Valve Shares Performance Numbers On Port of Left4Dead

New submitter nschubach writes in with an update on Valve's progress porting one of their games to GNU/Linux. From the article: "One factor in creating a good gaming experience is throughput. This post discusses some of what we've learned about the performance of our games running on Linux. ... After this work, Left 4 Dead 2 is running at 315 FPS on Linux. That the Linux version runs faster than the Windows version (270.6) seems a little counter-intuitive, given the greater amount of time we have spent on the Windows version. However, it does speak to the underlying efficiency of the kernel and OpenGL. Interestingly, in the process of working with hardware vendors we also sped up the OpenGL implementation on Windows. Left 4 Dead 2 is now running at 303.4 FPS with that configuration." nschubach adds "It seems there are good things coming out of this for both Operating Systems!"

Comments

Valve is big on Linux

[Anonymous Coward]

Just look at all the games available on Steam for Linux

Re:Valve is big on Linux

[Minwee]

% wine Steam.exe

Yeah, look at them all.

Re:Valve is big on Linux

[Anonymous Coward]

% wine Office.exe

Hey, look, MS has released Office for Linux! Let's give them a round of applause!

Re:Valve is big on Linux

[Minwee]

That's a great game. When did Valve release it?

Re:Valve is big on Linux

[pushing-robot]

I was wondering why the difficulty level increased with each version.

What does it tell you?

[Anonymous Coward]

What does it tell you when, to fully optimize something on Windows, you have to port it to Linux first?

Re:What does it tell you?

[FTWinston]

To be fair, it was their porting to to Open GL that improved the windows Open GL performance. What I find more interesting, to be honest, is that Open GL is (slightly) outperforming Direct 3D on a windows/nvidia box.

Re:What does it tell you?

[fuzzyfuzzyfungus]

Probably that, when dealing with GPU-limited things like framerates in a moderately intensive OpenGL application, a substantial portion of your performance is going to come down to the togetherness between your application and the GPU vendor's drivers, so working with said vendors might help...

Re:What does it tell you?

[synapse7]

Does the openGL driver/port have the same graphic detail as the directX version?

Re:What does it tell you?

[DrXym]

OpenGL sits over the top of DirectX on Windows so just using it would incur a penalty. I'm fairly certain that anyone porting a DirectX game to Linux using winelib or some commercial derivative would incur a penalty in the other direction.

Re:What does it tell you?

[Anonymous Coward]

OpenGL sits over the top of DirectX on Windows

No, it doesn't. That was the plan when they developed Vista, but it was scrapped after an outcry from half the industry. The OpenGL driver is just as low level as the DirectX driver on Windows.

Re:What does it tell you?

[kav2k]

[citation needed]

Re:What does it tell you?

[gman003]

Er, no, it actually doesn't.

Microsoft had planned to do so in Vista (they actually wanted to run old versions of DirectX on top of DX10, then the newest), but they scrapped that plan well before release after half the game industry, most of the professional graphics industry, and the graphics card companies themselves rose up in arms (Nvidia was actually planning on circumventing it, offering direct OpenGL).

So on any actually-released version of Windows, OpenGL is as low-level as DirectX.

Re:What does it tell you?

[FTWinston]

There's no mention of graphic detail in their article, so I'd presume (might be a big one) that they're running both at equivalent resolution, level of detail, and shader effects

Re:What does it tell you?

[kbonin]

This is not true - OpenGL is still a native driver. Microsoft wanted to cripple OpenGL in this way, but the CAD industry pushed back through Khronos and Microsoft agreed to retain native driver access.

Microsoft still hasn't updated the Windows OpenGL library bindings in something like a decade, so OpenGL developers all have to use the extension mechanism to access all of the new features.

Re:What does it tell you?

[Anonymous Coward]

OpenGL sits over the top of DirectX on Windows so just using it would incur a penalty. I'm fairly certain that anyone porting a DirectX game to Linux using winelib or some commercial derivative would incur a penalty in the other direction.

Games running on Wine/WineX/Cedega used to run faster than natively on Windows in a lot of cases, but I suspect that has to do with many functions not being implemented or just quick hacks to work rather than render correctly, not some magical performance increase while doing the same operations on the same hardware.

Back in the day, many games used to have a choice of rendering systems, D3D or OpenGL, (and or software even) and everyone knew different features were implemented on each. There's no reason to assume Valve's OpenGL port is feature complete.

Re:What does it tell you?

[jedidiah]

This is actually a mundane sort of thing in cross platform development. Your other "ports" aren't just a money pit. They allow you to stress your code in ways you might not have thought of. Typically the bugs you find in secondary platforms improve your product on it's primary platform.

Re:What does it tell you?

[hairyfeet]

Frankly it shouldn't be surprising, as OpenGL frankly isn't the focus on Windows so why should the Windows team optimize it? It doesn't gain anything for MS to have OpenGL work any better, DirectX works great and most cards frankly have a lot better DirectX than OpenGL support, and Windows is going to GPU accelerated everything while Linux can be stripped down like Win98 used to be so you can rebuild it and make a hell of a "bare metal" system for a specialized task...say gaming? Even if you don't strip it Linux distros don't usually have a bunch of GPU acceleration, nothing like Aero or Win 8 where every frame is GPU accelerated. Naturally using the GPU for more tasks is gonna lose a few frames for gaming especially when Joe Average don't know how to turn that stuff off.

So I don't know why anyone should be surprised at this, Linux has always been OpenGL, Windows DirectX, so having Linux do OpenGL better isn't surprising, anymore than Windows doing DirectX, its just what the platform is made for is all.

Re:What does it tell you?

[Ignacio]

So I don't know why anyone should be surprised at this, Linux has always been OpenGL, Windows DirectX, so having Linux do OpenGL better isn't surprising, anymore than Windows doing DirectX, its just what the platform is made for is all.

As a counterpoint, TFA said that their OpenGL port of L4D2 on Windows ran faster than the DirectX version.

Re:What does it tell you?

[FTWinston]

So I don't know why anyone should be surprised at this, Linux has always been OpenGL, Windows DirectX, so having Linux do OpenGL better isn't surprising, anymore than Windows doing DirectX, its just what the platform is made for is all.

Assuming I understand you correctly, I should perhaps have rephrased this:

What I find more interesting, to be honest, is that Open GL is (slightly) outperforming Direct 3D on a windows/nvidia box.

To:
What I find more interesting, to be honest, is that Open GL on a windows/nvidia box is (slightly) outperforming Direct 3D on the same box.

Re:What does it tell you?

[Joce640k]

Does the openGL driver/port have the same graphic detail as the directX version?

You think they don't know how to benchmark their own game?

Re:What does it tell you?

[Tawnos]

Clearly not. They give a bare number that doesn't indicate whether it is a maximum FPS or an average FPS. They provide neither the test setup (screen resolution, detail settings, etc), nor meaningful analysis of overall performance. For example, if the average FPS is lower on one platform, but the variability is also lower, the actual user-perceived performance will be better. No meaningful details are provided, just some ePeen number which is abstract of context. The statistician in me weeps.

Re:What does it tell you?

[Khyber]

"What I find more interesting, to be honest, is that Open GL is (slightly) outperforming Direct 3D on a windows/nvidia box."

Why would you find that interesting? It's been rather well-known OGL is superior to DX, just by the very nature of being extensible without having to wait for a new update to the OGL spec. If you want a new feature and the card has the power to do it, you can implement it directly into your engine and send the calls direct to the GPU.

Using DX, you're stuck with what you're given, and with the additional abstraction layers in DX, performance drops.

GOL has less overhead. No wonder it beats out DX, which is still partially CPU limited (moreso than OGL.)

Re:What does it tell you?

[BlackThorne_DK]

That only shows that they fail at communicating the results. Not that the result itself is bogus...

Re:What does it tell you?

[twocows]

The test setup is provided on their blog: http://blogs.valvesoftware.com/linux/

Re:What does it tell you?

[twocows]

Oh derp, you were referring to the options. And the blog was linked in TFA. Guess I'm the idiot.

Re:What does it tell you?

