37 replies to this topic

[MrJeffers]

Wintersdark, on 16 September 2016 - 10:17 AM, said:

Numbers seem pretty stable to me. Steam release brought in a vast number of "tourist" players, who left fairly quickly, then there was another drop around April. What happened in April? Was there a specific event, or was that a group of Steamies who played for a few months then stopped? I suspect the later, to be honest, and that's a pretty normal thing. I know that this particular game aside (and World of Warcraft back in 2005 -> ~2008, I never play any given game for more than 6 months.

But the last few months? Looks very much like solid play over weekend events petering off midweek, which seems pretty reasonable and stable to me.

Bud Crue, on 16 September 2016 - 10:27 AM, said:

April townhall is what happened in April. Phase 3 also went live I believe in April (I may be wrong on that but there was much hand ringing about it that month). I think those two event (separately or together) caused perhaps more than a few bitter vets that had been hanging on, to finally call it a day.

The above may be part of it, but also in April was when all the rules and schedules for the MWO World Championship were announced. There was a lot of flack for this, and over the following months a lot of player burn out in the registered teams playing to try and accommodate the schedule with a lot of teams (I think players as well) dropping out in the June/July time frame.

[Wintersdark]

Don't get me wrong, though, the part of their budget that does go into development is often, imho, sadly misspent... But yeah. There's not really going to be a heck of a lot of time and money for new stuff after operating expenses.

Felt it was important to point that out though, because people often think of how many mech packs are sold in a month, draw up rough figures, then get all googly eyes at the number... But that number gets way less impressive after you start considering all the other stuff.

Also taxes. BC's taxes are the second highest in Canada, after Nova Scotia. To give you an example:

In 2012, I moved from Vancouver BC to Calgary Alberta. I left a job paying $15/hr, 40hr's a week, and moved to Calgary. I was hired at $23/hr (roughly 50% more), and still worked 40 hours a week. I took home $860 every two weeks in Vancouver. In Calgary, I took home $830... Per week. And that's not mentioning the 12% sales tax vs 5%, but that's less relevant)

MrJeffers, on 16 September 2016 - 10:58 AM, said:

The above may be part of it, but also in April was when all the rules and schedules for the MWO World Championship were announced. There was a lot of flack for this, and over the following months a lot of player burn out in the registered teams playing to try and accommodate the schedule with a lot of teams (I think players as well) dropping out in the June/July time frame.

Another side effect here is that the tournament client isn't on steam, so people playing in the tournament are probably totally removed from the steam numbers. I can't imagine the tournament load+practice leaving you wanting to play on the regular live servers too.

Edited by Wintersdark, 16 September 2016 - 11:28 AM.

[Johnny Z]

Bud Crue, on 16 September 2016 - 10:23 AM, said:

You be quite!

I have it on good authority (2 very "vocal" people on this forum) that Energy draw is going to bring in dozens, scores, perhaps thousands...droves even...of new players to MWO! Wait times will be near instantaneous once Energy Draw is here. And Bucket Combining...are you even serious? Combining Buckets is the future of competitive game play! Moreover, like Energy Draw, Bucket Combining is what we call "new content" around here. And everyone knows that new content is what draws players into an existing game like bees to honey.

So you can just stop with your "nothing in the pipeline" talk. The future is in that pipeline. And it is a glorious future filled with a golden single bucket of players and an energy draw system that I don't quite understand...but I will love it and it will attract new players...because of some reason...because PGI knows best!

Hes a smarty pants.

[xTrident]

I thought MWO forums users didn't give two licks what the Steam numbers were? For some reason the non-steam MWO players frown upon those who use Steam. They're all immature mindless FPS players that wouldn't know what it took to play as a team and use strategy let alone what a quality game was... So wth is going on here? Steam users do not matter! And forget the fact that nearly every match I see same the people.

[Wintersdark]

xTrident, on 16 September 2016 - 12:32 PM, said:

I thought MWO forums users didn't give two licks what the Steam numbers were? For some reason the non-steam MWO players frown upon those who use Steam. They're all immature mindless FPS players that wouldn't know what it took to play as a team and use strategy let alone what a quality game was... So wth is going on here? Steam users do not matter! And forget the fact that nearly every match I see same the people.

