Pros/Cons of Server Based Refresh Times

mohio

In case anyone is unaware of what I mean in the title, what I would like us to discuss are the pros and cons of fixing refresh times as opposed to them being X amount of time from when you completed it. Clintman and maybe others have brought this up before, basically it would be everyone's nodes refreshing at midnight and noon EST (since these are the usual ending times for events) regardless of when you last played.

As I see it the Pros to implementing this are that you no longer have to concoct a weird schedule where you're playing 6-8, then 8-10, then 10-12 (all times eastern of course) or whatever the case may be for your play style. It also lets you play any time you want/are available during a given refresh without worrying about what consequences that will have on your NEXT refresh (and the one after, etc. etc.).

The only real Con I can see to this system is for people whose optimal play time is something like 11-1 EST, you'd have to really be aware of what time it is since nodes would refresh right in the middle of you playing. You could even argue this is GOOD for this person since they can theoretically kill 2 refreshes in one sitting, although that would require minimal/no rubber banding for all of us min-maxers here on the forum.

So - 2 questions for you guys.
1) Are there other Pros/Cons I haven't thought of or written about that are important to think about while evaluating this suggestion?
2) Would you prefer a server-based refresh time cycle (likely to be 12AM/12PM EST) to the current 12 hr since last completion of node refresh cycle?

NB: I know 8hr refreshes are possible but we haven't seen any of those in a while so the 12 hr ones are what I've been using as default, but keep in mind this could easily be done for 8 hr refreshes too (i.e. 8AM, 4PM, 12AM or swap AM/PM depending on start/end times)

NB2: Can anyone tell me if it's an 8 hr or 12 hr refresh for Simulator? Haven't started yet and really need to know.

Unknown

Not a big fan of this idea to be honest. At the moment, because you are factoring the time it clears nodes into how many refreshes you can do/want to do, you have a bit of wiggle room in when you play. Not playing for over 12 hours doesn't necessarily cost you a reset. On the other hand if you implement fixed refresh timers suddenly you're on the clock to play every 12 hours or miss one refresh out entirely.

The solution to people feeling they had to play constantly was rubber banding but that was unpopular enough to be severely lessened into a more grindy model for the most recent PvE event(s). Adding in fixed refreshes with ongoing minimal rubber banding just brings it to a new level of enforced play grinding....

rixmith

I'm not sure whether server based refreshes or personal refreshes would be better. I guess I'd have to try the server based to have a stronger opinion. The main thing for me is that I'd like the game to explain the behavior it wants from the players. Do you want me to play twice a day? Tell me so! Give me a countdown that show when the nodes will refresh so I understand what is going on. Then I can make a choice of whether I can meet the game's expectations or not.

Having to troll through large numbers of complex posts in an out-of-game forum in order to have any idea of how to play the game well doesn't seem right to me.

Wolarsen

Why not a more constant short time refreshing? I mean a small regeneration each few minutes?

Unknown

mohio wrote:

The only real Con I can see to this system is for people whose optimal play time is something like 11-1 EST, you'd have to really be aware of what time it is since nodes would refresh right in the middle of you playing. You could even argue this is GOOD for this person since they can theoretically kill 2 refreshes in one sitting, although that would require minimal/no rubber banding for all of us min-maxers here on the forum.

That's the only con you see? How about for the people who don't live in the US, where the ending times of events can often be terrible. Don't forget that rubberbanding, even minor rubberbanding will mean that you'll see the most points for playing during the last hour of a refresh. Am I expected to get up at 4am every morning (I'm a UK player) to max out my points before the event refreshes? How about at 4pm, which is when most of the people in the UK are still in work?

It wouldn't work unless they had zero rubberbanding in the event and the points at 5:01pm were the same as they would be at 4:59am, and that will never happen as everyone will end up on the same points by grinding each node to nothing.

Server based refresh times are a huge no-no for me.