[HerculesMO]

It's outperforming DX9. DX11 is a different animal.

Re:What does it tell you?

[Wraithlyn]

It's not "clear" at all. Just because they didn't provide such details doesn't mean they don't exist. Absence of evidence is not evidence of absence.

Valve is an exceptionally competent developer of high performance, low-level graphics software. The article says "The data is generated from an internal test case". It's reasonable to assume this internal test case is specifically designed for, oh I dunno, TESTING. Which means it's designed to put the different rendering pipelines through the exact same gauntlet (or as close as can be managed).

Why would you think that them running the test at inconsistent resolution/detail settings is even a remote possibility? Do such obvious test practices REALLY need to be spelled out for you?

Yes, they reduced the test data to a simple number (I thought it was pretty obvious this represented "average" FPS). It's an easier discussion point. What of it? It's a blog post, not a scientific paper.

Re:What does it tell you?

[Requiem18th]

It isn't surprising for anyone who knows DX and OpenGL well, but it is surprising for game developers and other graphic intensive users since the "everybody knows" that graphics on Linux sucks.

Re:What does it tell you?

[SplashMyBandit]

Hardware OpenGL has always been faster than DirectX (there were a lot of companies and talented engineers pooling ideas for OpenGL, plus it has been around a while and worked out a lo of kinks). This shouldn't be a surprise. There was a hiatus where DirectX was ahead on features (we'll ignore the OpenGL extensions, and consider Core only). Now OpenGL has feature parity wih DirectX and works on a lot more platforms (which you could take as OpenGL actually having more features) - including OpenGL's superior support for Windows XP.

Re:What does it tell you?

[Narishma]

Yes, it's mentioned in one of the replies in the comments.

Re:What does it tell you?

[hairyfeet]

Again this is surprising...why exactly? Valve has ALWAYS been more OpenGL focused, going back to HL1 where you would get better visuals and framerate if you chose OpenGL over D3D so why is it surprising that a game made by them runs better on OpenGL than DirectX?

This is like saying "Sony games run better on the PS3 than the PC" which of course would get a bug DUH! from everyone because it should be obvious, same here. Valve and ID have always been more OpenGL than DirectX, been that way since the beginning, doubt its gonna change barring OpenGL going the way of Glide. I'm sure the Valve guys know OpenGL like the back of their hands, know all the tricks to squeeze performance out of it, so not a shock when that is what we see in the FPS.

Re:What does it tell you?

[VGPowerlord]

As a counterpoint, TFA said that their OpenGL port of L4D2 on Windows ran faster than the DirectX version.

Which makes me wonder where Valve is hiding the OpenGL for Windows version... as far as I'm aware, all of Valve's Source games for Windows require DirectX 9.0c.

Re:What does it tell you?

[Belial6]

The statistician may weep, but the writer would rejoice. A single number is the correct way to deliver information given the context of the statement. The information you are looking for would be appropriate in a full review. This statement was from a quick blog post.

Re:What does it tell you?

[Anonymous Coward]

32 GB of RAM, i don't know any others off the top of my head. read better articles

Re:What does it tell you?

[Tawnos]

Don't get me started on the "absence of evidence" line. It's a BS saying by those who want to support a particular claim without having any way to show others they're not just making stuff up. This isn't even absence of evidence, though. It's poorly-stated evidence, or possibly even misleadingly-stated evidence.

When a company that is "an exceptionally competent developer of high performance, low-level graphics software" fails to communicate test results in a fashion which can be meaningfully compared by people who would be interested in such a developer's blog, it either calls into question their competence or the content of their communication. A competent group that realizes "hey, that's really cool, we got a higher average FPS" and also sees "that's weird, there are a lot of sections of benchmarking where the actual FPS drops to 15-20", might choose to communicate the first result and hide the second in order to drum up support. Simply reporting the average FPS does nothing to assure us that the performance is better for an end-user (other considerations include micro-stutter, periods of low playability,

Additionally, such "obvious test practices" do really need to be spelled out, and conformity results reported, for a reader to even infer a meaningful comparison. With the system they described, if they were pumping this out to a 1920x1080 display, the test results mean something completely different than if they were driving 1, 2, or 3 2560x1600 displays. I (and others in the field or even just interested in the field), would like to know this information so that it is meaningful, and not merely an "ePeen number".

Perhaps it's due to [H]ard|OCP that I've come to expect better benchmark reporting, but what is shown here disappoints. A test case can be designed to test a number of factors, so simply stating it's for "testing" indicates you're unaware of testing in the graphics area. Those factors include, but are not limited to, conformance, pixel throughput, vertex throughput, fill rate, average fps, memory utilization, GPU utilization, CPU utilization, inter-frame jitter, and average dropped fps. Not all of these always need to be reported, but some of them are linked, and should be reported together.

So yeah, one factor is marginally faster and was useful for finding a previously-missed bit of overhead. Cool, great work, etc. But don't report a single number which may or may not be misleading (and will obviously be touted by some a place to claim platform or API set x is better than y), without giving meaningful context to the results. As you said, Wraithlyn, they're experts. How about they demonstrate that in a meaningful fashion for the other experts who might gain some insight from their blog?

Re:What does it tell you?

[Khyber]

I still can't believe they've not made a simple WINE/DX API wrapper built natively into Linux. I remember GLide wrappers that worked very well under Linux when 3Dfx support SUCKED.

Re:What does it tell you?

[Anonymous Coward]

This is not correct. The OpenGL backend was abandoned with Source engine, and removed completely, while Direct3D9 back-end was integrated more closely with the actual engine. This was more than 10 years ago, so given the changes in hardware and driver infrastructure, all relevant expertise at Valve is Direct3D-centric.

The OS X OpenGL port piggy backs off the Direct3D9 back-end and uses an emulation layer, a custom written library which presents a Direct3D9-like interface to the program while emitting OpenGL calls. This is evident from the backtraces of OpenGL related crashes of Source engine games on OS X which have been posted elsewhere. It's unclear what the scope of this library is - i suspect it encapsulates the state management and shader translation.

On the other hand, NVidia has always said: if your application is geometry centric, if it makes a lot of draw calls and is light on actual GPU load, you SHOULD use OpenGL because this will yuild you up to 30% more performance on NVidia driver stack. Thankfully, Microsoft has improved things for Direct3D starting with Vista, but not enough to close the gap.

-Siana Gearz,
Singularity Viewer

Year of...

[Anonymous Coward]

LINUX DESKTOP

Re:Year of...

[Anonymous Coward]

And my distro is the best!

Re:Year of...

[Yvan256]

Linux Desktop? Nope. Desktops are dying anyway, almost everyone has moved to laptops.

Year of the Linux game console, perhaps?

Re:Year of...

[Deorus]

Linux Desktop? Nope. Desktops are dying anyway, almost everyone has moved to laptops.

This is unfortunate. Ever since I became a nomad (and switched to Apple) that I miss actually shopping for desktop hardware. Every time I enter a retail store and look at the high-end video cards I really really want to build a desktop, but it can't fit my luggage... The desktop PC is far from being dead and I am already missing it, I think it's gonna be one of those things that I will remember from early 21st century just like I miss tinkering with analog electronics in the 80s (no, I'm not old, I was born in that decade).

Year of the Linux game console, perhaps?

Rumor has it that Valve is building a console with PC hardware, so I wouldn't rule out that possibility. They feel that the Windows and Mac App Stores represent a threat to Steam as a third party, so this may be part of their strategy to build a platform of their own. Blizzard has expressed similar feelings, which makes sense if we consider the rumor that they had and probably still have a third party service like Steam planned for battle.net (at least according to the leaked schedules which have been quite accurate, though battle.net third parties is overdue at this point).

Re:Year of...

[Provocateur]

My question for Valve is, why did it take them this long to port
games to Linux? Did their janitor remove his headphones that
one time and told them, hey I know this! And get on the phone
to contact his homies about this crew (yea Valve, toldja before)
that needed help boosting the FPS on nvidia, and bring some
sandwiches, yeah Hold the mayo. Make it two. And no ice in the
soda, remember what happens when it melts?