Why? Oh, I'm sure there's people who take that angle. But hell, I've been here from 2012, and I jumped on Steam the moment I could. I love it's automated background patching, the overlay, easy install/uninstall, etc. I <3 Steam to bits.


You certainly see the same people for a large number of reasons (which I can explain if you don't already know - hint; it's not about population), and would continue to do so even if we had twice the population we currently do.

[xTrident]

Wintersdark, on 16 September 2016 - 12:44 PM, said:

Why? Oh, I'm sure there's people who take that angle. But hell, I've been here from 2012, and I jumped on Steam the moment I could. I love it's automated background patching, the overlay, easy install/uninstall, etc. I <3 Steam to bits.


You certainly see the same people for a large number of reasons (which I can explain if you don't already know - hint; it's not about population), and would continue to do so even if we had twice the population we currently do.

I really question that theory - seeing the same people even with twice the population... Of course twice the population of what we have currently is still going to be rather small. And yes, I know, I just doubt everything works so good in this game that I'd get teamed with the same people.... Right.

[justcallme A S H]

Wintersdark, on 16 September 2016 - 10:17 AM, said:

Numbers seem pretty stable to me. Steam release brought in a vast number of "tourist" players, who left fairly quickly, then there was another drop around April. What happened in April? Was there a specific event, or was that a group of Steamies who played for a few months then stopped? I suspect the later, to be honest, and that's a pretty normal thing. I know that this particular game aside (and World of Warcraft back in 2005 -> ~2008, I never play any given game for more than 6 months.

But the last few months? Looks very much like solid play over weekend events petering off midweek, which seems pretty reasonable and stable to me.

Faction Play - Phase 3 happened, or well, the lack there of.

Also there have been some BIG events in the last 2 months which is causing much more population activity than normal. Not for these events, figures would be lower IMO.

[Wintersdark]

xTrident, on 16 September 2016 - 03:27 PM, said:

I really question that theory - seeing the same people even with twice the population... Of course twice the population of what we have currently is still going to be rather small. And yes, I know, I just doubt everything works so good in this game that I'd get teamed with the same people.... Right.

Ok, math time!

Lets use these figures:

Average match time: 8 minutes
Average time spent building a match: 1 minute
Average time spent between matches: 1 minute.

So, in a 10 minute stretch, you're only actually building a match 10% of the time.

I calculated an average player count online over the last monthish at 1300. We'll call our theoretical doublecount then at 2600.

We have several buckets. We've got the solo queue, group queue, and faction play.

It's been a long time since PGI shared ratios, but pre-FW it was 85% solo 15% group. Lets go with 80% solo queue. This seems pretty fair to me, and honestly changes here don't make a huge difference.

Keep in mind these counts are highly generous: They're assuming that (of the solo players), 100% are chaining match after match with an average out-of-match time, including matchmaking/searching, of 2 minutes. Anyone online but tinking in the mechlab, store, or eating dinner? They're counted as people chaining matches.

So, 80% of 1300 = 1040, and of 2600 = 2080

Now, we've 5 tiers of gameplay, but there's some spread in matchmaking. So lets call this three buckets.

Now our buckets are 346 / 692. These are the numbers of players that you're likely to be matched with.

Except only the players searching at the same time count. That's 10% of those players (remember the time distribution above?) - so that'll be, rounding, 35/70 people.

Of those, you need 24 people in a match. So that's one and a half or three matches worth of people at a time.

Now, because you're leaving matches at basically exactly the same time every match, and you're already matched with the people as close to you as the MM can currently manage, and you're requeueing at roughly the same time, with 1.5 or 3x a match population to choose from you're absofreakinglutely (if I may use the technical term) going to see a heck of a lot of the same people in every match.

This will be worse, of course, in lower pop times and not as bad in higher pop times, but as the MM makes new matches it's always trying to get players as close together as possible so each time it's searching, it's going to match the same people together more and more often, not less - kind of a genetic sort of thing, where people ranked closely together (however moronic the ranking system may be) are going to see each other more and more often, and thus requeue at exactly the same time more and more often, making it even more likely for the MM to group them together again and again.


