Camelot Unchained’s latest battlefield test included nearly 3000 players and bots, surpassing Kickstarter promise

    
60

Camelot Unchained’s beta has gotten a bit wider this week with the addition of more backers into the testing phase. It’s still beta one; it’s just that beta two donors are now playing too. Make sense? Awesome. So what are they testing? The “massively multiplayer” part of the game’s warfare system.

“How do you define a massively multiplayer battle? One of our Backers asked on our Forums whether, based on what they saw today, we should use a new phrase instead of ‘Massively Multiplayer’ to define our battles,” City State says in its latest newsletter.

“That’s because of what our tests today showed as we continued our ongoing testing of the improved ability system. We had over 2,809 (which is why the date for this update is 2809!) ARCs (Autonomous Remote Clients) and players moving around an open battlefield using abilities, more often than we expect most players to do! Wrap your heads around that folks: 2,809 players and ARCs running around the screen in the same battle, in a tiny, by industry terms, space. And not 2,809 NPCs, but 2,809 connected clients, consuming bandwidth/servers like players and using the server-based physics system. That’s a heck of an achievement for the team, especially when you think about the fact that ‘500 people in the same battle, at 30FPS’ was the goal we stated during the Kickstarter. While there is still more work and performance improvement ahead, these numbers are much increased over our pre-Beta 1 numbers on the old system!”

Worth noting is the devs aren’t necessarily saying that 2809 people is ideal for fights in the game; the idea is that if the game can handle all of that in terms of rendering and networking, then the devs can layer in “NPCs, crazy siege engines, VFX, and more” and thereby “expand [CSE’s] vision for what a battle can look and feel like.”

“It is quite freeing in that way,” says the studio. “It also means that if we can do that, but not require the best CPU/GPU to play, more people can play our game.”

newest oldest most liked
Subscribe to:
Reader
Wrathbaby1993

What’s been created is impressive, or perhaps even the best thing so far ever made for getting this many people together in an MMO on the same screen and not expect it to be unplayable. Since there has been a lot of time and effort put into getting here, obviously this game is gonna be played mostly in a state of many people and Mass chaos. What are the plans for communication? I know it isn’t typical of an MMORPG to have local always on voice chat, but after playing an MMO called Foxhole I’ve realized that you can add significant value to any game, if you ease communication. It might not be much of a technical feat or current priority, but are there any plans or mantras or goals or ideas for CU on that front?

I currently play wow and my biggest complaint is playing for hours, and taking to no one. That’s certainly not a perfect parralel to this game, but it has grown noticeable, to the point I feel no reason to play.

Hamblepants
Reader
Hamblepants

I’d also love to know more about this.

Reader
Rolan Storm

This is great. I love it.

Random MMO fan
Reader
Random MMO fan

This is pretty cool that you managed your server to handle that many clients/players.

What about client side, Mark? Will there be any form of DX12 support? I’m saying this because I saw Matt Daniel’s post about user-created DX12 wrapper for GW2 which claims a pretty significant performance gains. Would be nice to see a proper support for it from developers themselves instead of relying on such user mods.

Reader
Mark Jacobs

Knowing Andrew, we will do what is best for the game’s Backers. He’s a brilliant engineer and he shares my concern(s) about always caring about our Backers first.

Reader
Arktouros

This is the only aspect towards this game I’m interested in. The #1 issue with most large scale combat games in the past has been absolutely atrocious performance when battle size scales up. I was there for WAR’s 800 man battle and it was a total crazy town of lag and server performance. Seeing all those characters out on a field at once is just awesome. Shame about the rest of the game.

Reader
Mark Jacobs

Sorry Ark, maybe the next game we make. As you know, I love PvE but it wasn’t the right fit. The good news is that when Camelot Unchained is successful, anything is possible. :)

Hamblepants
Reader
Hamblepants

Thought Ark was talking about the rest of WAR, not of CU…?

Reader
Arktouros

Oh no, I was definitely talking about CU.