Unknown

There's no way you end up losing time with serve based refresh. It's easiest to work backwards. Let's say the event ends at midnight, and it's 11am right now. You played 5 missions from 11am to noon. In the server based system you'd get those 5 missions, and then everything reset so you got 1 refresh left. Under the current system, you still have your 5 missions played, and then you'll have to wait until 11pm to play again and you only get a full refresh you're somehow able to crank out a full refresh in one hour.

Rubberbanding doesn't matter for the same reason. Anything you can do in a client-based refresh time refresh schedule you can do exactly the same thing on server based refresh because your missions, at worst, refreshes exactly as often as the client version so you can just match the same schedule. I guess people are thinking what if you say you did a full refresh 14 hours before an event end, and then with a server refresh schedule you completed it as soon as the event refreshed with 12 hours to go instead of 2 hours to go? But in that case you chose to play earlier. On the server refresh schedule you can just wait until 2 hours left to do the same missions even though refreshed 12 hours ago.

The con for the server based time is that in general everyone will have a slightly higher score so it is likely D3 set the progression rewards a bit higher, and it's disadvantageous to guys with the crazy schedule who could squeeze in the maximum number of refreshes.

Unknown

DaveyPitch wrote:

mohio wrote:

The only real Con I can see to this system is for people whose optimal play time is something like 11-1 EST, you'd have to really be aware of what time it is since nodes would refresh right in the middle of you playing. You could even argue this is GOOD for this person since they can theoretically kill 2 refreshes in one sitting, although that would require minimal/no rubber banding for all of us min-maxers here on the forum.

That's the only con you see? How about for the people who don't live in the US, where the ending times of events can often be terrible. Don't forget that rubberbanding, even minor rubberbanding will mean that you'll see the most points for playing during the last hour of a refresh. Am I expected to get up at 4am every morning (I'm a UK player) to max out my points before the event refreshes? How about at 4pm, which is when most of the people in the UK are still in work?

It wouldn't work unless they had zero rubberbanding in the event and the points at 5:01pm were the same as they would be at 4:59am, and that will never happen as everyone will end up on the same points by grinding each node to nothing.

Server based refresh times are a huge no-no for me.

You can fix the rubberbanding issue by calculating it at the refresh rather than dynamically, as I suggested here: viewtopic.php?f=8&t=6783

Unknown

Phantron wrote:

There's no way you end up losing time with serve based refresh. It's easiest to work backwards. Let's say the event ends at midnight, and it's 11am right now. You played 5 missions from 11am to noon. In the server based system you'd get those 5 missions, and then everything reset so you got 1 refresh left. Under the current system, you still have your 5 missions played, and then you'll have to wait until 11pm to play again and you only get a full refresh you're somehow able to crank out a full refresh in one hour.

Rubberbanding doesn't matter for the same reason. Anything you can do in a client-based refresh time refresh schedule you can do exactly the same thing on server based refresh because your missions, at worst, refreshes exactly as often as the client version so you can just match the same schedule. I guess people are thinking what if you say you did a full refresh 14 hours before an event end, and then with a server refresh schedule you completed it as soon as the event refreshed with 12 hours to go instead of 2 hours to go? But in that case you chose to play earlier. On the server refresh schedule you can just wait until 2 hours left to do the same missions even though refreshed 12 hours ago.

The con for the server based time is that in general everyone will have a slightly higher score so it is likely D3 set the progression rewards a bit higher, and it's disadvantageous to guys with the crazy schedule who could squeeze in the maximum number of refreshes.

If a sub is, say 2 days 12 hours long with 12 hour refreshes then, even though that's 5 refreshes in length you can't actually play 6 times because it leaves no time to actually beat the nodes. Instead you can play a maximum of 5 sets of missions fully refreshed. That leaves 12 hours which comprises the time you take to beat the missions and some left over for leeway. It means you can definitely go an entire 12 hour period without playing a node and still get a full set of refreshes in.

