Is 500 vs 500 PVP in Ashes of Creation a REALITY?

Nanfoodle

Short video on some real tech...

Slapshot1188

I used to find his videos informative but he went total fanboy/apologist a year or so ago.
This "proof" video is pretty sad.

Jacobs and Camelot Unchained showed 1500 bots fighting it out 4 years ago. Here is a piece from their Newsletter.  Not sure if they released the video (I think they did) but if have alpha Access you can find it there.

"2. Tech –  Bots: We’ve had over 1500 Bots running around, using abilities, and firing effects-laden projectiles at one another! This is more impressive than the last time we had 1500 Bots in a small area, as now Bots are now all individual player clients, use both melee and ranged abilities, use items such as arrows, and are stress testing our recent VFX improvements! And we’re still at a great framerate that we will continue to improve. Overall, the past three weeks have had a solid team effort to improve our code, and hence overall client stability and performance! Last night, we hit 2.1K Bots, comprised of 360 archers, about 1800 fighters, and at peak over 200K particle effects a second. Until we hit about 1800 Bots (in a small area, as usual), the game was quite playable on Mark’s rig."

--------------
From Massively
Mark Jacobs

Hey OC, great question, here’s a quick explanation:

1) Bots are “headless clients” running on separate AWS instances. This means that on different instances (servers), we are running 1500 separate clients. They are headless, because in our case, they aren’t attached to real players at keyboards, monitors, etc.

2) Bots are almost indistinguishable to the server from players. They are started up from player characters we created just for that purpose. As such, they have full access to all the skills, abilities, etc. as players.

3) Because Bots = players, the server treats them the same way as players (except they don’t get to ask for refunds!) and as such, consume the same amount of network traffic as players. So when a Bot is firing off an ability, it has to send the same message to the servers that players do and in return, are sent the same amount of information from the server.

4) Bots, in general, consume a bit more network traffic than players since they don’t spend any time talked, go off for “botty-breaks”, etc., like normal players do. In some tests we actually had them chatting so they could test the chat system.

5) Bots are not “NPCs” because those things are usually controlled from the same server(s) instance(s) and are generally not networked like players. We will, of course, have NPCs too but no NPCs were harmed in the making of these screenshots!

So, in terms of whether are Bots are the same as players in terms of Internet/network traffic and also client stability (they can crash too), etc., the answer is 100% yes! Otherwise, if they weren’t, we’d be, to put it mildly, lying and even though you and I disagree about some things, I hope by now you and our Backers know that I refuse to lie, I just rather deliver bad news. FYI, I wanted to have these kind of Bots for testing in WAR. Unfortunately, that was vetoed. Fortunately, this time my partner/co-founder Andrew couldn’t have been more in favor of it and was a strong proponent himself. And thanks to the other engineers at CSE and the considerable time spent getting them up and running, they are really incredibly useful for these types of tests, then, now and forever.

FYI, interesting historical note, the concept of using Bots for this kind of testing goes way, way back, even to the origin of the MUDs. One of the earliest members of Mythic Entertainment, and a fine gentleman who worked for me at AUSI (my first company, maker of the world’s third MUD, Aradath), Darrin Hyrup, had a similar system to stress test our MUDs. Very helpful then, incredibly helpful now. For example, In Beta, we’ll be using them every few weeks to beat on the current build to make sure that nothing bad has crept into the code networking-wise. No matter how good the programmers/designers/artists, occasionally things go into the game that looks fine but when systems are pressured, bad things can arise. The Bots aren’t a cure-all, but used properly they can cut down on the number of times things seem to work great when 40 people are playing during testing but go to complete crap when 4000 people are playing the same build when it goes LIVE.

-------------------

Similar to CU, what was demoed were "bots" and despite Jahlon saying "They're not just standing still" you can clearly see that they are in fact just standing still while executing the same action over and over again.


Can they deliver 500 vs 500?  I hope so... but this video was doesn't really help demonstrate anything of worth. 

When are the real 500 vs 500 player battles with spells and siege engines and all that?  At this point, the game needs less "promises" and more deliveries.  Don't show me what might happen, do it.  And then the game will sell itself.

[Deleted User]

The user and all related content has been deleted.

Nanfoodle

They have already done tests of 200 v 200 successfully but thats with only the Battleground running. They ran it at 4k on a 1080 ti at I think 60 FPS and it was lag free, this while running their over sized partial effects. The partial effects will be scaled before launch and that will only help with large scale and lag. As for bots, I dont see how that impacts people in a PvP battle, as bots dont normally get into that type of playing. They look for nods for crafting and gold. Also, botts in this game will be fun to mess with, as this is a open world PvP game. Want to spend an afternoon killing bots. Have fun. 

