Quoting: PompesdeskyDid Feral note how often the word "Dirt3" comes up in the comments on this site ? Certainly is a sign that this community likes it a lot ^^

Yes, we note many things in these forums and other similar places :)

View this comment - View article - View full comments

Quoting: elbuglionethat's reason way you rock, guys...
I just finish Life is Strange and i have to say, you are the best game developers in the Linux world.

Thanks :D You're too kind.

View this comment - View article - View full comments

There is a massive difference between:

1. Linux install base
2. Linux users who game
3. Linux users who game and have have hardware suitable to your title
4. Linux users who game, have hardware suitable and like the genre/type of game you're making

Asking and seeing what the community like in forums certainly helps as it gives you an idea of how popular the game would be among a smaller community, also helps get a feel for how hard it might be to get the news out to the Linux gaming crowd. Word of mouth is powerful in smaller communities as often it's hard to get the word out about your games, the response to enquires also helps you judge how many people will be excited and post about your game if you do release.

For example Feral are fairly well known in Linux gaming circles yet I still see loads of posts like "Oh Warhammer is out on Linux I never knew" or "Feral they make Linux games? Oh that's interesting". For an indie developer/publisher I'm sure it's even harder to get enough of a word of mouth about releases given we still have people not know about our releases and some of them are fairly high profile.

Simply put it's not about begging or asking for metrics but getting a feel of the community by looking at how they react to your questions and posts compared to similar posts and questions in the same forums. This helps build up a picture that allows people to make a decision.

View this comment - View article - View full comments

Quoting: OLucasZanellaFeral Interactive, the lack of complaining means there is a texture bug in Shadow of Mordor that hasn't been corrected in maybe a year. To me, it's all a message: "do not take us lightly, for we too have claws"

For people reading about this and wondering about the issue I have a few bits of information. We've had a beta patch via support since just after this happened, unfortunately for various reasons this beta update is not yet on the main branch.

There is also a launch script alternative that you can do using the current release version if you prefer. Anyone who wants info on the beta just contact our support and we'll give you all the details.

View this comment - View article - View full comments

Quoting: boltronicsI assume the launch issues were fixed by sym-linking (or renaming) the libraries to match the names the game binary was using to to look for them (or alternatively by updating the game binary to use the existing library name), as I pointed out in the Steam forum thread. Honestly, I don't know how that was missed by their QA people.

These launch issues only existed on unsupported distros that have issues with the Steam runtime which then can impact games in different ways. We constantly tweak our scripts and options to make the game run more reliably and workaround these issues on unsupported systems to help users who don't use the officially supported distros however we can't reasonably test and fully support all distros especially rolling release distros (which are constantly changing).

Quoting: boltronicsWhat I'm really interested in is seeing them finally get to the bottom of why the game wants to download an update every single time I launch Steam. It's really annoying!

We've been looking into it and Valve are investigating the issue but we don't have a solution yet. It seems related to ownership of multiple DoW2 games but we haven't pinpointed the exact cause yet.

Quoting: boltronicsAlso, it seems one of my recent Mesa updates (which I compile myself) broke the game, so I'll need to look into that before I can play it again - and possibly do a regression test.

If you play on the bleeding edge of git you have to expect to get cut every once in a while. :) I'm sure it will get fixed quickly many of the most active Mesa developers have DoW2 keys so they should be able to easily repro issues :)

View this comment - View article - View full comments

Quoting: 0aTTHere is a good video:

https://www.youtube.com/watch?v=ozhXDhGozqA [External Link]

Even with a GTX 1070 the Game can't hold the 30 fps line. For me it's a riddle why all claim it would run with 60 fps on a GTX 970. Is there a video to prof this?

I can't believe this after what I know from ATTILA. ATTILA has on my GTX 970 system a similar performance as Warhammer in the above video. And what's really crazy: Whenever the fps rate breaks sharply, the GPU utilization breaks also. This also happens when I play at the lowest level. I have a i7 6700k CPU. With other games I never witnessed such behavior.

There must be a bottle neck somewhere. I love the Total War series and will buy Warhammer anyway. But it's a little bit frustrating to know that the performance will not increase even I update to a GTX 1070.

There is definitely something wrong with your setup, the main thing I can see immediately is you are using a driver with known performance issues. Everyone else on the similar card or older are getting double the fps (Liam's video is a good example of the performance you should expect and most users are experiencing). There are even videos of people playing triple screen with much higher settings than you are using!