WAR though was also largely disappointing as well. I loved the hell outta that game, I even bought myself a Griffon/Manticore. Yet I will never forget the way they didn’t fix a single issue for around 3 months and I watched our alliance of over 1000 people evaporate into about 25 of us as a result. Like how do delete your customer base in one easy step lol

Reader
Arktouros

I absolutely don’t want a PvE title. If I want PvE I’m going to go play a single player title where other players won’t invariably ruin my good time.

I’m 1000% about massive, large scale PvP titles. I’ve played just about every major open world PvP title since Ultima Online. The drought of PvP oriented titles over the last decade was soul crushing and it’s been fantastically exciting to see so many new titles being catered towards the competitive player again.

That said there are certain design decisions you’re going with I don’t care for. We’ll see what happens for development, things can always change, but not overly optimistic it’ll end up keeping my attention long term.

Reader
Mark Jacobs

Got it. I thought you meant that you played WAR for both the battles and the PvE and that you were disappointed in CU because it had no PvE but that you were excited by the size of the battles.

My mistake. :)

Reader
Loyal Patron
Patreon Donor
Armsbend

I like the term ‘when’. Confidence is always key.

Reader
Shadex De'Marr

I hope so because I will be honest that I was disappointed to see that PvE was not going to be a part of CU. I have been a dyed in the wool DAoC fan since launch but PvP is stressful and it was important to be able to leave the frontier and go back to your realm to relax and group for content with your friends. With CU that is not an option it seems so people will instead just log out when they have had their fill of PvP and go play something else. That is now a player not in game and potentially just staying and playing that other game from then on.

I know PvP games are really popular right now and I think CU will do well. I am just curious for how long. People say if they wanted a PvE game they would go play WoW or a single player game but I have also heard that if people wanted a PvP game they would just go play one of the thousand battle royals on the market. A lot of what I feel is CUs current draw are those of us that remember Mark’s amazing game DAoC. RvR was incredible but it wasn’t the only thing that made DAoC memorable or well made. I was one of the ones that past on backing CU when I found out PvE was removed. I truly do hope that after this you will entertain doing another DAoC. But by then it will be another 10 years down the road.

Hamblepants
Reader
Hamblepants

There’s PvE stuff to do in CU – crafting, building, trading, gathering.

You just can’t level a combatant class by doing it.

Reader
Shadex De'Marr

PvE is defined as player ‘versus’ environment. Having crafting, gathering, and building in a game does not mean it has PvE content. Would one classify Battle Royal titles as a PvP/PvE game because you gather materials and build things? CU decided not to go the route of PvE with CU and that was their choice. I personally just feel that it slighted a lot of original DAoC fans that enjoyed the game for more than just RvR.

Hamblepants
Reader
Hamblepants

No, building/gathering/crafting actually is PvE.

Player – you.

Vs. – Being challenged by, and trying to overcome ____.

Environment – the stuff around you in-game (anything that isn’t living human beings playing live through their computers.)

Yes crafting, gathering, building are all PvE, just maybe not the PvE you like.

Question: do you think gameplay requires direct player combat against NPC’s for it to be classified as PvE? Cause that’s not how I’d define it.

Also, CU is going to have direct player combat against NPC’s, this has been confirmed since the Kickstarter… the Depths, killing animals (possibly powerful ones) for their hides, killing guards (possibly powerful ones) for their hides (j/k to avoid being killed or to attack structures), and possibly also NPC factions that attack the player factions or defend certain territories.

You just don’t get to level up a combat class from it.

And yes, I do classify games like Fortnite as having PvE.

Edit: linking MrBTongue’s video on MMO’s cause when am I gonna get another relevant chance to do that: https://www.youtube.com/watch?v=nvK8fua6O64 . And yea it does eventually get around to talking about MMOs.

Reader
yoh_sl

Sounds like the network stress test and physics system is working rather well.
Seems that things are going rather smoothly all things considered.
Now if you can just stick the landing with the combat system as a whole, you should have a real winner on your hands. Thou that is still very much work in progress.

Thou that being said I honestly have a hard time picturing it in my head, the combat system that is. I really don’t know what to expect at this point. It kind of feels like I am looking at less then half a jigsaw puzzle, seemingly with a lot of pieces from several different puzzles all meshed together, and I’m looking at this wondering “what the hell am I even looking at?”.
I honestly just don’t know what the end produce is even meant to look like, so it’s kind of hard to tell how far you even are along, even if the gameplay isn’t the main priority right now.

