Final Fantasy XIV 1.23b - Would you Play it if Available?

[Gamer3427]

Kind of how it is like now with the telegraphs never lining up with when damage is calculated.
But for the love of all gods, the ui is still slow and awkward. I dont get why people feel like they improved it, when its still the same ''delay on every ui action''.

I agree with Raikki above. I think this is something to do with your connection. I have satellite internet, and am pinging at about 700ms from my server's IP address, and the only thing I've ever noticed is that a I have less of a dodge window for attacks than other players due to a half second lag time. I've only ever noticed any UI issues if I'm also downloading a lot or loading a lot of videos on my laptop as well while playing, (I'm on PS4 for reference). To me, the problems you describe as having sound like what it'd be like if I was trying to play while pushing my internet's capacity to the max, while a storm was interfering with my connection to the satellite......

[Aeyis]

I'm not experiencing any problems with this. Are you playing with exceptionally high ping or anything by chance? The UI is really snappy for me but I'm also only 30ms from the NA datacenter.

As for throwing away the good with the bad, there are definitely a few things where I'm not happy with the direction they took in 2.0, but I wouldn't classify it as "a lot". AV/CC runs were terribly tedious; leveling wasn't very good (still not strong but a little better); primals were pretty decent but I like them here, too; crafting I did like a little better in 1.x, though.

I will say, one big thing I preferred was the relic process in 1.x. I really don't care for it here.

I'm not surprised, even in 1.0 some people said they didnt notice.
Im talking about the UI delay here. How it ''lags'' every time you use the ui to open a window or do an action (start/finish craft for example, or place items in storage)

In the past I believe SE made statements that the delay was needed for extra data safety (better backing up or something), altho Ive always felt it has more to do with console limitations.
Its nothing to do with (only) server latency, as the delay is far greater then the expected 120ish (if east coast server) or 180 (if west coast server) ms server ping.

Its also present when checking for your characters position, making it a bit of a trial to move out of telegraphs on time.
For comparison, I'd have to check for the exact time, but the delay+ normal server latency allows for heal clipping at about 25% of the cast left.

In your case, because of your very low ping it might just be in the zone of ''bearable'' where as an extra 50-100 pings might make it fall outside of that zone for me, I suppose thats quite possible.



Edit: just checked. Spellcasting is canceled with a delay of > .5 seconds. So positions are updated only every 0.5 seconds in the best case scenario for me.
So in the worst case scenario (absurdly high ping ~200) it means a 0.3 second delay on updating positions.


I'm not aware if the UI delay is an issue that arises from a combination of poor server location (EU datacenters are located in the us/canada?) and the server delay, like the combat issue does.

I agree with Raikki above. I think this is something to do with your connection.

Its not a connection issue, its an actual server delay that SE confirmed existed. 0.7 seconds is actually smaller then the UI delay.

Incidentally as you are on a console, that might be a reason you dont see the delay. That or you arent sensitive to ui delays.
A forced delay doesnt feel remotely the same as a delay caused by irregular/high ping.

You need a better graphics card then, because the V1.0 UI was hella-slow because it didn't use the hardware mouse cursor at release... Here I can even find a webcomic making fun of it...


But the one referring to what you're talking about is here:

That has nothing to do with what I talked about tho. Ive never had any issue with getting my mouse to the right place in 1.0 or 2.0

Perhaps you guys dont craft a lot or something, never use storage, or sort items...or perhaps its a eu only issue.

[Raikki]

Im talking about the UI delay here. How it ''lags'' every time you use the ui to open a window or do an action (start/finish craft for example, or place items in storage)

I honestly for real don't experience any of that, and I'm pretty picky about UI responsiveness.

Spellcasting is canceled with a delay of > .5 seconds. So positions are updated only every 0.5 seconds in the best case scenario for me.

This is incorrect, but it's easy to see why people think this. Spells actually complete casting well before you see the cast bar complete, but there's a delay added before they appear, likely put in to allow for animations to sync up. It has nothing to do with position updates. Most Paladins take advantage of this behavior without really thinking about it when they learn self-stoneskin timing. Boss attacks that appear near the end of the cast bar never interrupt the stoneskin because the stoneskin is really already done casting.