Re:Year of...

[TubeSteak]

There is a difference between your computer's desktop and a desktop computer.
Reports of the Linux Desktop's death have been greatly exaggerated.

Re:Year of...

[Ratchet]

I think they aren't so much building a game console as they are building a spec for a Linux based gaming PC for everyone to get behind. That makes more sense to me. I, quite literally, only use Windows to play games. Every thing else I either do on my phone, tablet, or already just as easily could do under Linux. If they can make it easy for us hardcore gamers to transition to Linux, then I doubt any of us would bother with Windows again.

The only issue is support for all this cutting edge hardware I have. Linux is always a problem there, but if gamers start to flock, I hope so too will the companies that make our gaming hardware.

Re:Year of...

[poetmatt]

I'm guessing it was solid influence plus just taking a long ass time? I'm not anyone with a view into valve though, but my thoughts are that:

Valve is so big now that if they go "we're going to Linux" they can bring it en masse. Back when valve was relatively new that wouldn't have exactly happened like that.

Re:Year of...

[_Shad0w_]

Valve projects run on the basis of "work on what you think brings value to our customers and to the company", so someone would have to have decided "I think a Linux port is a good idea, so I will start working on one" and encouraged other people to join the project. All of their desks are on wheels, so people working on stuff can move their desks together - they just unplug from where they are and plug in where they go.

Valve has a flattened hierarchy, there are no managers. Team leaders aren't appointed, they just happen because the naturally turn out to be the team leader. Remuneration is based on perceived worth by your peers. Their employee handbook is an interesting read - the PDF of it was publicly available, last time I checked.

Re:Year of...

[Anonymous Coward]

Thats why you buy yourself a Bitfenix prodigy. You'd probably be amazed to find that the latest top-end consumer CPUs and a 680 will work extremely nicely on a Mini-ITX board, which is not all that large.

Re:Year of...

[Anonymous Coward]

Anecdotal ignorance strikes again! THE PC IS DEAD. EVERYONE WANTS A TABLET blah blah blah blah. You should get an analyst job. You seem to have the required cluelessness.

Re:Year of...

[h4rr4r]

Gabe basically answered this already, Windows 8 is why.

Re:Year of...

[synapse7]

Linux is bringing the desktop back! MATE, the savior of the desktop.

Re:Year of...

[hairyfeet]

Yeah ya know, every time I hear someone say "Oh Gabe is doing this because the Windows appstore is a threat" I just have to ask...has Gabe never TRIED GFWL? Its fucking horrible! The damned thing can't even show you WINDOWS games when you are on Games for WINDOWS Live, nope those are at the bottom of the page under a dozen Xbox games! Its crash prone, slow as hell to log in, matchmaking sucks hairy balls, its a total POS!

The only "threat" the MS Appstore holds to Valve is the Cut The Rope, Angry Birds type of "gamers" that frankly wouldn't know WTF Steam was or that they even HAD those popcap style games in the first place! Mark my words, a year and a half from now old Gabe will still be laughing all the way to the bank while the MS Appstore will be added to Zune Market and GFWL in the "WTF were they thinking?" list.

If ya ask me I'd say its pretty fricking obvious what old Gabe is doing...he is thinking "Steambox" with COTS parts similar to XBox 1 and a stripped down Linux with Steam for all the apps. Makes sense, have a combo game/media center where the console devs can sell to PC and console users with a single port, the only catch I see is getting the blood sucking publishers like EA and Activision to go along. Lets face it, those corps never met a customer they didn't want to assrape and the second they see its for the living room they are gonna be rubbing their hands together and thinking about how hard they are gonna hit that wallet. One of the big selling points of Steam is how many good games you can buy there cheap, its literally "push button to get game" with prices lower than going out for pizza.

Personally I hope he pulls it off, I'd love a console where I can play my Steam games and just log into my account to have all my PC games right there to play with the boys, but i bet other than valve games the other publishers will screw the whole thing with their greed. Hey Valve, if you wanna do the whole Steambox, know what would be a GREAT launch title? How about Episode 3 huh? Just make sure we PC gamers can have it too okay?

Re:Year of...

[Orga]

" Desktops are dying anyway, almost everyone has moved to laptops."

as a gamer I went this route.. once.. bought a pricey gaming laptop... too hot, always had to change thermal paste, didn't keep up well with the games. I gave up and built out a new desktop.

As long as thermal issues remain in computing and vendors refuse to standardize and allow upgrading of laptop components desktops sized computing will not die.

Re:Year of...

[Anonymous Coward]

He mentions laptops and you talk about tablets? You need to improve your trolling skills.

Re:Year of...

[Deorus]

Thats why you buy yourself a Bitfenix prodigy. You'd probably be amazed to find that the latest top-end consumer CPUs and a 680 will work extremely nicely on a Mini-ITX board, which is not all that large.

It's not just the box, it's also all the peripherals. A laptop has everything built-in.

Re:Year of...

[PerfectionLost]

Games are the only reason I use Windows at home.

Re:Year of...

[Anonymous Coward]

Everyone is not your small circle of LAN buddies, kiddo. I won't get a laptop for any serious use just because I can never upgrade the parts that I would really want to.

Full desktop user right here.... unless you found a way to fit a full-sized nvidia or amd GPU into a laptop.

Sigh. Moron.

Re:Year of...

[KhabaLox]

My Debian brings all the beards to the yard
And they're like, it's better than yours
Damn right it's better than yours
I could fork it and I wouldn't even charge

Re:Year of...

[Anonymous Coward]

Parent post says "almost everyone". Your small circle of LAN buddies is the exception, not the majority.

Re:Year of...

[JazzLad]

Sigh. Moron.

Nice of you to sign your posts when posting anon, but I don't think you're a moron at all ;)

I love laptops. They're portable & easy to take everywhere. I even do most of my work on one. That said, when I go home to play games, there is no way in the hot place that I want to game on a laptop. I don't need portability in my gaming, but I do need a non-mobile graphics card, full-size keyboard and mouse and multiple large monitors. Of course I can plug these into a laptop (except the graphics card, of course), but what's the point? If I'm tethered to the rest of this, what benefit does the laptop give except to play at the homes of friends (sans monitors)? I'm no teenager (not for nearly 20 years now), so I don't really ever feel the need to play games elsewhere (and if I do, it's not like my rig can't be moved).

Nope, outside the very young set, I can't see gaming on laptops to be very popular.

Re:Year of...

[Anonymous Coward]

I think Gabe is more afraid of Microsoft trying to pull something like making their app store the only way to install ANY program once we get to Windows 9 (maybe 10) than he is of the service quality being comparable to/better than Steam. Which to be honest wouldn't surprise me (at times the Desktop in 8 when I tried it made me feel like it was only there for backwards compatibility).

Re:Year of...

[Anonymous Coward]

That's a very interesting way to run their business. I want to wheel my desk out side. Down the hill...

I have to say, Valve is one of the only companies out there, now that Blizzard has done their EA/Activision D3 scenario, that actually seems to be "good" in the sense of being very, very serious about delivering a good product for their customers. Seriously, it took them 11 years to make D3, and I'm tired of it within 2 weeks? As is everyone I know that bought it? RIP OFF, WANTS MY MONEY BACK. I wonder if Blizzard as a great developer has simply run it's course. Every dog has his day, and since they were acquired by Activision.. I fear for them. What's next, churning out a version of WOW or Diablo every year like the NHL franchise? SC/Diablo crossovers with super mario and link? Gah.

And if Valve can get every game on Steam to run on Linux, I will definately be drinking that kool-aid. The ONLY reason I still use Windows is because it is the best platform for gaming on the computer. And Steam is oh - so - tasty. Just picked up Transformers last night for like $8 - can hardly wait to tell my little guy, he's gonna be stoked!

Re:Year of...

[morari]

I love Linux. I've had some distribution or another set a second partition since the late 90s. It's great for basic usage and applications. Your average user will miss out on nothing by switching to Linux. But for me, it's not something I can use full time, try as I might. Games are a big part of it, but there are other things as well.