Slapshot1188

Nanfoodle said:

They have already done tests of 200 v 200 successfully but thats with only the Battleground running. They ran it at 4k on a 1080 ti at I think 60 FPS and it was lag free, this while running their over sized partial effects. The partial effects will be scaled before launch and that will only help with large scale and lag. As for bots, I dont see how that impacts people in a PvP battle, as bots dont normally get into that type of playing. They look for nods for crafting and gold. Also, botts in this game will be fun to mess with, as this is a open world PvP game. Want to spend an afternoon killing bots. Have fun. 

I think something was lost in translation.  The bots in the video and in CY are what the quotes were about.   Both use bots to "simulate" real players but it's not even close to being the same thing.

Nanfoodle

Slapshot1188 said:

Nanfoodle said:

They have already done tests of 200 v 200 successfully but thats with only the Battleground running. They ran it at 4k on a 1080 ti at I think 60 FPS and it was lag free, this while running their over sized partial effects. The partial effects will be scaled before launch and that will only help with large scale and lag. As for bots, I dont see how that impacts people in a PvP battle, as bots dont normally get into that type of playing. They look for nods for crafting and gold. Also, botts in this game will be fun to mess with, as this is a open world PvP game. Want to spend an afternoon killing bots. Have fun. 

I think something was lost in translation.  The bots in the video and in CY are what the quotes were about.   Both use bots to "simulate" real players but it's not even close to being the same thing.

I get what you are saying. Yes real world and simulations are always different. But being able to run 1300 moving assets is an accomplishment, even at 30 FPS. ESO used to run Cyrodiil at 600v600 with lag and now you cant do 50v50 without lag in ESO. Large scale PvPers have not had a real home since DAoC. So fact that someone is working on this tech is a real happy place for many RvR fans. 

Slapshot1188

I was going to explain again... but perhaps the best response is simply:

We will see

TwistedSister77

I'm not holding my breadth... but hope someone can do it on modern games.

Scorchien

tzervo said:

Wargfoot said:

To be honest, 50 v 50 with no lag would be game changing.
I don't even see where 1/10 of the claims here have been successful.

Dunno, I think GW2 WvW, Albion and Foxhole are already doing fine with 50 vs 50 (or maybe it is less and I am too optimistic?). Not completely lag-free but pretty close (although I know that people had issues with GW2 WvW and will slap me in the face for saying I never had serious problems with it )

I'll add I've been in quite few 500 vs 500 or more ( i remember a 650 vs 650) on The Warhammer Emu server , and it performed well , now they have a guide up with everything you need to do along with addons to achieve this but it does work . 

Iselin

Nanfoodle said:

Slapshot1188 said:

Nanfoodle said:

They have already done tests of 200 v 200 successfully but thats with only the Battleground running. They ran it at 4k on a 1080 ti at I think 60 FPS and it was lag free, this while running their over sized partial effects. The partial effects will be scaled before launch and that will only help with large scale and lag. As for bots, I dont see how that impacts people in a PvP battle, as bots dont normally get into that type of playing. They look for nods for crafting and gold. Also, botts in this game will be fun to mess with, as this is a open world PvP game. Want to spend an afternoon killing bots. Have fun. 

I think something was lost in translation.  The bots in the video and in CY are what the quotes were about.   Both use bots to "simulate" real players but it's not even close to being the same thing.

I get what you are saying. Yes real world and simulations are always different. But being able to run 1300 moving assets is an accomplishment, even at 30 FPS. ESO used to run Cyrodiil at 600v600 with lag and now you cant do 50v50 without lag in ESO. Large scale PvPers have not had a real home since DAoC. So fact that someone is working on this tech is a real happy place for many RvR fans. 

You missed a 3rd side in your ESO Cyrodiil numbers but as far as I know, even in beta, they never did 600 x 600 x 600 there. 250 x3 they did for sure and the game actually released with that and they might have done larger than that in beta but I don't remember any bigger.

So 750 all total for ESO was definitely done with good performance back then but they can't do that many any longer. Haven't played for a year and it was down to about 150 X 3 at that point. It may be lower now with all their performance issues.