I guess I can understand peoples frustration when there is no apparent end or milestone in sight. (probably by virtue of the fact they just don’t know what to look for) The work your doing right now isn’t exactly the most glamorous.

Hamblepants
Reader
Hamblepants

That’s something I’ve had trouble wrapping my head around too – just what exactly is an “hour in the life” of a CU player going to look like, what’s the rhythm and pacing going to be (if I’m understanding you correctly).

Reader
yoh_sl

I can generally picture what the game as a whole is going to look like, as that has been fairly well explained at this point. Even so, there are some major parts of the game that are complete mystery boxes, at least they are to me.
It could be a case of me just not paying enough attention because life, or just not understanding what to look for.

Thou personally I am willing to just sit and wait in the fog for it to clear, as it inevitably will in time. And as MJ said, their not exactly in the business of promoting the game at this point or making it look good, as they have their hands busy just building the damn thing.

At this point, my confusion is really more my problem then theirs. And that’s fine.

Hamblepants
Reader
Hamblepants

I hear that, but I don’t think it’s cause of a lack of understanding by you – probably just the “they’re busy building it” part.

Also, and I actually like this:

– they need to get the promised features in, connected, working, and fun before it’s clear what the entire game will play like – so any commitments to detailed descriptions of how it will play at this point would just be guesses, OR they’d be (potentially) committing themselves to something unbalanced/unfun without knowing it.

Reader
yoh_sl

While I understand that position, I don’t agree.
If history has taught us anything, building of a foundation on sand is inevitably going to lead to ruin, as was the case with Warhammer Online. While the game had a myriad of problems, chief among them was performance issues. If the game runs like dogshit, it doesn’t matter what it’s features are. And I think MJ understands that all too well.
I believe creating gameplay before your engine is ready to support it is very much putting the cart before the horse. Even if it is frustrating to watch from a players perspective.

Good things take time. And even thou sure I would like to have a better understanding of what the combat system is even going to look like in the end, I don’t think they should start working on it until their engine is bloody well capable of doing everything they need it to.

Hamblepants
Reader
Hamblepants

Not sure what you’re disagreeing with, but I do agree that making sure the engine can handle the gameplay is #1 priority.

Reader
yoh_sl

Well your previous comment made it out that they need to build out the skeleton of the game, and it basic features first. So that players would have an understanding of what the full product will look like.
I think that’s jumping the gun.

I think the foundation that you build your house on is by far in the way the most important element. Everything else will come in time and be better for it.
So long as they don’t run out of money midway, which doesn’t seem likely, it’s just a waiting game.
But I do understand why people are getting antsy.

Hamblepants
Reader
Hamblepants

I mean, having the basic features is a necessary first step to understand what the game plays like, I don’t really see a way around that – though goodness knows people seem to think there is.

Reader
Mark Jacobs

As you folks know, I can’t say the words “Build the foundation first” enough times to both my team and our Backers. That’s what these tests are showing, that the aspects of the foundation are in really good shape or else these kinds of tests would not be possible.

Next up, more work on the ability system, abilities and the player ability builder for the game. :)

Alyn
Reader
Alyn

Soon!

Reader
Loyal Patron
Patreon Donor
Kickstarter Donor
Dean Greenhoe

My interpretation of a test of 3k clients all in one small area:

This equates to visually realistic battle for 1000 players all firing off effects, siege effects, viable ammo from all sources, animals, smoke, weather, veil storm, structures, rubble, wagons, vox’s, npcs etc.

Bottom line, realism fora very large battle. A good start..

Reader
Mark Jacobs

You know Dean, I couldn’t have said it better myself. :) Thanks!

Reader
rafael12104

Wow. Nice! Congrats CU.

Reader
Mark Jacobs

Thanks, we’re getting there.

Reader
Siphaed

I get their proud of this, but maybe it’ll be more notable when the “NPCs, crazy siege engines, VFX, and more” are actually part of the game. Right now it looks like a very low rez open field battle from a game some 10+ years ago. At 23 FPS to boot.