I'd say that the lack of Adobe's Creative Suite is a huge sticking point for me. I use just about every one of those programs on a near daily basis. It might work well enough in WINE, but that's not exactly the kind of experience most users are looking for anything. There are certainly no open-source alternative that stack up against Photoshop, After Effects, and Premiere. Hell, there aren't really any proprietary products that work as well as Adobe's suite does for most things.

We've seen the light of hope that is games before. I'm sure we all remember Loki Software. It was a great concept, seemed to have a ton of support, and the ideology was even spreading throughout some of the better developers, like Id Software. Then it went belly up, almost quicker than it hit the scene to begin with. I hope that's not the case now with Valve, as Linux has matured quite a bit. Still, the worry is there. They can either put the effort forth and force hardware developers to get in line, or they can fail. Really though, between a handful of Steam games and Minecraft, I would be covered in Linux. :P

Finding the solution to Adobe's absence is an altogether different problem though. They sell remarkably expensive software to professionals. They don't see the money in supporting Linux. The upfront costs would surely be high, and that isn't helped by the perceived unwillingness of Linux users to pay for any software, let alone pricey software. If they weren't in it for the money, they wouldn't be throwing out toy-like applications for tablets.

Re:Year of...

[gman003]

The only "threat" the MS Appstore holds to Valve is the Cut The Rope, Angry Birds type of "gamers" that frankly wouldn't know WTF Steam was or that they even HAD those popcap style games in the first place!

Just an FYI, Popcap themselves were actually one of the first third-parties to sell on Steam. I remember when early versions of Steam listed every free game on Steam in your library. At the time, that was what, Lost Coast and Codename: Freeman?

Then Popcap came in, and had demos for roughly ten million games[citation needed]. It rather irritated gamers, having to scroll through so many games they don't actually have or care for. They fixed that in a patch a few weeks later.

There are a lot of other "casual" games on Steam - check the "Family" genre. I actually ended up with several by buying several "everything this company made" packs.

Re:Year of...

[Anonymous Coward]

Which is why, contrary to the part you quoted, people are not moving away from desktops for serious gaming. Laptops are for farmville. Desktops are for just about everything else which requires an actual GPU.

Re:Year of...

[evilRhino]

They miss out on certain DRM content like Netflix and iTunes.

Re:Year of...

[Nemyst]

Two words: Internet Explorer.

Just as awful as GFWL and we're still trying to get rid of it a decade later.

Re:Year of...

[Narishma]

Haven't you heard? Consoles are dying too. Everybody is playing Facebook and mobile games.

Re:Year of...

[Anonymous Coward]

Well, that explains why HL2 Ep.3 is taking so damn long. Completely organic structures are just as problem ridden as super strict hierarchies.

Re:Year of...

[Anonymous Coward]

Dying is not an appropriate word. Desktops will be around a long time. I have six desktops in house and just bought my first laptop a few months ago. When I want a mobile computing device I'll use my mobile phone as all those other devices != convenient.

Re:Year of...

[vux984]

Desktops are dying anyway, almost everyone has moved to laptops.

As much as i dislike Steve Jobs, he got this right: "Desktops are like trucks".

Desktops are like trucks in a world without cars., and then introduce cars and see what happens to trucks.

Introduce cars (laptops) and what happens to trucks (desktops)?

A lot of people who used to buy trucks will now buy cars. But a lot of people will still buy trucks, because they need trucks. The market for trucks will shrink, but its not going to die.

I guess to carry the analogy further...tablets are like scooters or maybe those little smart cars. They aren't going to kill cars (laptops) or have any further effect at all on the market for desktops (trucks).

Re:Year of...

[arth1]

It's not just the box, it's also all the peripherals. A laptop has everything built-in.

Including a mouse, joystick, high end graphics card and a keyboard with thirty macro keys.
Nod, nod.

Re:Year of...

[Jaqenn]

With 'good-enough' integrated video currently shipping on current-gen Intel processors, I think we'll start seeing laptops that can dock onto an external high end video card.

When you're on the go you've got your small-time games and traditional laptop full of movies or term papers or whatever.

When you dock it at home you've now got the horsepower to play graphically intensive games without the full additional price or mental juggling of a second gaming PC.

Re:Year of...

[WorBlux]

TV didn't kill the radio, it just displaced some of it's contents. Desktops aren't dying, people just won't be using them as much or for as much. There are still use cases where having a real keyboard and mouse are vital.

Re:Year of...

[morari]

That is a fair point for those people that use such services. Personally, I can't think of any DRM-based distribution services that I use outside of Steam however. I only get discs from Netflix, and absolutely hate iTunes.

Re:Year of...

[Anonymous Coward]

I'm sure we all remember Loki Software. It was a great concept, seemed to have a ton of support, and the ideology was even spreading throughout some of the better developers, like Id Software. Then it went belly up, almost quicker than it hit the scene to begin with. I hope that's not the case now with Valve, as Linux has matured quite a bit.

Loki failed because it concentrated purely on Linux at a time when there wasn't enough demand for Linux games. I don't know if the demand is there now, but Valve certainly isn't going to abandon Windows anytime soon.

Re:Year of...

[Orga]

Sounds like a nice idea but I think you'll find along with that fancy video card you'll also need an additional power supply and still be limited to your laptop screen?

I don't see that happening, the price of this "dock" which holds a video card, power supply and ports is going to pricey. External video card would also require some very nice connector, thunderbolt or something like that.. and still quite a distance to travel from the cpu and system ram.

Re:Year of...

[morari]

I should have specified. I don't think that failing in Linux would drag Valve down. I do think there is a very real possibility of Valve's Linux venture not taking off however.

Interesting bit from the article

[gman003]

What I found interesting was how much an improvement this is from their initial port.

Their very first version ran at a full six frames per second (167ms/frame). They've now gotten it up to 315 fps (3.17ms/frame).

That's some pretty impressive work. Pity the article is so light on the details of how they did it (I'll spare you reading the article: they found places where it ran slow due to the kernel, they found places where it ran slow making OpenGL calls, and they found places in the driver itself that ran slowly - that's about as much detail as the actual article gives you).

Re:Interesting bit from the article

[kav2k]

What I find really interesting is the fact that this port spurred impovements in proprietary OpenGL drivers, in close collaboration with manufacturers.

This push by Valve may benefit everyone, even people who never will use Steam.

Re:Interesting bit from the article

[somersault]

The article gives about as much detail as you need really. The heap stuff sounds familiar, I think they mentioned that they were going to optimise it pretty early on. It is nice that they've achieved such speed gains though, especially on the driver side.

Re:Interesting bit from the article

[Anonymous Coward]

That's some pretty impressive work.

I too find 6fps in L4D to be very impressive. By comparison it takes Valve 3.14159fpd (frames per day) to Crysis. Very, very impressive!

Re:Interesting bit from the article

[gman003]

I would have liked much more detail. What shaders were the most problematic to optimize? What sections of the code perform differently (the netcode? the AI? the animation?)

Valve normally goes into a lot more detail. I remember their TF2 art presentation going on and on about shader algorithms - lots of equations, and pictures showing the effect of each term.

Re:Interesting bit from the article

[eddy]

The initial port probably ran on a MESA (software) backend. Hence 6 fps.

Re:Interesting bit from the article

[Ynot_82]

I'm sure lots more details are coming.

Valve are doing a big presentation on their Linux adventures next week at SIGGRAPH (6th - 9th Aug)

Re:Interesting bit from the article

[somersault]

Well, it is just a blog post - perhaps they'll do a more in depth presentation at a later point? :)

Re:Interesting bit from the article

[poetmatt]

Let's be careful with statements, please. They didn't say it ran slow due to the kernel or OpenGL, that is BS. They are acknowledging their own errors, not OpenGL's and not the kernels. They also did say exactly what they did to fix the situation.

