29 replies to this topic

[Kunae]

Tolkien, on 15 November 2013 - 12:51 AM, said:

Spoiler

Regional servers are a tricky one. PGI has always said that they will do it eventually, and I seem to recall them saying it would be shortly after launch (might even have been shortly before launch at one point, I think the quotes are in ATD if anyone wants to dig).

There are two big problems with regional servers and they both come down to population.

1) How many players will be needed to make community warfare work? The way they have described it, it seems to have at least the 6 houses, the clans, lone wolves, and player run merc corps. Even if there was only 1 Merc Corp this still splits the playerbase into 9 factions.

2) Compare the maps between the PGI proposed community warfare and the MultiplayerBattletech 3025 maps of the inner sphere>



Literally hundreds of worlds with an unknown number of connections to neighbors - if they implement the jumpdrive the way it worked in battletech lore I think you can have connections to 1-10 other worlds. Compare this to the much more modest MPBT3025 map>



I haven't counted them but there are maybe 75 territories (plus solaris....cough!) and they have a very simple to understand 2D border structure so you can have a proper 'front'.

My long winded point is that I don't know if the game will survive having the population split between 3 regional servers, and then into 9+ factions, and then spread across hundreds of worlds.

This might be fantasy thinking but maybe they'll be able to introduce a system where the core game remains unified, but the matches can be hosted on servers local to the majority of players. That would probably be best.

You are putting waaaay too much faith in PGI.

CW will be 3 orders of magnitude simpler than that, and more dumbed down.

[Tolkien]

I don't think I'm putting too much faith in PGI, I think I'm being a realist. Here's a link to something I posted back in December 2012.

http://mwomercs.com/...ost__p__1624485

In this long winded post I state my point of view that customization+role warfare+clan tech would make the game a nightmare to keep in balance.

Quote

I implore the devs to have the courage to give us more by giving us less. To have the vision to see where role warfare and customization and clan technology will take the game, and to reverse course before it's too late. I am proposing nothing less than making this game much more similar to what MPBT3025 was by removing customization, removing clan tech (at least for the time being), and focusing on delivering on a much narrower front.

I would rather play a modernized and streamlined MPBT3025 that will evolve and grow with PGIs ideas than the game that seems to be the logical result of 'role warfare' continuing to try to square the circle. There would be a lot of crying and gnashing of teeth since they would have to take away a lot of the options we have now, but I honestly think that role warfare and excessive player freedom to customize will end this game before it can get off the ground.

Now you can debate whether the game is off the ground yet or not, but it has at least launched. However I will point out that the player numbers from open beta can be calculated based on the infographic they released. In this thread I use the numbers they give us to calculate that during open beta up to mid july we had an average of 1623 players in matches at any given time.

http://mwomercs.com/...ost__p__2717775

If you've been keeping an eye on the forum population numbers and if you assume that they are proportional to each other, we have seen about a 50% roll off since July in active players. During the weekends the number of players on the forums would spike to about 2500+, this last weekend I saw it around 1300.

Granted we had our highest ever players the weekends following Phoenix which is good news in that it shows that people come back when big content drops, but the fact that it has fallen off so fast afterwards means the game is still not sticky enough to hold people for long.

Regional servers for the whole game are a dead end. You cannot take less than 1623 players, split them into 9 factions, then split that into 3 regions, then split that over 10 combat zones (there would only be 6 people per team in each).

I think they should copy the MPBT3025 model then use Mercenary contracts to make merc players fight for whichever side is losing badly (probably marik or liao). That would allow them to keep the game in dynamic balance by just having merc contracts get sweeter for whichever side is getting hammered.

Unfortunately I am afraid they are going to press ahead with a grandiose and baroque system that will not work well with the reality of limited players.

Edited by Tolkien, 15 November 2013 - 07:05 AM.

[DaisuSaikoro Nagasawa]

MonkeyCheese, on 12 November 2013 - 04:04 AM, said:

Quit whining, consider yourself lucky that you live in Australia where the internet is half decent in relation to NA based servers compared to other countries.

I have been playing this entire time at about 250-280 ping and yes while mwo hit detection has been up and down for everyone in the long run it has been perfectly stable and I have countless screenshots and stat screens where I have plenty of kills/damage in a wide variety of mechs.

Hell a couple of hours ago when crappy primetime AU internet/some local problem was pushing my ping up to 320ish ping I tailed a yellow open rear ct jenner and nailed it twice in my cicada lb10x for the kill.

Oh and a fun little fact for anyone who has ever played with or against me, thanks to "reasons" in my house I have always played over a "usually" 100mbs wifi connection.

And I have no problem landing hits on lights.

Edit- now if they would fix that entirely separate problem of the super spider..

Wow, that's a very non sympathetic jerky post.

I also wonder when and/or if PGI will add servers elsewhere... it's a valid question and didn't deserve such douchery.

[Tolkien]

See my post above Daisu.

I hold out hope that they will introduce combat host/instance servers to the regions, but hopefully they will not split the player base to regional servers - I think the sub 1600 active player average is too low to allow splitting.

I feel your pain about ping as even after switching from cable to a fiber connection I can't get much below 130.