So, there you have it.

That is why you see so many players over and over again, and why even doubling the population wouldn't significantly change that.

[Lukoi Banacek]

Plus plenty of people padded the Steam count and then stopped using Steam to run the game (myself included).

[AphexTwin11]

Wintersdark, on 16 September 2016 - 04:02 PM, said:

Ok, math time!

Lets use these figures:

Average match time: 8 minutes
Average time spent building a match: 1 minute
Average time spent between matches: 1 minute.

So, in a 10 minute stretch, you're only actually building a match 10% of the time.

I calculated an average player count online over the last monthish at 1300. We'll call our theoretical doublecount then at 2600.

We have several buckets. We've got the solo queue, group queue, and faction play.

It's been a long time since PGI shared ratios, but pre-FW it was 85% solo 15% group. Lets go with 80% solo queue. This seems pretty fair to me, and honestly changes here don't make a huge difference.

Keep in mind these counts are highly generous: They're assuming that (of the solo players), 100% are chaining match after match with an average out-of-match time, including matchmaking/searching, of 2 minutes. Anyone online but tinking in the mechlab, store, or eating dinner? They're counted as people chaining matches.

So, 80% of 1300 = 1040, and of 2600 = 2080

Now, we've 5 tiers of gameplay, but there's some spread in matchmaking. So lets call this three buckets.

Now our buckets are 346 / 692. These are the numbers of players that you're likely to be matched with.

Except only the players searching at the same time count. That's 10% of those players (remember the time distribution above?) - so that'll be, rounding, 35/70 people.

Of those, you need 24 people in a match. So that's one and a half or three matches worth of people at a time.

Now, because you're leaving matches at basically exactly the same time every match, and you're already matched with the people as close to you as the MM can currently manage, and you're requeueing at roughly the same time, with 1.5 or 3x a match population to choose from you're absofreakinglutely (if I may use the technical term) going to see a heck of a lot of the same people in every match.

This will be worse, of course, in lower pop times and not as bad in higher pop times, but as the MM makes new matches it's always trying to get players as close together as possible so each time it's searching, it's going to match the same people together more and more often, not less - kind of a genetic sort of thing, where people ranked closely together (however moronic the ranking system may be) are going to see each other more and more often, and thus requeue at exactly the same time more and more often, making it even more likely for the MM to group them together again and again.


So, there you have it.

That is why you see so many players over and over again, and why even doubling the population wouldn't significantly change that.

TL;DR - but seriously that was a good take on it

I for one, personally enjoy the Steam overlay quite a bit

Edited by AphexTwin11, 16 September 2016 - 05:07 PM.

[xTrident]

Wintersdark, on 16 September 2016 - 04:02 PM, said:

Ok, math time!

Lets use these figures:

Average match time: 8 minutes
Average time spent building a match: 1 minute
Average time spent between matches: 1 minute.

So, in a 10 minute stretch, you're only actually building a match 10% of the time.

I calculated an average player count online over the last monthish at 1300. We'll call our theoretical doublecount then at 2600.

We have several buckets. We've got the solo queue, group queue, and faction play.

It's been a long time since PGI shared ratios, but pre-FW it was 85% solo 15% group. Lets go with 80% solo queue. This seems pretty fair to me, and honestly changes here don't make a huge difference.

Keep in mind these counts are highly generous: They're assuming that (of the solo players), 100% are chaining match after match with an average out-of-match time, including matchmaking/searching, of 2 minutes. Anyone online but tinking in the mechlab, store, or eating dinner? They're counted as people chaining matches.

So, 80% of 1300 = 1040, and of 2600 = 2080

Now, we've 5 tiers of gameplay, but there's some spread in matchmaking. So lets call this three buckets.

Now our buckets are 346 / 692. These are the numbers of players that you're likely to be matched with.

Except only the players searching at the same time count. That's 10% of those players (remember the time distribution above?) - so that'll be, rounding, 35/70 people.

Of those, you need 24 people in a match. So that's one and a half or three matches worth of people at a time.