[Aeyis]

This is incorrect, but it's easy to see why people think this. Spells actually complete casting well before you see the cast bar complete, but there's a delay added before they appear, likely put in to allow for animations to sync up. It has nothing to do with position updates. Most Paladins take advantage of this behavior without really thinking about it when they learn self-stoneskin timing. Boss attacks that appear near the end of the cast bar never interrupt the stoneskin because the stoneskin is really already done casting.

Im not sure why you assume this.


This is how I tested it:
Start a cast while moving, take note of the timer; repeat a hundred times. Nothing to do with when it ''completes'' I simply took note of how long it took for the timer to stop, while casting during continued movement.


Spells complete before the bar finishes because of the delay. I guess you arent as picky about cast delays are you are about ui delays. :P (then again, you dont notice the ui delay either)
What you seem to consider a novel feature is the very (combat/ability related) delay that I'm talking about.


Rather then making a real issue into a none-existant thing because you dont experience (notice it) yourself, your welcome to correct my methods of determining the delays presence.

[Raikki]

Spells complete before the bar finishes because of the delay. I guess you arent as picky about cast delays are you are about ui delays. :P (then again, you dont notice the ui delay either)

I don't "notice" delay because it doesn't exist for me. It's amusing that you purport to know better than I how "laggy" my game is and insist that I must have some bizarre perspective on time to account for the discrepancy.

your welcome to correct my methods of determining the delays presence

Sure thing. The most conclusive method is to use something like Wireshark to record network traffic while you cast cure spells on yourself, then isolate the packets associated with the beginning and ending of your casts and compare the time delta with the cast time of the spell.

For a more complete picture I rigged up a program to accurately record the timestamp when I pressed Stoneskin (on PLD with no spell speed obviously, so a full 3.00 second cast according to the game), so we can look at the complete time between me pressing Stoneskin and the Stoneskin cast being completed (and thus uninterruptible):

Cast 1: 2.749 seconds
Cast 2: 2.572 seconds
Cast 3: 2.614 seconds

This was in a housing zone. The variance between trials should be lower in a raid since they use a higher server refresh rate. All 3 casts nevertheless completed much more quickly than the advertised 3.00 second cast time (which is how long the cast bar displays).

(mods: while I do not believe anything in this post violates the letter or spirit of the ToS, if you deem otherwise feel free to redact some or all of it.)

-----

It's actually pretty easy to see why they'd do this. With no buffer time you'd end up with situations where you're seeing interrupts on your client during your successful cast animation. Just as an example, if someone had 200ms latency:

+0.000s - Player clicks Stoneskin, casting animation and cast bar begin instantly
+0.200s - Server receives Stoneskin cast command
+3.100s - Server processes Stoneskin being interrupted at 2.9/3 seconds
+3.300s - Player receives Stoneskin interrupt message 300 milliseconds after their cast bar completed -- frustrating and not desirable behavior.

[TheRogueX]

SE posed OP's question to users. ...the general answer was "no".

Personally, I'd like to see a return of 1.0, even if it was just a single-player game at that point. There's something kinda special about seeing one's own character in the cutscenes, not to mention the wealth of lore/dialogue that wasn't carried over into 2.0 or put on the 2.0 collector's edition bluray.

Err, well no. They posed the question to a tiny amount of players, compared to what we have now. The OP wasn't asking if 1.0 players would go back; his question was would you play if they brought it back, and it was aimed at the general population, not just the Legacy players.

For me, a non-legacy player, what story there was is the only reason I'd like to see it brought back.

[Dante_V]

Nah. I have no desire to do that at all.

[Felis]

For me, a non-legacy player, what story there was is the only reason I'd like to see it brought back.

the original mainstory or the 7th umbral story?

[Aeyis]

Sure thing. The most conclusive method is to use something like Wireshark to record network traffic while you cast cure spells on yourself, then isolate the packets associated with the beginning and ending of your casts and compare the time delta with the cast time of the spell.