Quoting: 0aTTI think it could be the asynchronous compute weakness of nvidia cards. I was wondering why Feral used an AMD card on there TWW live stream.

We used an AMD card to show off that due to the great progress of Mesa we could support Total War: WARHAMMER which I think was a pretty big deal especially for people who have AMD GPUs :)

Nvidia performs just fine as always, the only performance issues we have see are down to people using unsupported drivers or other hardware.

View this comment - View article - View full comments

Quoting: Guest

Quoting: edddeduckferalThe centralised server concept was popular in the late 90s but it has increasingly been replaced by peer to peer based multiplayer which is more robust and removes the overhead of maintaining dedicated server hardware.

I wasn’t necessarily thinking of dedicated servers; even in P2P game clients have to synchronize somehow or you can end up with two players believing they have each killed the other one :). But yes I haven’t followed the state of the art of network game programming recently…

Amusingly that example you used is what can happen when things go wrong.

I recall a couple of matches I played on a game a few years ago during development when that exact scenario would play out. As the games got more out of sync your opponents moves would get worse (as they didn't know what you were doing) and you'd start to win. Sadly same thing happened on both machines and you'd end up both winning and freaking the results code right up.

View this comment - View article - View full comments

Quoting: numasan

Quoting: edddeduckferalIn theory, however many WM's can have a large negative impact on game performance depending on the various settings and features it might use.

Really? Looking at these tests with AMD and Intel GPUs, I wouldn't call it a "large" impact, but who am I to argue with Feral:
AMD [External Link]
Intel [External Link]
I understand that Feral only wants to officially support one environment, though.

I would take the results of a single test like the ones above as one single example in a clean environment. It's not that a performance delta always happens (as those tests prove usually the difference is minor) however it *can* happen.

Usually things are fine but then you'll find someone has lost performance and in the end a WM or more usually a WM setting has been isolated as the cause of many weird edge cases. It's more a point of reference about an area that can cause issues not that is always causes issues.

In most cases WMs work just fine but if you have an issue checking your WM is something worthwhile (after you've checked your drivers).

Quoting: numasanRegarding the MP difference, it really is a bummer. Reading the CoH2 answer, I thought it was an engine specific issue? Will all ports suffer from this? I looked forward to this game, mostly because some of my friends have been pushing and recommended it when it came out (and I like the Warhammer fantasy universe), but now I have to tell them we can't play together, and they wont understand the technical reason :( They most likely have moved on to other games anyway, so in a way the MP issue doesn't matter anymore I guess.

In any case, thanks Feral for working on the port!

We've done loads of games with multiplayer and only a few can't play with Windows due to the reasons we posted on that link. So far these have been games by Relic (DoW2 / CoH2) and Creative Assembly (Total War franchise). We have many other games like the racing games from Codemasters and the XCOM series that support cross platform multiplayer.

Quoting: GuestI’m surprised that there are math issues for multiplayer. Isn’t there a single server responsible for making sure everybody is in sync, taking important decisions and sending back the new positions and events to all players? In my (limited) experience with multiplayer game development it is a requirement.

The centralised server concept was popular in the late 90s but it has increasingly been replaced by peer to peer based multiplayer which is more robust and removes the overhead of maintaining dedicated server hardware.

There are many reasons for choosing one over the other which you can check out but as with all things multiplayer code and protocols has gotten more complex over time compared to the good old days of hosting a quake server on a headless box in your cupboard. :-(

[edit] Typo fixes

View this comment - View article - View full comments

Quoting: Eike

Quoting: edddeduckferalThe games are released on Ubuntu, results on Ubuntu are accurate and are very similar to SteamOS but much easier to use and test. Using an unsupported distro and WM can sometimes cause issues you don't see on supported systems.

I'm expecting that, if no bugs occur, performance on different WMs shouldn't differ much. Do you have experience on this?

In theory, however many WM's can have a large negative impact on game performance depending on the various settings and features it might use.

View this comment - View article - View full comments

Quoting: renegat0x0So I see the benchmarks were executed on ubuntu. I am not sure if it is a good approach. The games are released on steamos. I do not know if it is hardware or software issue, but I see a big difference when running on mate and on openbox with compton. The latter solution is much closer to steamos. I have nvidia card 950 and am running linux mint 18 mate.

The games are released on Ubuntu, results on Ubuntu are accurate and are very similar to SteamOS but much easier to use and test. Using an unsupported distro and WM can sometimes cause issues you don't see on supported systems.

View this comment - View article - View full comments