I'll spare you on your details with the reality of what they said:

Their goals:
Performance improvements fall into several categories:
Modifying our game to work better with the kernel
Modifying our game to work better with OpenGL
Optimizing the graphics driver

The results:
An example of the first category would be changing our memory allocator to use more appropriate Linux functions. This was achieved by implementing the Source engine small block heap to work under Linux. The second category would include reducing overhead in calling OpenGL, and extending our renderer with new interfaces for better encapsulation of OpenGL and Direct3D.
The third category is especially interesting because it involves working with hardware manufacturers to identify issues in their drivers and, as a result, improving the public driver which benefits all games. Identifying driver stalls and adding multithreading support in the driver are two examples of changes that were the result of this teamwork. That's not a valve benefit, that's "all linux games" benefit.

Re:Interesting bit from the article

[adisakp]

What I found interesting was how much an improvement this is from their initial port.

Their very first version ran at a full six frames per second (167ms/frame). They've now gotten it up to 315 fps (3.17ms/frame).

That's some pretty impressive work.

That happens on nearly every Engine port. For example - Mortal Kombat on the Playstation VITA Handheld Console.

I worked on the team porting Unreal from PS3 / XBOX 360 to PS Vita at Netherrealm Studios (which we did in house separate from Epic's efforts). We ported over a NULL driver and then got the basic graphics up and running. Our initial port ran at 6 FPS. The shipped game ran at 60FPS with frame syncing and 80-90FPS at Speed-of-Light (frame syncing off).

You write a lot of code quickly to just get things working and once they are, you figure out the bottle necks and optimize code and assets from there.

Re:Interesting bit from the article

[adisakp]

Although in this case, I'm sure a number of the bottle necks were in the system and drivers which Valve apparently worked with hardware manufacturers to fix. That's a good thing for everyone.

Re:Interesting bit from the article

[gman003]

Sorry, I sort of glossed over that distinction.

The problem is in how the code works with the kernel (or GL driver). That can be fixed either by reworking how the code calls it, or reworking how the kernel (or driver) works internally. I referred to this, ambiguously, as "problems with the kernel", not "problems working with the kernel".

As far as kernel stuff, they seem to have done it entirely on their side. I imagine most of it was memory allocation - Linux's malloc() has much different performance characteristics than Window's malloc(), and that's 90% of your kernel calls right there.

The GL stuff they fixed in both places. In some, they were using it in a sub-optimal way. Sometimes they had to work with the driver team to get it fixed in the driver.

Re:Interesting bit from the article

[Yvanhoe]

Actually this is what the article suggests but in the comments, a member of the team explains that the only driver engineers they worked with were the intel guys in charge of their OSS driver. They had no answer from ATI or nVIDIA. It is in the 10 or 20 first comments. Check one with an answer.

Re:Interesting bit from the article

[SplashMyBandit]

Probably they changed from immediate moe rendering to Vertex Buffer Objects, Frame Buffer Objects, Pixel Buffer Objects and the like. By using more modern (as in the last decade years) OpenGL constructs you can get awesome performance, since the GPU and video RAM are doing nearly all the work (rather than transferring across from main RAM and have the GPU stall while waiting for the CPU).

Re:Interesting bit from the article

[Anonymous Coward]

Wrong, he said they only worked with Intel specifically on OSS drivers. They've been working with ATI and nVIDIA on PROPRIETARY drivers very much, and have made that very clear.

Re:Interesting bit from the article

[gman003]

Ah, that explains things. Valve does present fairly often at SIGGRAPH. If it's coming up, they may have written this short blog post as sort of a lead-in.

Re:Interesting bit from the article

[epiphani]

I imagine most of it was memory allocation - Linux's malloc() has much different performance characteristics than Window's malloc(), and that's 90% of your kernel calls right there.

*twitch* malloc() is not a kernel call, its a library function only. man 2 brk.

Re:Interesting bit from the article

[epiphani]

And, if memory related stuff is anywhere close to 10% of their system calls, they're doing it wrong.

Efficiency

[defender.tx]

Takeaway: the linux kernel and OpenGL are arguably more efficient than their Windows-based counterparts. I think a lot of people have thought this to be true for years, but it's always nice to see solid comparisons. It's a good time to have a linux box!

Re:Efficiency

[Urza9814]

Yea, I noticed several years ago, back when I used to play WoW and my computer could barely handle, that it would run faster in Linux on Wine with OpenGL than it would on Windows XP. I mean I'm talking ~5fps on windows to ~15fps with better graphics on Linux -- not really playable on Windows, barely playable on Linux.

Re:Efficiency

[Anonymous Coward]

I'm no Linux-user, but even I understand that things are going to run more efficiently on Linux. You don't run Windows for its speed and efficiency, no matter how much you try and optimize.

If Valve wasn't expecting a Linux build to run faster, have they really done enough research to justify their expansion plans? With this, it seems more like just a knee-jerk reaction to not liking either Windows 8 or the Surface brouhaha.

Re:Efficiency

[Anonymous Coward]

I wouldn't say so. It means, you can create more efficient code, when you can look at the underlying sources, instead of guessing the inner workings.

Re:Efficiency

[gman003]

They weren't expecting it to run more efficiently *given the level of work they invested*. They've sunk years of work into making the Windows version run quickly. Getting a Linux port to run faster, only months after getting their initial port running (the first running Linux version ran at 6 fps under the same test), is impressive.

Re:Efficiency

[Anonymous Coward]

To play a game that Windows users stopped playing two years ago?

How's the Linux port of Super Mario Bros.? Blazing?

Re:Efficiency

[Anonymous Coward]

Even guys at Windows know this, I've heard talk of getting rid of certain windows pieces such as DirectX and .NET

Re:Efficiency

[Xest]

You may well be right, I'd be cautious in taking anything away from this particular article though, as Valve have made it quite clear now that they have a vested interest in seeing Windows fall as they see Windows 8 as a genuine threat to their existence if people start buying games directly from the Windows Marketplace rather than Steam. When they referring to Windows 8 as a general "Catastrophe" which is probably a bit of a stretch, even if it maybe is for them, then it's hard to see them as objective on this issue.

Not that this is likely to be an unpopular move here, nor is it necessarily a bad thing if a company like Valve is helping Windows fall a peg or two, but now Valve has a clear political motivation to attacking windows, it's hard to see anything anti-Windows they mention as necessarily objective. It's also quite possible that Valve's Windows engine actually just isn't well optimised, and that now that they're moving it to another platform it's given them chance to rewrite components that were long overdue for a rewrite. I believe at least some of the foundations and design of the Source engine actually stem all the way back to the Quake 1 codebase for example.

Re:Efficiency

[arth1]

With this, it seems more like just a knee-jerk reaction to not liking either Windows 8 or the Surface brouhaha.

Well, that is what it is, and that is what is expected. Valve/Steam is in the business of desktop gaming. Windows 8 (and Gnome 3) will steer people away from desktop computers on to tablets and other devices primarily intended for consuming and not input heavy work.
If Gabe didn't react, I would be more worried.

Re:Efficiency

[Anonymous Coward]

If Valve wasn't expecting a Linux build to run faster, have they really done enough research to justify their expansion plans? With this, it seems more like just a knee-jerk reaction to not liking either Windows 8 or the Surface brouhaha.

According to MS and Valve, the Linux market is as large, or larger than OSX. Are you saying Apple's OSX is a knee-jerk reaction to Windows?

The real problem is, far too many people, especially on slashdot, have been barfing out known invalid numbers, and stale by half a decade, desktop numbers for Linux. Worse, when these were corrected, they were almost always troll moderated in an effort to censor, such that few learn, much more accurate numbers existed and have existed for a long time now.

The simple fact is, the year of the Linux desktop quietly came and went HALF A DECADE AGO. Linux on the desktop is a success*. The only things holding Linux back are games, tax software, and pre-loads. Valve is attempting to fix games. With browser based tax software, that largely addresses that. And companies like Dell is once again trying test pre-loads.

Its important to understand that Dell has always seen value in Linux pre-loads. The problem has always been Microsoft extorting Dell to prevent that. As such, Dells linux pre-loads have always horribly suffered at the hands of anti-competative Microsoft.

The simple fact is, anyone who says the Year of the Linux Desktop has not arrived is also saying Apple's OSX is a failure. You can't have it both ways. Either Apple has a desktop success, as has Linux, or they both are failures. Which is it?