If, on the other hand, you refresh every 12 hours then that is no longer the case. If the nodes refresh, then 12 hours pass without me playing and they refresh again i've missed out completely. So for one it effects people who wish to play the full number of refreshes. To go further if I only wanted to do 3 run throughs because the event has rubber banding I would still have plenty of leeway (in fact even more since max rubber banding normally happens much earlier than 2 refreshes + 3 play times form the end of an event).

It increases the scheduling requirements of the game and the only benefits I can really think of are:

1) If you can't add 12 hours to the time you finished playing and remember it this solves that problem for you.....
2) If you can be free for the time it takes to clear the nodes either side of a server set refresh then you can play less often, but the same amount of time, for the same rewards (depending on whether rubber banding is present ofc, if it is you'll not benefit as much from it).

For me, those don't outweigh the increase in play time restrictions to do well.

Unknown

bonfire01 wrote:

Phantron wrote:

There's no way you end up losing time with serve based refresh. It's easiest to work backwards. Let's say the event ends at midnight, and it's 11am right now. You played 5 missions from 11am to noon. In the server based system you'd get those 5 missions, and then everything reset so you got 1 refresh left. Under the current system, you still have your 5 missions played, and then you'll have to wait until 11pm to play again and you only get a full refresh you're somehow able to crank out a full refresh in one hour.

Rubberbanding doesn't matter for the same reason. Anything you can do in a client-based refresh time refresh schedule you can do exactly the same thing on server based refresh because your missions, at worst, refreshes exactly as often as the client version so you can just match the same schedule. I guess people are thinking what if you say you did a full refresh 14 hours before an event end, and then with a server refresh schedule you completed it as soon as the event refreshed with 12 hours to go instead of 2 hours to go? But in that case you chose to play earlier. On the server refresh schedule you can just wait until 2 hours left to do the same missions even though refreshed 12 hours ago.

The con for the server based time is that in general everyone will have a slightly higher score so it is likely D3 set the progression rewards a bit higher, and it's disadvantageous to guys with the crazy schedule who could squeeze in the maximum number of refreshes.

If a sub is, say 2 days 12 hours long with 12 hour refreshes then, even though that's 5 refreshes in length you can't actually play 6 times because it leaves no time to actually beat the nodes. Instead you can play a maximum of 5 sets of missions fully refreshed. That leaves 12 hours which comprises the time you take to beat the missions and some left over for leeway. It means you can definitely go an entire 12 hour period without playing a node and still get a full set of refreshes in.

If, on the other hand, you refresh every 12 hours then that is no longer the case. If the nodes refresh, then 12 hours pass without me playing and they refresh again i've missed out completely. So for one it effects people who wish to play the full number of refreshes. To go further if I only wanted to do 3 run throughs because the event has rubber banding I would still have plenty of leeway (in fact even more since max rubber banding normally happens much earlier than 2 refreshes + 3 play times form the end of an event).

It increases the scheduling requirements of the game and the only benefits I can really think of are:

1) If you can't add 12 hours to the time you finished playing and remember it this solves that problem for you.....
2) If you can be free for the time it takes to clear the nodes either side of a server set refresh then you can play less often, but the same amount of time, for the same rewards (depending on whether rubber banding is present ofc, if it is you'll not benefit as much from it).

For me, those don't outweigh the increase in play time restrictions to do well.

Let's just take Thunderbolt Mountains since that was 2D 12H with 12 hour refresh schedule. Let's say you're able to do everything mission in 2 hours, then currently your schedule would look like:

2D 12H - start refresh 1
2D 10H - refresh 1 complete
1D 22H - start refresh 2
1D 20H - refresh 2 complete
1D 8H - start refresh 3
1D 6H - refresh 3 complete
0D 18H - start refresh 4
0D 16H - refresh 4 complete
0D 4H - start refresh 5
0D 2H - refresh 5 complete

So you get a total of 5 refreshs complete with the 2 hours to spare.