[Heffay]

Tolkien, on 15 November 2013 - 07:11 AM, said:

I think the sub 1600 active player average is too low to allow splitting.

[Tolkien]

Heffay, on 15 November 2013 - 07:22 AM, said:

[/size]

They released an infographic near the end of open beta that gave a concrete number on the hours of combat spent during open beta (up to July) along with the amount of XP earned. It gives some pretty good insight into how many matches on average are going on at any given time, and from that we can get the average number of players online, and the answer was 1623.

The calculations are here: http://mwomercs.com/...75#entry2717775

Edit: Here's the link to the PCGamer article from July that has the infographic in it http://www.pcgamer.c...rify-engineers/

Edited by Tolkien, 15 November 2013 - 07:31 AM.

[627]

Australian Servers! Finally!!!

Edited by 627, 15 November 2013 - 08:35 AM.

[Heffay]

Tolkien, on 15 November 2013 - 07:29 AM, said:

They released an infographic near the end of open beta that gave a concrete number on the hours of combat spent during open beta (up to July) along with the amount of XP earned. It gives some pretty good insight into how many matches on average are going on at any given time, and from that we can get the average number of players online, and the answer was 1623.

The calculations are here: http://mwomercs.com/...75#entry2717775

Edit: Here's the link to the PCGamer article from July that has the infographic in it http://www.pcgamer.c...rify-engineers/

Nice math. I think it may be fairly accurate, at least within an order of magnitude. Before they got rid of the number of people online counter, I would see anywhere between 2000-5000 people playing. Since that was generally peak time and your 1600 number is averaged over the day, that would seem to fit in nicely.

No idea if those numbers are growing or shrinking, but I would guess there was a bump when the game when live and for the release of Project Phoenix as well.

Still, very interesting.

[Tolkien]

Heffay, on 15 November 2013 - 08:59 AM, said:

Nice math. I think it may be fairly accurate, at least within an order of magnitude. Before they got rid of the number of people online counter, I would see anywhere between 2000-5000 people playing. Since that was generally peak time and your 1600 number is averaged over the day, that would seem to fit in nicely.

No idea if those numbers are growing or shrinking, but I would guess there was a bump when the game when live and for the release of Project Phoenix as well.

Still, very interesting.

The numbers they released don't give any indication of minimum or maximum, or the variation over the day, but I do find them believable especially given the old player meter used to show 'a couple thousand' playing.

If we can use forum population as a proxy for in game population then there is good news and bad news. The good news is that population more than doubled after phoenix dropped, which means that big content releases can still bring many players around. The less good news is that things have rolled off again fast since last month.

I remember you posted in that thread talking about AOE:Online and how it was 'free to play, the wrong way', so I believe you watched the presentation. If so you'll understand that I think the ability the MWO team has to email past players is going to be critical in making sure things don't go the way they did with AOE. Here they can at least let people know when content drops.

[MonkeyCheese]

Daisu Saikoro, on 15 November 2013 - 07:04 AM, said:

Wow, that's a very non sympathetic jerky post.

I also wonder when and/or if PGI will add servers elsewhere... it's a valid question and didn't deserve such douchery.

Wasn't trying to be a jerk at all, was just saying that there are worse places to live on this planet in terms of internet speeds and reliability, the OP was saying that things were impossible, when coming from the same country and with the same conditions I know that they are not, the game is in a up and down hit detection cycle yes but outside of the spider mech and its well discussed current hit detection problems things are fine

LongJohnSilver, on 12 November 2013 - 12:20 AM, said:

There are plenty of occasions were my weapons will hit a 'mech standing still and nothing will happen no damage to them but I can clearly see that my weapons hit.

and I assume that the OP is talking about the spider battlemech here and the exact same problem that is happening with other mechs

to Quote myself from another thread

MonkeyCheese, on 14 November 2013 - 01:19 PM, said:

What you explain is hit detection, clearly when you do hit a spider it takes damage like that ac20 example, so the hitboxes are actually working as armor and internals do take damage, problem is as pgi have said in the command chair is that there is a problem with the spider hit detection.

If anyone remembers the old days with knockdowns that is a perfect example, 1. Mech falls over, 2. Enemy's shoot at Mech laying on ground, 3. When Mech quickly goods into stand up animation it jumps on the enemy's screen into a different location. It is clearly a similar problem with the spider skeleton, just because you shoot it on you screen didn't mean that is exactly where it is on the server.

Well that's my understanding unless the spider goes into godmode when overheated but I think the server location is a better reason.

Sad to say it but just don't even bother shooting at spiders right now unless its with lrms preferably with tag/narc/arti or unless you have multiple streaks, sure they may splat everywhere now days but they will still kill a spider.

MonkeyCheese, on 14 November 2013 - 01:54 PM, said:

Example here, server is clearly having similar jumping around issues with the "real" location of the spider.

These complaints of hit detection are just as much problems with the game code as they are with server/player locations and the problem is clearly more on the game code side of it as hit detection/server/yourmechs location and aim when compared to your enemys mech location has been a problem since closed beta.

The game is perfectly playable for people in Australia and many other parts of the world with semi decent internet but some people in Australia still have sub standard internet.