YashaX

Nanfoodle said:

They have already done tests of 200 v 200 successfully but thats with only the Battleground running. They ran it at 4k on a 1080 ti at I think 60 FPS and it was lag free, this while running their over sized partial effects. The partial effects will be scaled before launch and that will only help with large scale and lag. 

I have not seen any footage of real 200 vs 200 player battles in Ashes, do you have a link for this?

*edit: I found one for 100 vs 100. Looks in a better state than CU was during beta, but far from lag free and 60fps:

[Deleted User]

The user and all related content has been deleted.

Vrika

Nanfoodle said:

Slapshot1188 said:

Nanfoodle said:

They have already done tests of 200 v 200 successfully but thats with only the Battleground running. They ran it at 4k on a 1080 ti at I think 60 FPS and it was lag free, this while running their over sized partial effects. The partial effects will be scaled before launch and that will only help with large scale and lag. As for bots, I dont see how that impacts people in a PvP battle, as bots dont normally get into that type of playing. They look for nods for crafting and gold. Also, botts in this game will be fun to mess with, as this is a open world PvP game. Want to spend an afternoon killing bots. Have fun. 

I think something was lost in translation.  The bots in the video and in CY are what the quotes were about.   Both use bots to "simulate" real players but it's not even close to being the same thing.

I get what you are saying. Yes real world and simulations are always different. But being able to run 1300 moving assets is an accomplishment, even at 30 FPS.

Getting that many moving assets is only one part of the challenge.

Another part is when actual players have huge amount of different equipment and skills in play - as opposed to those bots who use just a few.

Then another huge part is having the server deal with all the participants moving around - as opposed to that video where you had one camera moving around while the bots stayed in one place so that the server didn't have to update everyone's positions to everyone.

[Deleted User]

The user and all related content has been deleted.

[Deleted User]

The user and all related content has been deleted.

[Deleted User]

The user and all related content has been deleted.

Iselin

@Wargfoot


IDK about others but when I talk about having fun in RvR with 500 v 500 v 500 I'm thinking of a large zone or zones with many, many objectives that contribute to an overall win.

In Dark Age of Camelot a push for a relic (which gave you nice bonuses in both PvP and PvE if you were successful in stealing one form the other side) could have as many as 500+ players on your side in the raid.

But that was probably 100+ going after non-relic keeps as a faint in both of the other realm's frontiers, a couple of hundred in strategic choke points to distract and prevent reinforcements and the rest sneaking around the edges of the map to go after the relic keep. It was very rare to have more than 80-100 v. 80-100 in any one spot.

It was a blast to coordinate and pull that off.

I have no idea what 500 v 500 in one spot would play like as I've never seen it but I suspect it would be a very crappy, chaotic experience.

[Deleted User]

The user and all related content has been deleted.

Scorchien

Iselin said:

@Wargfoot


IDK about others but when I talk about having fun in RvR with 500 v 500 v 500 I'm thinking of a large zone or zones with many, many objectives that contribute to an overall win.

In Dark Age of Camelot a push for a relic (which gave you nice bonuses in both PvP and PvE if you were successful in stealing one form the other side) could have as many as 500+ players on your side in the raid.

But that was probably 100+ going after non-relic keeps as a faint in both of the other realm's frontiers, a couple of hundred in strategic choke points to distract and prevent reinforcements and the rest sneaking around the edges of the map to go after the relic keep. It was very rare to have more than 80-100 v. 80-100 in any one spot.

It was a blast to coordinate and pull that off.

I have no idea what 500 v 500 in one spot would play like as I've never seen it but I suspect it would be a very crappy, chaotic experience.

Similar to what Iselin describe above,in Warhammer during the 500vs 500 ive been in there has been 20+ Warbands per side there are 25 in a WB and 2 Keeps and 8 control points..

    In a Lake .. So for a large part of the battle it is spread out over the lake as the 2 sides battle for resources to attack or Defend a Keep .. The Keep battles are also fairly spread out as the Keeps are Huge ..

Ungood

Iselin said:

I have no idea what 500 v 500 in one spot would play like as I've never seen it but I suspect it would be a very crappy, chaotic experience.

NGL, I don't get this urge to want these big battels.

It's going to end the way these games always end, with large Zerg Guilds ruling the landscape, were the vast majority of the members accept that they are just grunts in what is a player ruled military industrial complex. 

All the special snowflakes, with their 5 person statics or small guilds, will end up crying that they are little more than speedbumps in the grand order of things, and that is on their best days, at worst they are cussed out and told to leave the zone so the larger war ready guilds can cap the zone with their members.