If this is on server refresh, we'll assume all missions refresh on 12 hour intervals, then:

2D 12H - event begins
2D 12H to 2D 0H - Do refresh 1
2D 0H - missions refresh
2D 0H to 1D 12H - Do refresh 2
1D 12H - missions refresh
1D 12H to 1D 0H - Do refresh 3
1D 0H - missions refresh
1D 0H to 0D 12H - Do refresh 4
0D 12H - missions refresh
0D 12H to 0D 0H - Do refresh 5

Note that every refresh you have a 12 hour period to do it, and you can complete 5 of them, same as the client model. You will do as good or better in all cases. Note that if the event went for say 2D 16H instead, under the server model you'd have 4 more hours to do your 6th refresh since the missions would've refreshed at the old 0D 0H mark, while under the current model you do not have enough time to do a 6th refresh no matter what (you completed your 5th refresh with 2H left, so adding 4 more hours and you're still short of the next refresh by 6 hours).

And since with a server based model you can do at least as many missions as the old model, this also means rubberband is irrelevent because you can always choose to simply do exactly the same number as the old model if there's some magic way to approach the events.

Unknown

Phantron wrote:

Let's just take Thunderbolt Mountains since that was 2D 12H with 12 hour refresh schedule. Let's say you're able to do everything mission in 2 hours, then currently your schedule would look like:

2D 12H - start refresh 1
2D 10H - refresh 1 complete
1D 22H - start refresh 2
1D 20H - refresh 2 complete
1D 8H - start refresh 3
1D 6H - refresh 3 complete
0D 18H - start refresh 4
0D 16H - refresh 4 complete
0D 4H - start refresh 5
0D 2H - refresh 5 complete

So you get a total of 5 refreshs complete with the 2 hours to spare.

If this is on server refresh, we'll assume all missions refresh on 12 hour intervals, then:

2D 12H - event begins
2D 12H to 2D 0H - Do refresh 1
2D 0H - missions refresh
2D 0H to 1D 12H - Do refresh 2
1D 12H - missions refresh
1D 12H to 1D 0H - Do refresh 3
1D 0H - missions refresh
1D 0H to 0D 12H - Do refresh 4
0D 12H - missions refresh
0D 12H to 0D 0H - Do refresh 5

Note that every refresh you have a 12 hour period to do it, and you can complete 5 of them, same as the client model. You will do as good or better in all cases. Note that if the event went for say 2D 16H instead, under the server model you'd have 4 more hours to do your 6th refresh since the missions would've refreshed at the old 0D 0H mark, while under the current model you do not have enough time to do a 6th refresh no matter what (you completed your 5th refresh with 2H left, so adding 4 more hours and you're still short of the next refresh by 6 hours).

And since with a server based model you can do at least as many missions as the old model, this also means rubberband is irrelevent because you can always choose to simply do exactly the same number as the old model if there's some magic way to approach the events.

Well firstly it doesn't take everyone 2 hours to do a refresh, in fact lots of ppl only run a limited number of nodes a limited number of times to avoid scaling issues. As I get further below 2 hours to clear I get more and more "time to spare" as you put it. Lets say, for arguements sake, I take 1 hour to clear the nodes i'm interested in, I now get:

2D 12H - start refresh 1
2D 11H - refresh 1 complete
1D 23H - start refresh 2
1D 22H - refresh 2 complete
1D 10H - start refresh 3
1D 9H - refresh 3 complete
0D 21H - start refresh 4
0D 20H - refresh 4 complete
0D 8H - start refresh 5
0D 7H - refresh 5 complete

Now, since I see no reason why I have to use the 7 spare hours I have to sit there twiddling my thumbs at the end. Instead I could let 14 hours elapse on 3 of my refreshes before starting, or 16 on one and 14 on another and still have a spare hour kicking around. These are both things that are impossible to accomodate with fixed server refreshes. Even if it took me 2 hours to clear the nodes I want to do (it doesn't) I could still have 13 hours off on 2 refreshes or 14 on one and get everything done, again something a fixed schedule does not allow for.