Now, because you're leaving matches at basically exactly the same time every match, and you're already matched with the people as close to you as the MM can currently manage, and you're requeueing at roughly the same time, with 1.5 or 3x a match population to choose from you're absofreakinglutely (if I may use the technical term) going to see a heck of a lot of the same people in every match.

This will be worse, of course, in lower pop times and not as bad in higher pop times, but as the MM makes new matches it's always trying to get players as close together as possible so each time it's searching, it's going to match the same people together more and more often, not less - kind of a genetic sort of thing, where people ranked closely together (however moronic the ranking system may be) are going to see each other more and more often, and thus requeue at exactly the same time more and more often, making it even more likely for the MM to group them together again and again.


So, there you have it.

That is why you see so many players over and over again, and why even doubling the population wouldn't significantly change that.

Solid points and I don't deny any of it... But you did say it yourself and I bolded such. Population matters and math aside higher populations means playing with less of the same people. That's my point. And as I said, doubling the current population - 2,600... Well that really isn't **** which is also my point. There simply isn't much of a population in this game and that's too bad.

Edited by xTrident, 16 September 2016 - 06:28 PM.

[Wintersdark]

xTrident, on 16 September 2016 - 06:24 PM, said:

Solid points and I don't deny any of it... But you did say it yourself and I bolded such. Population matters and math aside higher populations means playing with less of the same people. That's my point. And as I said, doubling the current population - 2,600... Well that really isn't **** which is also my point. There simply isn't much of a population in this game and that's too bad.

yeah, that's my point. Even "not as bad" in my post, if you assume peak time plays AND double the population from peak time, you'll still see the same names in matches, likely at least 30, probably as much as 50% of the players being that same, and that's doubling peak time player counts, not average.

For population to stop this you'd need literally orders of magnitude more concurrent users, due to how many times you're splitting the "current players" into different buckets.

We'll never see that, even if PGI starts doing everything right.

[jaxjace]

*Takes a drag*

As soon as PGI gets their **** together or sells the rights to another developer

*exhales*

[Dee Eight]

Not a game session goes by that I don't end up somehow with a "new" player, or many, because of how the PSR jumps quickly over relatively few games if they're high damage output wins. Also since I only use the N.A. servers, it often ends up being the game forms up pulling players from the other servers to fill out numbers (especially at oddball hours of the morning / night /pre-dawn/whatever), and by that time matchmaker has totally thrown tier #s out the window. I had a game last night with two blue on blue kills (I was one of them) and the guy who killed me tried denying over and over he'd done it, and the other one blamed it being 2:30am when someone else "came towards him" 30 seconds into the game (it was 4am EST though, so I don't know where he was actually). I looked up leaderboards afterwards (faction and quick tables) and neither had more than 20 games in the whole month. One had been a member three years, and only had THREE total quick games for this month.

[Wintersdark]

Dee Eight, on 18 September 2016 - 10:42 AM, said:

Not a game session goes by that I don't end up somehow with a "new" player, or many, because of how the PSR jumps quickly over relatively few games if they're high damage output wins. Also since I only use the N.A. servers, it often ends up being the game forms up pulling players from the other servers to fill out numbers (especially at oddball hours of the morning / night /pre-dawn/whatever), and by that time matchmaker has totally thrown tier #s out the window. I had a game last night with two blue on blue kills (I was one of them) and the guy who killed me tried denying over and over he'd done it, and the other one blamed it being 2:30am when someone else "came towards him" 30 seconds into the game (it was 4am EST though, so I don't know where he was actually). I looked up leaderboards afterwards (faction and quick tables) and neither had more than 20 games in the whole month. One had been a member three years, and only had THREE total quick games for this month.

See my math above. That's why.

[Moldur]

What I find most alarming is that the population dropped more steeply in the summer months. That is pretty unusual for a video game. People are out of school, people take more time off, etc. Even worthless games usually get more people in the summer.

[FallingAce]

The numbers were in steady decline until PGI decided to run an event EVERY weekend. Something like 11 weeks in a row.

[Odium]

FallingAce, on 18 September 2016 - 12:12 PM, said:

The numbers were in steady decline until PGI decided to run an event EVERY weekend. Something like 11 weeks in a row.

Not going to lie I only log in to play events. They give me goals to focus on.