Slapshot1188

YashaX said:

Nanfoodle said:

They have already done tests of 200 v 200 successfully but thats with only the Battleground running. They ran it at 4k on a 1080 ti at I think 60 FPS and it was lag free, this while running their over sized partial effects. The partial effects will be scaled before launch and that will only help with large scale and lag. 

I have not seen any footage of real 200 vs 200 player battles in Ashes, do you have a link for this?

*edit: I found one for 100 vs 100. Looks in a better state than CU was during beta, but far from lag free and 60fps:

Yeah I remember that video.   For an Alpha I think it ran fine.  The problem is that these guys (and Jahlon now) are always so over the top with GLORIOUS ASHES crap...

Just deliver!  if its 300 people on screen, and it runs well and it's FUN then that's a fucking homerun.  if its 1500 people on screen, and its runs well, and its a mass of shit then it's a FAIL.

I hate the model today where these streamers are paid advertisements and even worse in Ashes where these guys get a piece of every sale from every person they recruit forever.  I do not believe a word any of them say.

Just let me know whats actually IN the game when it launches.  If they manage a fun RvR experience I'll be all over it like white on rice.   But they aren't going to trick me into plopping down hundreds of dollars to go look at their Alpha because of some bot video.

Mendel

Large scale activities in MMORPGs have always had difficulties.  Even PvE raids with 70 people were almost impossible in the days when the chat channels only held 50 people.  Coordination was a nightmare, and tactics became meaningless as the only thing that really worked  consistently for everyone was "stay with the group".  I.e., zerg tactics.

A game may have the ability to run 500x500x500 static bots on an internal test.  The real test is to add in multiple active chat channels, unpredictable non-scripted behavior, and server (and network) contention.  Then factor in the odd graphics/CPU generated lag on various clients on substandard PCs.  Adjusting everything to make a smooth, playable experience is a considerable task that few have been able to achieve.

Good luck to those trying.

nurso

Slapshot1188 said:

Jacobs and Camelot Unchained showed 1500 bots fighting it out 4 years ago. Here is a piece from their Newsletter.  Not sure if they released the video (I think they did) but if have alpha Access you can find it there.

Small correction: Officially there were 2809 human players and ARCs (autonomous remote clients, no bots!), see https://camelotunchained.com/v3/humongous-multiplayer-battles-friday-january-25th-2809/.

Unofficially, they increased the number of ARCs by quite a bit just or fun. 

nurso

Wargfoot said:

...and lets think about this 500 v 500 nonsense.

For your part, the only thing you'll be worried about are the people who are in range to either attack you or be attacked by you.  What is the dream here? 

Scalability. Somewhere on the world map a small fight develops between two small groups, say 5 vs. 5, and more and more combatants join. Then a third and fourth faction become aware, and suddenly it's 500 vs. 500. There's no lag, all players averaging 60+ fps, and it's all non-instantiated. That's the dream.

nurso

Wargfoot said:

I don't know if the bot test is the best approach.

Maybe have the people who kicked other teams by synchronized casting of lag generating skills in New World should be called in to do the testing?

Bots don't intentionally try to bring the server down.

So in CU's case, they use Autonomous Remote Clients (ARC) which, as @Slapshot1188 mentioned, are "'headless clients' running on separate AWS instances". These ARCs are normal game clients just without keyboard, mouse, etc. controlled by a simple program.  Thus they are "indistinguishable to the server from players", especially when it comes to network traffic. In fact, ARCs "consume a bit more network traffic than players".

And, again in CU's case, these ARC-tests were designed to push the server to its limits.

Kyleran

nurso said:

Wargfoot said:

I don't know if the bot test is the best approach.

Maybe have the people who kicked other teams by synchronized casting of lag generating skills in New World should be called in to do the testing?

Bots don't intentionally try to bring the server down.

So in CU's case, they use Autonomous Remote Clients (ARC) which, as @Slapshot1188 mentioned, are "'headless clients' running on separate AWS instances". These ARCs are normal game clients just without keyboard, mouse, etc. controlled by a simple program.  Thus they are "indistinguishable to the server from players", especially when it comes to network traffic. In fact, ARCs "consume a bit more network traffic than players".

And, again in CU's case, these ARC-tests were designed to push the server to its limits.

Yet when FSR launched on Steam last year it had (still has) a player limit of what 10, or other ridiculously small number?

Good thing they never got more than 3 who ever tried it.

So far the public engine performance has yet to be seen.