I didn't argue that you wouldn't get to do as many refreshes, I argued it's just ANOTHER set of restrictive fixed playing times for no gain.

Unknown

Let's take this schedule:

2D 12H - start refresh 1
2D 11H - refresh 1 complete
1D 23H - start refresh 2
1D 22H - refresh 2 complete
1D 10H - start refresh 3
1D 9H - refresh 3 complete
0D 21H - start refresh 4
0D 20H - refresh 4 complete
0D 8H - start refresh 5
0D 7H - refresh 5 complete

So let's say you want to find a 15 hour block where you can walk away from the game, then one possible way to it would be:

0D 21H - start refresh 4
0D 20H - refresh 4 complete
0D 8H - refresh complete (but still on break)
0D 5H - begin refresh 5
OD 4H - refresh 5 complete.

Now compare this to the server schedule:

1D 0H - missions refresh
1D 0H to 0D 12H - Do refresh 4
0D 12H - missions refresh
0D 12H to 0D 0H - Do refresh 5

You can simply pick 0D 21H to start refresh 4 and 0D 5H to start refresh 5 and get exactly the same result. In fact, you have 24 - (time it takes you do 2 refresh) hours on the server schedule. If it takes you an hour to a refresh, then you can squeeze in 2 refresh with up to 22 hour between them. Namely, you'd do this:

0D 24H - start refresh
0D 23H - finish refresh 1
0D 12H - refresh complete
0D 1H - start refresh 2
0D 0H - finish refresh 2

And you'd have a 22 hour period where you don't have to do anything (between 0D 23H and 0D1H).

Unknown

Phantron wrote:

-snip-

And you'd have a 22 hour period where you don't have to do anything (between 0D 23H and 0D1H).

Only has the same effect if you completely remove rubber banding which is not what the thread is about and not something that has been suggested by the devs, they just remove/almost entirely remove it in certain PvE then throw it back in for the next. This means doing 2 refreshes back to back either side of a reset (which i noted in a prior post as an option for server based refreshes) means the second set of nodes are worth bugger all. The player based refreshes work to give you wiggle room both without AND with rubber banding present.

Also worth noting that I get my break if and ONLY if I can be there for 2 hours across a server rest. So in order to give flexibility in play time I need to JUST schedule my life around the game for more than the ending time..... awesome solution

Unknown

bonfire01 wrote:

Phantron wrote:

-snip-

And you'd have a 22 hour period where you don't have to do anything (between 0D 23H and 0D1H).

Only has the same effect if you completely remove rubber banding which is not what the thread is about and not something that has been suggested by the devs, they just remove/almost entirely remove it in certain PvE then throw it back in for the next. This means doing 2 refreshes back to back either side of a reset (which i noted in a prior post as an option for server based refreshes) means the second set of nodes are worth bugger all. The player based refreshes work to give you wiggle room both without AND with rubber banding present.

Also worth noting that I get my break if and ONLY if I can be there for 2 hours across a server rest. So in order to give flexibility in play time I need to JUST schedule my life around the game for more than the ending time..... awesome solution

You need to name a particular playing schedule that somehow isn't possible with the server. We'll keep this simple. 24H left on event with 12H refreshes. Put any way of playing you believe is advantageous and outline the time frame you play and explain why it is not possible to do exactly the same thing with a server model.

NorthernPolarity

bonfire01 wrote:

Phantron wrote:

-snip-

And you'd have a 22 hour period where you don't have to do anything (between 0D 23H and 0D1H).

Only has the same effect if you completely remove rubber banding which is not what the thread is about and not something that has been suggested by the devs, they just remove/almost entirely remove it in certain PvE then throw it back in for the next. This means doing 2 refreshes back to back either side of a reset (which i noted in a prior post as an option for server based refreshes) means the second set of nodes are worth bugger all. The player based refreshes work to give you wiggle room both without AND with rubber banding present.