For a more complete picture I rigged up a program to accurately record the timestamp when I pressed Stoneskin (on PLD with no spell speed obviously, so a full 3.00 second cast according to the game), so we can look at the complete time between me pressing Stoneskin and the Stoneskin cast being completed (and thus uninterruptible):

You fail to mention just what program you used and how it works.

Cast 1: 2.749 seconds
Cast 2: 2.572 seconds
Cast 3: 2.614 seconds

This was in a housing zone. The variance between trials should be lower in a raid since they use a higher server refresh rate. All 3 casts nevertheless completed much more quickly than the advertised 3.00 second cast time (which is how long the cast bar displays).

(mods: while I do not believe anything in this post violates the letter or spirit of the ToS, if you deem otherwise feel free to redact some or all of it.)

-----

It's actually pretty easy to see why they'd do this. With no buffer time you'd end up with situations where you're seeing interrupts on your client during your successful cast animation. Just as an example, if someone had 200ms latency:

This doesnt make sense. If you used programs to accurately note the cast time, then if theres a buffer there it should have the same value (all casts should take roughly as long). This is assuming that you did your test with stable ping.

+0.000s - Player clicks Stoneskin, casting animation and cast bar begin instantly
+0.200s - Server receives Stoneskin cast command
+3.100s - Server processes Stoneskin being interrupted at 2.9/3 seconds
+3.300s - Player receives Stoneskin interrupt message 300 milliseconds after their cast bar completed -- frustrating and not desirable behavior.

Your hypothesis falls flat on its face however, as I took measure of the casting time during continous movement. In all cases it took >500ms to interupt the cast.


I'm also not sure if you are aware how this normally works. Under normal conditions (without a delay between updating, and assuming your example of 200ms ping) it would only take 1x 200ms to update your movements. In other words all your actions would be updated with 200ms delay, and they would happen 200ms later.
In other words something like:

With no buffer time you'd end up with situations where you're seeing interrupts on your client during your successful cast animation.

Can not happen. Atleast without buffer time, interesting is that it can happen with buffer time.


Add to that the fact that 200ms ping is really high by todays standards.


What I take from your post is that your readings, despite the programs you used arent more accurate then my method. And interesting they both take note of the very delay I argumented still exists.


So, I asked if you could confirm via better methods that the delay does not exist, your answer is that it does indeed exist?

I don't "notice" delay because it doesn't exist for me. It's amusing that you purport to know better than I how "laggy" my game is and insist that I must have some bizarre perspective on time to account for the discrepancy.

My apologies. I thought you were seriously arguing a point, instead of flaming.

The delay exists, as you yourself have proven. (altho the way you proved it doesnt make sense..)
I have also argued that its possible that users of consoles dont notice this delay, perhaps it does not even exist for them.
If you would instead ignore these arguments and choose to flame, thats your call.


Fact: the delay exists.
Not a fact: whether it effects only regions, or only effects pc. Whether everyone is capable of noticing delays to the same degree. What the best ff game is.

[Raikki]

You fail to mention just what program you used and how it works.

Anyone with a modicum of programming experience can reproduce the simple program I made. Just bind a specific key to Stoneskin, then write and run a program that uses a low level keyboard hook to watch for that key being pressed and record the exact time. Run this in parallel with Wireshark and combine the data.

This doesnt make sense. If you used programs to accurately note the cast time, then if theres a buffer there it should have the same value (all casts should take roughly as long).

The server has a set rate that controls how often it processes events. In open world zones, this seems to be around 3-4 times per second, while in coil and primals it's much faster (seems to be doubled, but further testing would be needed to pin down both these numbers). Even game servers for twitchy FPS games work this way, although some have very high rates (high-end Counter-Strike servers often processed at 100Hz). High rates are going to greatly increase server load and network congestion in relation to how many entities there are, which is why they're reserved for content where response time is more important (and there are fewer entities).

My apologies. I thought you were seriously arguing a point, instead of flaming.

I tried to shed some light because you clearly don't understand what's going on and how the game works. Whether you choose to absorb new knowledge or ignore it is up to you. You complained about not being able to cancel casts at the end of the cast bar and I showed you why.