Listen, their aim was “500” in an all out, large scale battle in a fully fleshed out MMORPG. And now they’re pushing 3k with bots…..as a networking sim or something? How many of those 2,809 were actual players? No, I’m not sold on the idea that their virtualized clients aren’t bots. Not unless they’re running 1:1 on single machines all connected to the server. But that isn’t the case.

It is still core building. Still in the engine mode. The engine that delayed the game significantly -including a combat reboot from scratch- yet still isn’t completed. Mechanics are bare bones or nonexistent. Items aren’t there. Classes -Crafting, in particular- aren’t there.

Other than making a field combat demo, what makes this a game? Let alone an MMORPG?

I’m sorry. It’s just my frustrations with this project grows as progression shows little to no gain in comparison to many other projects of the same stature. Furthermore the “game” doesn’t look to be releasing this side of Late 2020. [Perspective: The Kickstarter finished in May 2013.]

Hamblepants
Reader
Hamblepants

The bots are to stress their server, proxies, and to give us an idea of performance on our clients.

They say the bots are similar to real players in terms of all of the above.

If you think they’re wrong, I’d love to get more insight into why you think that. Genuinely not something I know much about.

Crafting is in game, as are items, just not a ton of them. Not sure how crafter class is gonna work, they still need to work on that for sure.

Point me to the 10+ year old game with 2500+ people networked together plz in a battle where they’re all within view distance and they are actually 3D rendered models with comparable graphics?? People keep mentioning this but nobody will actually point me to it :S . No, EVE doesn’t count lol.

Agreed that it won’t release before Late 2020 though.

Reader
Siphaed

https://www.eurogamer.net/articles/2015-01-26-planetside-2-gets-guinness-world-record-for-biggest-fps-battle

Game was developed over 10+ years ago, but released 7 years ago. And although not “2500+”, it was well over 1,000 REAL PLAYERS.

Hamblepants
Reader
Hamblepants

Ok, let’s set aside 2500+ for a min and just go with your 1,000 players.

Here’s what I said originally: “Point me to the 10+ year old game with 2500+ people networked together plz in a battle where they’re all within view distance and they are actually 3D rendered models with comparable graphics??”

Here’s a link to a video (apparently) of that battle you mention.

The biggest within-view-distance battle I can find is at the 1-hour mark, with ~50 people (based on counting pips of all colours on the minimap).

Couldn’t find a point in the battle with more than that (though I didn’t watch every minute of it, just skipped through to find biggest battles within the larger engagement that was taking place across the very large map).

If you’ve got some footage, pics of 1,000+ players all in the same small-to-medium-size space, and within view distance, please share!

Edit: pretty damn cool video, though!

Reader
Mark Jacobs

Yep, damn cool. And I also checked a number of other videos and I couldn’t find any videos either that showed hundreds, let alone 1K players fighting right next to each other or even within the space shown in our screenshots.

And until we hit about 2,400. Even when we yanked all the players and ARCs together, both the client and server worked well. At 1,800 after we yanked them, it was perfect. At 2,400 we had a brief pause as all the ARCs were trying to fire off their abilities at the exact same time since when we yank them, their ability timers reset and they sync up.

No disrespect meant to PS1 nor PS2, but what’s in the screenshots is a totally different challenge than getting 1K players in the same segmented and/or large map from both a rendering and networking angle as you know. If players are far enough away from each other, you don’t have to render them. Our render distance defaults at 4km which is why you can see not only castles, trees, etc., in the distance but players as well. In terms of networking, if you are a certain distance from other players, they don’t need to know about your location or them you, which reduces the network traffic since sending out packets to 200 people standing near each other is harder than sending out packets to 200 people spread out. It’s one of the reasons why even games like Dark Age of Camelot, Warhammer, ESO, etc., have issues with large-scale battles. Even if you don’t think that 1 ARC = 1 player, the fact is that all the connected clients need to know about all the other connected clients positions or you would be walking through all of them, they would pass in and out of view, etc. That didn’t happen in our test until things started breaking down because we couldn’t send out enough packets per server tick like when we tried to pull in 2,500 ARCs/players at the same time and made them stand next to each other at the same time. :)

The bottom line is that if networking 1000s of people in close combat was so easy, there would be more videos from games that have done it. The fact is that it simply is a real tough problem to solve due to both rendering and networking issues. And while we are late, our small team is well on its way to proving that we can make those kind of battles happen.