Also worth noting that I get my break if and ONLY if I can be there for 2 hours across a server rest. So in order to give flexibility in play time I need to JUST schedule my life around the game for more than the ending time..... awesome solution

Im trying really hard to understand your argument, but all im seeing are that it affects the psychology of the player: the systems are functionally identical. You say that if a player finishes a clear and the server immediately refreshes then the player is screwed if he does the next clear due to rubber, but theres absolutely no reason why he couldnt wait until the next refresh to maximize rubber.

Another thing you say is that the restrictions have no gain: i think the gain is that you arent in the dark about when refreshes happen. I hate having to track that i started node x at time y and finished my last node z at time w. Making it server based would relieve this issue completely since i can just schedule a time in each 12 hour block closest to when the server ends and is reasonable for me to play during and not have to worry about tracking when i play.

Like phantron said, give an example where the current system is functionally better than the fixed time system right now. I think youll find that the differences between the two are purely psychological.

I think the only thing bad with the suggestion is that it may remove some of the mystery of the system (and d3 loves having it be mysterious), and having two back-to-back refreshes be possible might not be good for average joe who just finished a clear and thinks "oh **** i have to do it again?". Intuitively the current system makes more sense and thats why i dont think demiurge will bother changing it. The new system would be better for us hardcore players in terms of managing our clears, but the elegance in the current system probably makes more sense as a whole.

Unknown

having it at 24hr left with 12hr refresh doesn't let me give a good example since, if you want to place particularly well and there is rubber banding you'll be playing at 12hr + 2 node clears to go then at the end of the event which is possible in both examples.

What i'm saying is fixed server side refresh times is more restrictive over the course of a sub.

So i'm playing a 2d 12hr sub and I want to get a decent score running up to the last 2 runs mentioned above without scaling. My best bet is for my first run to be all nodes x1 so I can see them then maybe another go on each essential if they're worth a decent amount. I'll then wander back to just do 2x on essentials and maybe next most valuable node(s) to stay reasonably in touch, rubber band last but one go hitting the nodes I need then do whatever seems necessary in the run up to closing.

Under a server based refresh system, in order to maximise my rubber banding I HAVE to play at set times. Although there is 12hrs between refreshes, playing at any time other than fairly close to the refresh timer costs me points meaning I either do worse of have to grind more nodes hurting my scaling. Doing the nodes I want just before a refresh then instantly running the nodes again is a PARTICULARLY bad plan from a points gained perspective.

So... assuming the above sounds reasonable here is my arguement...

Lets say i'm playing the simulator right now. I won't do every node cause of scaling and it takes me an hour ish to do the higher value nodes 1-2 times, with a plan to do a bit more 15 ish hours before the end then get up stupidly early to do a final clear before the end. Now the part that's fixed for me is a late as reasonably possible last but one clear and a clear when times coming up since that's what gets me a good placing in a rubber banding sub. Prior to that I just want to get a decent number of points with the minimum nodes cleared to keep scaling reasonable.

So that means I want to do the full number of refreshes with a good amount of rubber banding.

If the refreshes are server side then 5pm start with refreshes at 5am, 5pm, 5am, 5pm then finish at 5am. My issue comes with the fact that I have 4 fixed play deadlines 12 hours apart rather than JUST one fixed deadline at the end. If I intend to, say play at 4pm but get held up or have something else to do i'm screwed, one refresh gone, points gone, need to grind more, get scaled more etc etc.

In the current system I have the event end as a deadline and, practically speaking, if I want to place well i'll have a softer deadline to play my last refresh before the end as late as possible to rubber band and have time for a final clear before the end. The other refreshes have a lot of leeway. I can aim to play at a certain time but if I get cought up for an hour or 2 it's fine, i'll play then. As I posted above I have about 7 hours of leeway in a sub where my clears are an hour if I want to do max refreshes. there are NOT 5 fixed 12 hour windows where I have to find time to play or lose a refresh.

