server that require communication with the client can only occur in intervals of server ticks, forcing most durations to be rounded up to the next tick interval. But it's usually only used for power activation time.

All good, except for the part about network lag.  The server clock alignment issue seems to be independent of network lag.  It even affects when NPCs are allowed to do things, because (obviously in retrospect) although the game animates at 30 frames per second the game doesn't make decisions like "should this NPC shoot at you" at the same frequency.  There are a number of different "clocks" in the game server that do things at regular intervals: process animations, process combat power execution, process user input, process motion, process NPC AI.  Although there were some inconsistencies in the theory I could never fully resolve, it appeared that processing user combat inputs (i.e. activate a power) happened on a 8 per second clock.  But for complex reasons even I cannot fully explain the "0.132" factor in the Arcanatime equation has to do I believe with an alignment between 30 per second clocks and 8 per second clocks, analogous to a "beat" in music (what is amazing is the fact that if you use 0.133 instead of 0.132 in the calculation you actually quickly find corner cases that are dramatically different than the observed behavior because of a shift in roundoff - the number is not "about" 0.132, but exactly 0.132, at least in the overall average case).

Incidentally, I didn't name this theory and calculation "Arcanatime."  I actually called it "server-side combat clock tailgating" which I knew was too long, but I couldn't think of an obvious short name or acronym like I did for other things (like DPA, which is actually one of mine). I believe it was Werner that began calling it "Arcanatime" when he started testing the theory, and the name stuck.  If my recollection is correct Werner was the first player (he was certainly one of the first) to publicly test the theory in the context of attacking pylons.  Attack chain calculations are pretty precise, and the attack chains used on pylons for practiced players is also very predictable, so it is possible to predict how long it will take to bring down a pylon with a specific repeated attack chain.  The calculations were always off by about five percent or so, which everyone generally attributed to lag effects of some kind or an unknown factor in how powers recharged.  When Werner recalculated his chains under Arcanatime and ran a pylon test to compare, his predictions ended up within less than 1% of the measured time, a dramatic increase in accuracy.  And thus, Arcanatime became the established way to calculate attack chains when a lot of accuracy was desired.

[Arcana]

I can't help but consider my ridiculous anagram and spelling out of what KABLAM stands for getting roped in with the inflammatory stuff to be a major internet tragedy. That's several minutes of concerted activity now lost to the aether. Maybe we can get an in-game memorial in one of the successors?

https://www.youtube.com/watch?v=6bOy3RNyWME

[Vee]

This one?

it lives! i might shed a tear of joy :P

trying to recover the anagram from erased bits on a piece of scrap paper. will probably take longer than it initially did but i have no limits on the amount of time i'll spend on the pointless and silly.

[LateNights]

https://www.youtube.com/watch?v=6bOy3RNyWME

Is there no way to properly embed a YouTube video?

[ukaserex]

It's not relevant at all - but rarely is anything I read in here relevant.

But, I thought Talladega Nights was okay. And I thought Pulp Fiction was okay. Neither impressed me as an experience to share with someone else.

Now, Catch Me if You Can - that was an awesome flick. Taking the labels off of model airplanes...don't think it would really work, but still, genius. Ill-focused genius, but genius none-the-less.

[Vee]

When Werner recalculated his chains under Arcanatime and ran a pylon test to compare, his predictions ended up within less than 1% of the measured time, a dramatic increase in accuracy.  And thus, Arcanatime became the established way to calculate attack chains when a lot of accuracy was desired.

presumably any remaining small discrepancy can be attributed to gravity waves or a reasonable beyonder-based facsimile.

[Angel Phoenix77]

With your "One day the phoenix will rise again" post signature, I was expecting a flaming Homer to fly out of the bushes.

that would be awesome :)