* Part of the problem here is, far too many ignorantly believe that Linux's desktop success is defined as Microsoft's complete loss on the desktop. Such expectations is frankly, idiocy. Regardless, for whatever reason, that's what many want to believe.

Re:Efficiency

[gman003]

Much of the engine design dates back to the Quake 2 engine, but none of the code does.

History time!

First came Quake, which was coded completely from scratch. Then came Quake ][, which was a significant overhaul and massive rewrite. Some of the code was saved, but the engine design itself was changed quite a bit. Both of those engines were written by id Software.

Valve licensed the Quake ][ engine, and improved on it, adding a new renderer (a DirectX one, IIRC, but they kept the OpenGL and software renderers) and several other nice features. They used this for Half-Life, Team Fortress Classic, Counter-Strike 1.6, etc., and also offered it for relicensing under the name GoldSrc.

Over the years and years it took to make Half-Life 2, they rewrote literally the entire engine. Not a single line of code remains from Quake ][. They rewrote the renderer (several times), added all kinds of animation goodies, integrated Havok physics, and so on. But they kept the same basic client/local server/server design, the same general layout. It's much like how GNU made the basic Linux toolset - they copied the design of UNIX, but did not use any code from it.

Valve has continued to use and upgrade this engine, calling it Source to confuse everyone. They've offered it for license, and at one point were seen as a decent competitor, but they've really fallen behind in the post-UnrealEngine 3 world. I half-suspect they'll be either doing a total engine redesign, or giving up and licensing someone else's engine.

Re:Efficiency

[cyber-vandal]

Really? People will play stuff like WoW and Skyrim on tablets? Somehow I doubt it.

Re:Efficiency

[Anonymous Coward]

Hmm, sounds about right expect that Valve used the Quake 1 engine and heavily modified that.

Wikipedia to the rescue!

http://en.wikipedia.org/wiki/GoldSrc

And your not completely wrong, it does use a bunch of bug fixes from the Quake 2 engine. And its not a version of the Quake 1 engine but Quakeworld engine.

Atleast that is what I found in the english and dutch article, take their validity with a grain of salt.

Re:Efficiency

[Anonymous Coward]

It's been my experience as well that many games run at higher Fps under Linux plus wine versus windows, howeverI had assumed it was because many wine functions were just stubs not necessarily better graphics API plus kernel. I was pretty sue that the Linux kernel would be better as it seemed like Microsoft hadn't changed the nt kernel much in years.

Re:Efficiency

[jedidiah]

If Windows users aren't playing it anymore then it was a crap game to begin with. Good games stand the test of time and even get run in emulation 30 years later. Don't waste my time or money with dreck.

Plenty of people still play Super Mario knockoffs (or the real thing in emulation). Mario never went anywhere.

Re:Efficiency

[arth1]

Really? People will play stuff like WoW and Skyrim on tablets? Somehow I doubt it.

You missed the point, being that if people convert to tablets, there won't be any future Skyrim class games.

Re:Efficiency

[BenLeeImp]

I suspect they'd still exist, but just on consoles, which is sad.

Re:Efficiency

[MikeBabcock]

This has always been true in my experience. Quite a few Windows games running under Cedega were faster for me in Linux than on Windows with the same hardware. Native ports are an even bigger improvement; I found everything from NWN to Quake3Arena much faster on Linux than on Windows.

Re:Efficiency

[IpSo_]

I highly doubt they will license another engine given their recent Source Filmmaker work. Thats an impressive piece of technology and I have a hard time seeing them let it all go to waste.

Re:Efficiency

[Anonymous Coward]

Assuming they want to attack windows (and not just have microsoft and apple play nice), they might use the TRUE fact that linux is faster to attack them. It's a non sequitur to say: they attack ergo they cook numbers.

Re:Efficiency

[mellyra]

They aren't really porting the game for its own sake - they are porting the most recent version of their Source engine and L4D 2 is basically one huge glorified testcase.

Re:Efficiency

[Krneki]

I tried it too and it was faster on Linux, but .....most of the graphic effects were missing, so this might be the reason.

Re:Efficiency

[Urza9814]

Really? I never noticed any missing -- then again, I had the graphics settings as low as possible just to get the damn thing to run.

Re:Efficiency

[Anonymous Coward]

You're right, HL1 was based on the Quake engine, not Quake II, even though HL1 was released almost a year after Quake II. Some of us still remember this sort of thing!

Re:Efficiency

[Anonymous Coward]

Sound like a cpu bottle neck.

Re:Efficiency

[Anonymous Coward]

Good lord I wish people would stop forcing their own opinion onto everyone else.

I LIKE Gnome 3, and I'm sure there are hundreds of thousands, if not millions of people like me who actually like the improvements.

If you don't like it, then use LXDE, use XFCE, use KDE, use ratpoison for all we care, but stop trying to convince the world that no one likes Gnome 3, because obviously the developers do, and Fedora is still one of the most popular Linux distros, so it wasn't so bad that everyone jumpped ship for Mint or whatever else.

I've not met any person who REALLY hates the new Gnome except for the loud minority on the internet.

Re:Efficiency

[Anonymous Coward]

Err you're history time seems to be at odds with history.

"It is fundamentally just a heavily modified Quake 1 engine. There are about 50 lines of code from the Quake 2 engine, mostly bugs fixes to hard problems that Carmack found and fixed before we ran into them."

Pulled from http://en.wikipedia.org/wiki/GoldSrc

Re:Efficiency

[Blakey Rat]

They weren't expecting it to run more efficiently *given the level of work they invested*. They've sunk years of work into making the Windows version run quickly.

Yeah... but since 2007 or so they've done hardly anything with the engine (apart from building new content for it, of course.)

Getting a Linux port to run faster, only months after getting their initial port running (the first running Linux version ran at 6 fps under the same test), is impressive.

No doubt, but I wager that if the same effort were applied to Windows alone, the results would have been similarly impressive.

Re:Efficiency

[Anonymous Coward]

I had similar results years agao with EverQuest under Wine (actually, I was probably using Cedega at the time). Even though Wine was having to convert the DX path to OpenGL, After moving to Linux, I could finally walk through the Bazaar without staring at the ground.

Re:Efficiency

[steelfood]

That's because most optimizations happen at a higher level than the kernel and drivers. This is true for programming in general.

It doesn't take long to adapt the kernel and driver pieces to remove the bottlenecks--not if the original code was structured correctly anyway.

Re:Efficiency

[morgauxo]

Yeah, that was my experience too.

Re:Efficiency

[Anonymous Coward]

>I mean I'm talking ~5fps on windows to ~15fps with better graphics on Linux -- not really playable on Windows, barely playable on Linux.

It's probably because your OpenGL renderer didn't even try to put out the effects it didn't support, namely shaders.

Keep on drinking that kool-aid, though.

Re:Efficiency

[morgauxo]

When I compared running WoW using OpenGL on Linux vs Windows I didn't see any difference in content at all! Just a faster FPS in Linux. When I compared OpenGL on either to DirectX on Windows I did see it looked different. It seemed more "cartooney" to me in Direct X. I couldn't decide if either way looked objectively better or worse than the other, just different. I prefered the OpenGL look but that may just be because I was used to it. Later I found that WoW actually would run in DirectX mode on my copy of Wine. I didn't know DirectX was in there! I didn't run it enough to see if anything was missing. It was way too slow!

Re:Efficiency

[gman003]

You're talking "algorithm optimization". Using binary searches instead of looping through the whole array, for instance. I imagine those are all pretty cross-platform - O(n log n) will beat O(n^2) no matter what operating system you're on.

But games tend to care a lot more about performance, and so they do a lot lower-level optimizations than most applications bother with.

For instance, which is faster - rendering a triangle strip, or a triangle fan (ie. glBegin(GL_TRIANGLE_STRIP) or glBegin(GL_TRIANGLE_FAN))? It's entirely possible that one is faster on a different OpenGL implementation.

Or is it faster to run a DoF shader per-object, or run it as a post-process effect? Again, depends more than a bit on the GL implementation.