Now you can argue that I could play early in one 12 hour block and late in another to potentially get a long gap BUT that is both bad for rubber banding and I AM STILL having to play once in each arbritrary 12 hour block. I don't have that situation currently. While I have to find time to play a certain number of times and there are soft restrictions on when those have to be there is nothing so unneccesarily draconian.

NorthernPolarity

bonfire01 wrote:

Under a server based refresh system, in order to maximise my rubber banding I HAVE to play at set times. Although there is 12hrs between refreshes, playing at any time other than fairly close to the refresh timer costs me points meaning I either do worse of have to grind more nodes hurting my scaling. Doing the nodes I want just before a refresh then instantly running the nodes again is a PARTICULARLY bad plan from a points gained perspective.

So this is key idea that is confusing you. This statement applies to BOTH the current system and the proposed system. I can say the exact same thing about how it is right now: if the event is 12h refreshes and the sub is 1.5 days, doing the refresh at any other time besides the 25h, 11h, and 1h mark is just as inefficient.

bonfire01 wrote:

"Lets say i'm playing the simulator right now. I won't do every node cause of scaling and it takes me an hour ish to do the higher value nodes 1-2 times, with a plan to do a bit more 15 ish hours before the end then get up stupidly early to do a final clear before the end. Now the part that's fixed for me is a late as reasonably possible last but one clear and a clear when times coming up since that's what gets me a good placing in a rubber banding sub. Prior to that I just want to get a decent number of points with the minimum nodes cleared to keep scaling reasonable."

To make my original point even clearer, here is your words quantified into a simulator example:
1h before end - final clear
15h before end - second to last clear
random times before that - "decent amount of points clear"

Now, take this play schedule, and apply it to the server suggestion. The total number of points that you gain from either system due to rubberbanding is EXACTLY the same (assuming global point leader gains points at the same rate in both suggestions). You're still only experiencing the psychological effects of the change because having forced refreshes makes it clear what the optimal strategy is in order to maximize points. If you want to maximize points in the current system, you have to do exactly the same thing: the only difference is that its not explicit that that's what you need to do.

Again, if you want to show that your logic makes sense, you need to give a concrete example that proves what you're saying. Nothing forces you to play more in the server-refresh case: playing following the exact same schedule in either case (assuming each clear is spaced at least 12+ hours apart) will result in exactly the same amount of points gained from rubberbanding,

Unknown

A server based refresh schedule would make the runaway victories even bigger for events with weak/no rubberbanding. That is, someone who is dominating the opposition probably isn't going to rearrange his life completely to not miss any refresh, but he'll certainly grind it out every refresh if it takes no special effort to do so. However, we're talking about guys who have mindboggling high scores compared to the average guy, and you're really looking at say a guy dominating his bracket by 6000 or just 5000 points here. Sure, things might look more depressing than usual, but you were never going to catch those guys anyway under any system.

If anything, the higher points from those guys still help you a bit with higher rubberbanding for progression rewards. Even though the Thunderbolt Montains bracket had very weak rubberbanding, surely the fact that some guys are racking up 20K or whatever points in that sub bracket still helped those who are behind to hit the 45K Daken goal.

Unknown

I think player based refresh times are the way to go. What I think would work the best is have events run for a length of time, say a week or a week and a half. Make all 3* covers be additional progression rewards instead of rank rewards. Give people a fixed number of refreshes, with the countdown starting whenever they start the event, and nodes resetting to max points every 12 hours (and not increasing at all in the interim), regardless of when/how much they've each been played. Tally the totals once all players have finished an event or node, and give out placement rewards based off that, like ISO, hero points, tokens, but not the covers (since they're progression rewards instead).

Makes people feel like they have something to work toward, but not feel like it's a big gamble whether or not they actually get anything. Just make sure the progression rewards are attainable without grinding the game like it's a job. And tone down the scaling to keep it all playable.

Unknown

I surrender, not got time to argue more about it 8)