I’m heading home, have a great night!

Reader
Kickstarter Donor
JoeCreoterra

As someone who’s worked on these issues I wouldn’t really say “real tough problem”… I would actually say “mind numbing, head desk bashing, brutally hard problem” to be honest :)

The server interest management on that many players is extremely hard, I applaud you guys for really pushing this as I’m guessing you’re going not only based on distance but when crowding either going with scaled/lower tick rates and/or dynamic load balancing of the cluster… not to mention a ton of optimizations on actions/triggers. Extremely hard stuff that’s never really been a solved problem.

On top of that you have the rendering and scaled client prediction having to worry again about interest management… and also the LoD on models/textures/shaders, etc.

From what I’ve seen in your videos/updates your tech looks pretty epic :)

Reader
Mark Jacobs

Thanks Joe, yeah, we are trying to solve a number of rather tough challenges. We’re not there yet (we need to finish off networked building destruction of a block-based structure), but we’re getting pretty close.

The best part is that in some areas we have done a lot of optimizations but overall, not as many as people might think. We definitely have a lot of room for continued performance improvements on both the client and server side. And since we will have lots of graphic settings so players will always be able to have a good experience even on less-than-stellar PCs.

Thanks for the kind words, especially since you’ve been through the process yourself!

Reader
Mark Jacobs

1000 is not 2,809 and it also didn’t have true server-based physics . As anybody who was in our test can tell you, we didn’t have any performance problems till we hit about 2,500 and some of that was due to our continually yanking all the people/ARCs together in a truly tiny space, like right next to each other. :)

PS1 and PS2 are great achievements but as tough as 1K is, 2K is a lot harder than simply saying it’s twice as hard.

But as per below, we know we’re behind on gameplay but that is changing and will change faster now that we have more engineer time to devote to gameplay. And hopefully even more in the coming months.

Reader
Oleg Chebeneev

Arent you impressed with all the promises that devs throw at us while sitting on coach? How dare you?

Hamblepants
Reader
Hamblepants

edit: replied to wrong post.

Reader
Mark Jacobs

OC & Siphaed, ask the folks who were in the tests if they saw:

1) Lots of players (140ish) running around a tiny (by battleground standards) space. 140 at 60+ FPS (with vsync off) would be achievement enough but add 2000+ plus ARCs, that makes it a little more than just a small achievement.

2) Lots of abilities being fired off by both players and ARCs, at about 2,500 before things started to go south.

3) Any server crashes prior to us actually trying to crash the server.

And Siphaed, you’re wrong on a few points, the main one is the 10 year old game. Please show me one that that could handle 1K players in a battle at a high FPS. In our game, on a decent rig, you can get 60 FPS by turning off shadows and render distance. Can you even show me one today that can do that and at the same time, have a server-based physics system that prevents cheating?

Some of your and OC’s points regarding gameplay are fair ones which we’ve addressed before but not accepting the technical achievements that we have done as progress is both wrong and unfair. And mainly because if we didn’t do this now and focused on gameplay and then things went south on launch, I bet you guys would be saying “MJ, you should have known better from Warhammer! Didn’t you learn anything?” And you would have been right.

Yeah, I learned one lesson. That no matter how much people tell me to rush it before it is technically able to handle a load nor how many people tell me not to worry about it, we can fix it after launch, that I will ignore them and do what is right for all of our Backers, that is create a game that won’t fail at launch because the engine can’t handle it.

Call it Vaporware Unchained all you want OC, feel free to do it now if you feel the need to, because no other game in the world can do what we can do now and we are proving that every week to the people that come into the game.