Kernel-level, does malloc() work best with 1kb chunks? 4kb chunks? Any size at all? Is it more efficient to just malloc() massive 128mb chunks and use internal functions to allocate and manage it? Depends, I suppose, on your malloc() implementation, which last I checked was different between Linux and Windows.

Re:Efficiency

[morgauxo]

Of course it's a knee jerk reaction to not liking something about Windows or Microsoft! I'd love to finally see 'the year of the Linux desktop' about as much as anyone but I can still be realistic. Almost all of their market either gave in and switched to Windows or always ran Windows in the first place years ago! While there are still plenty of us out there who would love to see Linux gaming take off most would still just keep Windows around if the games only ran on it.

Re:Efficiency

[VGPowerlord]

Yeah... but since 2007 or so they've done hardly anything with the engine (apart from building new content for it, of course.)

To be fair, they're testing this with a game released in 2009 that's using the 2008ish Left 4 Dead game engine.

I know I haven't kept track of what was added in the latest engines (Portal 2, DOTA2 Beta, CS:GO Beta)... although the only update Wikipedia mentions is the cloth simulation in DOTA2.

Re:Efficiency

[Anonymous Coward]

I could finally walk through the Bazaar without staring at the ground.

I remember doing that. Now I feel old and sad.

Re:Efficiency

[Nemyst]

Considering CS:GO is still using Source, I think they're mostly aiming to have a decent in-house engine that they're continuously upgrading with a focus on being as light as possible so that it runs on as many configurations as possible.

Remember that the name "Source" is a bit misleading, as what we're seeing is a continuum of engine versions on a host of different games. While Half-Life 2 ran on Source, HL2 Episode 2 also ran on Source despite being noticeably prettier, and L4D and TF2 also run on Source, but once again different versions of it.

Unlike Epic, id or Crytek, they don't feel the need to version their engine, but improvements are being made to it with each new game and update they put out.

Re:Efficiency

[PenisLands]

Heh heh. Mike BabCOCK. COCK.

Re:Efficiency

[Anonymous Coward]

Valve licensed the Quake 1 engine, not 2. All the big talk at the time was how Valve had added colored lighting.

Re:Efficiency

[Anonymous Coward]

Hater detected^^^^ You mad, bitch?

Re:Efficiency

[Anonymous Coward]

GoldSrc is derived directly from Quake's engine, not Quake II's. http://en.wikipedia.org/wiki/GoldSrc quotes Valve's Ken Birdwell saying it has only about 50 lines of code derived from Quake II.

Re:Efficiency

[arth1]

That's a rather explosive rant as a reaction to "(and Gnome3)".

Why so protective? And Anonymous? Are the two related? I can't imagine someone without a deep vested interest would react with this much outrage, to something that wasn't even a criticism.
It borders on scary.

Re:Efficiency

[Anonymous Coward]

They didn't license the Quake 2 engine.....it was actually the original Quake engine.

http://en.wikipedia.org/wiki/Half-Life_(video_game)#Development

And a followup note over on the Quake 2 wiki page:

"Valve Software's 1998 Half-Life, which went on to sell over eight million copies, was originally going to use the Quake II engine during early development stages. However, the final version runs on a heavily modified version of the Quake engine, GoldSrc, with a small amount of the Quake II code."

Re:Efficiency

[Anonymous Coward]

More correct history time! ;-)

From an ancient blog:

What came first, the chicken or the egg? What is Half-Life built on, Quake 1 or Quake 2? These questions pop up pretty frequently, and neither seems to have an accepted answer. In an effort to extinguish the argument, I've asked the people who know best. About Half-Life, that is. We're not touching the question about the chicken.

Ken Birdwell explains it like this:

"It is fundamentally just a heavily modified Quake 1 engine. There are about 50 lines of code from the Quake 2 engine, mostly bugs fixes to hard problems that Carmack found and fixed before we ran into them."

At its core, it's a Quake 1 engine. You can tell this by comparing Half-life's map compiling tools with those shipped with Quake1. You'll find very minor differences -- none of them are fundamental. The core rendering is architecturally identical to Quake1, the only "significant" change is removing the fixed palette, making map lighting RGB instead of 8 bit, and converting software rendering to be 16 bit color instead of 8 bit color, which was pretty easy and only required minor code changes. Our skeletal animation system is new, though it was heavily influenced by the existing model rendering code, as were a lot of our updated particle effects, though less so with our beam system. Decals are totally new, our audio system has some major additions to what already existed, and at ship time our networking was almost totally Quake1 / QuakeWorld networking but about a year later Yahn rewrote most of all of it to be very different in design. The most highly changed sections are the game logic; ours being written in C++ and Quake's being in written interpreted "Quake C". Our AI system is very very different from anything in Quake, and there's a lot of other significant architectural changes in the whole server and client implementations, though if you look hard enough you can find a few remnants of some nearly unmodified Quake1 era entities buried in places.

Jay Stelly adds, "We also took PAS from QW and/or Q2 and a couple of other minor routines I can remember (no more than 100-200 lines of code there). There was some feature overlap (as Ken mentions) like game code DLLs and colored lighting, but we developed our own solutions to those independent of Q2."

So there it is. This should put some arguments to rest. Half-Life is based on Quake 1, although it has a very small amount of Quake 2 code. Yahn notes that "we did use some of the winsock functions from Q2, that's about it. Probably more than 50 lines, but nothing too interesting."

So GoldSrc and thus Half-lIfe was actually based on Quake 1, not 2.

Re:Efficiency

[Anonymous Coward]

"Over the years and years it took to make Half-Life 2, they rewrote literally the entire engine. Not a single line of code remains from Quake ][."

This is true only because it was never based on Quake 2. It was based on Quake 1 as others have clearly pointed out.

Also, if you've ever seen the source (which many have, it was leaked remember?) you'll see that some sections are in fact line for line identical to the GPL'd Quake 1 code, despite your supposed rewrite where not a single line of code remains.

The entire foundation is largely the same, which does fit in with what the GP said about the source architecture being built on a near 20 year old foundation. Sure some key components have been re-written, such as the renderer and so forth, but the design and some things like resource management, initialisation etc. remain the same. These are things that would have to be rewritten for a Linux port, so could well be a factor in the speed up if rewritten to use a more modern design and techniques.

Your history lesson is only a history lesson if you're intending to completely rewrite history.

Linux supreme!

[Anonymous Coward]

Deal with it, menial M$ users...

What about quality?

[AwesomeMcgee]

I'm just curious; how does the visual quality stand up I wonder? I always thought OpenGL didn't have the same amount of features that Direct3D does.. Can someone here speak to that?

Re:What about quality?

[nschubach]

From what I understand, OpenGL has more features because they support extensions where Direct3D has to wait for Microsoft support. Granted, this requires that the developer actually use those extensions. The blog is rather light on details/screenshots.

Re:What about quality?

[Anonymous Coward]

From what I understand, they can do about the same, but some things are much easier in each.

Also, they both have version adoption problems:
DX9 is the standard, DX10 is common, and DX11 is problematically new for developers.
OpenGL 2.1 is the standard, OpenGL 3 is common, but OpenGL > 4 is problematic for developers.

Re:What about quality?

[91degrees]

In practice though, Direct3D will usually support the functionality as soon as its available. MS works pretty closely with the 3D graphics hardware companies since good D3D support is in everyone's interests.

Re:What about quality?

[nullchar]

640,000 shaders should be enough for everyone!

Re:What about quality?

[Anonymous Coward]

nschubach is correct. OpenGL has feature parity with DX for a very long time now. And as he said, there are numerous optional extensions which are only available for OpenGL.

That's one of the big advantages of OpenGL in that its far easier and faster to get newer advanced graphics capabilities into OpenGL. Whereas with DX, you're forced to wait on MS and because of their policy, frequently forced to wait on hardware vendors to catch up.

If you want a superior graphical experience, OpenGL is where its at. The only trick is to stop getting developers to accept MS' bribes and to start developing for OpenGL again. In the long run, its better for everyone, except MS.

Re:What about quality?