Or and Siphaed, the reason our ARCS = players is that it doesn’t matter if the ARCs are sending their data from one machine or 10 machines. Each connection is separate just as it is with players. The only real difference is that our connectivity doesn’t test player-induced lag. But we have simulated that in the past and will be doing that again. It’s the same thing as somebody multiboxing a game from their house. All the connections from the house are going through the same pipe in their house, but they are all separate connections. Same thing for us.

Oh, and even if you gents don’t believe us on the fact that ARCs are the same as players for networking purposes, we still could render 2,809 of them and have some connectivity costs since they were all there. Rendering that many animated objects on a screen isn’t bad either though we have been up to 10K as well.

Reader
Bruno Brito

You prefer your turds floating in space, isn’t that right?

Hamblepants
Reader
Hamblepants

I mean, if they don’t speed up progress on the actual game itself, tests like these don’t really mean much.

So in that sense, I agree with Siphaed.

Gameplay-related stuff still seems to be added at a slow pace, and I’m pretty in-the-dark as to why.

I’d love for CSE to explain this, with specifics, cause I’m frustrated too.

Reader
Mark Jacobs

HP, that’s easy, because if you are using engineers to do certain things, they couldn’t do other things at the same time. As I’ve been saying for weeks, we’ve been focusing on the ability system’s performance which meant that we couldn’t build new abilities until we were sure that everything was working fine. We ended up taking up a lot of the time of our lead gameplay engineer, George, Andrew, and the two guys who were working on the system already. Now that we are confident about where we are, we’ve starting adding new abilities. We even moved one of the programmers to sit next to Ben to help speed up the process (this is why Max was ‘evicted’ from his sear which was next to Ben.

And I’m sorry HP, but this tests mean a lot regardless. Even if you thought I was wrong that we had to nail this now, the fact is we would have had to nail it at some point. So, even if this doesn’t speed up progress on gameplay (but it will), the fact is, as per above, if we didn’t test this now we would have had to test it later. And we all know what would have happened if something went wrong late in the process.

Also, I said last year that we would be do everything possible to make sure we didn’t have an issue with performance regression sneaking into the code and the way to do that was to have large-scale tests running constantly to confirm that. Well, once again I’m sticking to what we laid out for this game’s development and we hope, success.

CU in the game!

Hamblepants
Reader
Hamblepants

Ahhh ok, I misunderstood, and think I understand now – so the test was to see the effects/function of the current ability system at a large scale?

If so, then this test makes a lot more sense to me.

Thought it was just a random “check out how many bots we can have networked!” test.

Also gives me insight that the “lack of gameplay being added” assessment was wrong on my part, because the ability system is gameplay.

Either way, thanks for taking the time to explain.

Reader
Mark Jacobs

Yes HP, that’s it exactly! If you remember the update where I said that we were looking into the performance of the new ability system, this is how we do that. We can make any predictions we want, but until we are testing the systems LIVE, we can’t be 100% sure. So, we run tests like these and I didn’t like what I saw. That led to getting some additional folks involved and now we’re in a great position. The difference this time between say WAR’s development and CU is that we have the tools and nobody telling me that I shouldn’t worry about that stuff. :)

As far as your thanks, no worries and thank you of course for your support, patience, and kind words.

Hamblepants
Reader
Hamblepants

Sweet, glad to hear it, confusion led to frustration for me. Glad that’s cleared up.

Re: support and patience, happy to.

Also, while you’re here, thanks for giving Tim, Andrew, Tyler, Ben, Max and everyone free rein to just be themselves when on stream, be silly, joke around. It’s damn refreshing after seeing so many carefully-curated developer interactions.

I like that that’s how your studio is run, it’s genuinely one of the #1 reasons I’m happy to back this game.

Reader
Mark Jacobs

You’re welcome. As to the studio, we are as we appear to be. We don’t need fancy videos, gimmicks, etc., we just talk to you folks and if other people listen, great. We’re not trying to impress people at this point, we are just doing what we’re doing to make a great game and telling our Backers about our progress and/or failures. And I wouldn’t change a thing about our approach.

Hamblepants
Reader
Hamblepants

Was in this test, it was pretty bonkers, hadn’t been a part of an MMO-thing on this scale before.

I love Mount and Blade/Warband, so getting to be part of a test like this was pretty exciting for me.