[defender.tx]

See for yourself. Download the Heaven DX11 Benchmark. Run it in DX11 API mode and then OpenGL mode and see the difference. There is a difference, but it's fairly minor. Usually DX11 or OpenGL will be the first to support a new feature, then the other supports it in the next release. The end result is that they normally produce a similar graphics quality.

Re:What about quality?

[Ironhandx]

With the latest iterations of OpenGL there is pretty much feature parity. Some of the bleeding edge tessellation and such deals don't have feature parity and bloom doesn't have as much impact but to be honest they look shit to me most of the time anyways.

Bloom can make otherwise totally photo-realistic games look cartoony. Which is great, if you're into that, I'm not.

However I think Valves involvement is even making those issues go away, since they're even working on OpenGL itself.

Re:What about quality?

[gman003]

They're basically the same at this point.

There have been times when either was ahead (early versions of Direct3D lacked many features, while some versions of Direct3D were a leap ahead of OpenGL). As of right now, they're pretty much tied for features.

That does not, however, mean that different versions of a game, using different renderers, look the same. But that's because the game uses different code to do it, and if they don't take care to make them look the same, they often will not.

Back when I was playing Half-Life, the OpenGL renderer looked much better on my computer (it seemed to use better texture filtering than DirectX). However, it somehow caused an audio problem when underwater - you couldn't hear anything. So for the levels with water in it, I generally switched over to DirectX just for a bit. It was most likely just because the renders stuck to default settings, not bothering to set a texture filter mode, and the drivers decided on different defaults for each.

Re:What about quality?

[Anonymous Coward]

Can someone here speak to that?

No more management-terms, please!

Can someone here talk about that?

Re:What about quality?

[LanMan04]

Just play it on a Mac (which uses OpenGL).

It looks great.

Re:What about quality?

[Anonymous Coward]

and runs at 10fps

Conserve Energy!

[Yvan256]

Save the Earth, use Linux!

...

You don't care about the environment? You know, that place you live in? The only one available with no alternatives?

Okay then.

Save your wallet, use Linux!

Re:Conserve Energy!

[Anonymous Coward]

It's more, "Max your shit either way, but FPS-bate about your imperceptible 3% gain in framerate on a three year old game"

I know this is super exciting in terms of Linux gaming to actually have games to play, but you have to understand this isn't exactly making Redmond quake in fear or any of the Windows gamers who stopped playing L4D2 two years ago feel bad.

Re:Conserve Energy!

[PremiumCarrion]

Uh, use of linux will surely lead to us running out of a natural resource which cannot be replenished. Frames.
By burning more frames per second we risk our childrens future, those poor bastards will have to view TV on e-ink at this rate.

Re:Conserve Energy!

[Anonymous Coward]

I'm a faggot

There, FTFY

Great

[Yvan256]

Can we bring those improvements to Mac OS X?

Re:Great

[Anonymous Coward]

The Mac games that use OpenGL.
The games that use Cider, which is a wrapper using a long-time fork of Wine? Nope. DirectX-to-OpenGL translation isn't going to get better.

Source games will get better, other devs wont care, and Sims 3 will never be more performant (cider).

Re:Great

[Anonymous Coward]

Why would they?

OS X doesn't have the user base windows has or the potential to increase it in any way, like Linux does.

Re:Great

[Anonymous Coward]

Good troll, considering Steam for Mac OS already exists. Idiot.

Re:Great

[Anonymous Coward]

even though they SHOULD NOT exist...

Re:Great

[Anonymous Coward]

Funny thing, the way it's going Valve will probably abandond OSX once they close it down completely.

Re:Great

[Anonymous Coward]

I think Nabe newell actually cares about their mac customer, but it is difficult for him too. Blame steve jobs and tim cook. They have a tendency of ignoring third party vendor when they request apple to fix some parts of their broken os or add graphic capabilities to their macs. Added to the fact that apple breaks something with every other release, I dont think developers will be too happy fixing something on apple's whim.

Re:Great

[Anonymous Coward]

The Mac games that use OpenGL.
The games that use Cider, which is a wrapper using a long-time fork of Wine? Nope. DirectX-to-OpenGL translation isn't going to get better.

Source games will get better, other devs wont care, and Sims 3 will never be more performant (cider).

And who honestly thinks Linux will see a better ratio of native to Cider/Cedega ports than OS X?

Re:Great

[Anonymous Coward]

And considering that the OS X user-base has been increasing faster than the Windows user-base since 2006!

Re:Great

[jedidiah]

Not terribly impressive considering that Microsoft has relegated Apple to single digit market share for about 20 years.

Re:Great

[Anonymous Coward]

I think Nabe newell actually cares about their mac customer,

Bull shit. If they gave a damn about their mac customers they would rewrite their piece of shit steam client.

Re:Great

[Ignacio]

Windows loves developers, OS X hates developers, Linux IS developers.

Re:Great

[Anonymous Coward]

Windows loves developers, OS X hates developers, Linux IS developers.

OS X hates developers so much that it gives it's development tools away for free.

Re:Great

[Anonymous Coward]

Sure. Right after Apple releases iTunes for Linux.

Re:Great

[Anonymous Coward]

Why would Valve care about iTunes for Linux?

Re:Great

[Anonymous Coward]

Rewrite after a few months? Are you completely stupid?

Re:Great

[morgauxo]

I wouldn't know about that but too bad they don' t do the same for iOS.

Re:Great

[Anonymous Coward]

The development tools for iOS and OS X and the exact same developer tools. Meaning both, iOS and OS X dev tools are given away for free.

Re:Great

[Nemyst]

So just like Windows and Linux (too many to link), basically?

Re:Great

[VGPowerlord]

Windows loves developers, OS X hates developers, Linux IS developers.

OS X hates developers so much that it gives it's development tools away for free.

...but only if you own a Mac. Which makes sense for OSX tools, but that also applies to iOS tools...

And are you implying that you can't get Microsoft's development tools for free?

Re:Great

[SplashMyBandit]

Great post. If only I hadn't blown 30 mod points in the last few days I would have given you +1 Insightful.

Re:Great

[Anonymous Coward]

Steam for Mac

Steam for Mac OS X was launched worldwide on May 12, 2010

2 years is more than enough time.

Re:Great

[Anonymous Coward]

And are you implying that you can't get Microsoft's development tools for free?

Nope I implying that the asinine assertion that Apple hates developers is false.

Re:Great

[Anonymous Coward]

Therefore Windows loves Linux? :-P

Re:Great

[Anonymous Coward]

Do you mean "Windows loves Linux"?

0_o. Is this /. or am I dreaming?

Re:Great

[arth1]

The development tools for iOS and OS X and the exact same developer tools. Meaning both, iOS and OS X dev tools are given away for free.

Free as in "but you have to buy overpriced hardware and OS from us before you can use it"?
Unlike the free (in both senses of the word) gcc not requiring a GNU machine with a GNU OS, and even the free (as in beer) Microsoft Visual Studio Express not requiring a Microsoft PC?

That kind of free? Ah, ok. Most of us call that kind of free "bloody expensive".

Re:Great

[Anonymous Coward]

Yeah, pick on the wording rather than the issue. Cute.

Re:Great

[Anonymous Coward]

Are you retarded? What issue is there?

Re:Great

[Anonymous Coward]

Only idiots rewrite software thinking it will change much. You already sucked at writing it, why would you suck less writing it a second time instead of simply modifying your working code?

Especially when that retard (or you) claims "2 years" is "more than enough time" for a rewrite, as if you should rewrite everything from scratch about every 2 years.

Even having to explain this is idiotic.

http://www.joelonsoftware.com/articles/fog0000000069.html

Re:Great

[Anonymous Coward]

You still need to buy the overpriced OS from Microsoft before you can use their free Visual Studio Express.

If I compare the price of a Mac mini, which includes the OS, to the price of a stand-alone Windows OS license, I'd rather buy the Mac mini.

Re:Great

[arth1]

If I compare the price of a Mac mini, which includes the OS, to the price of a stand-alone Windows OS license, I'd rather buy the Mac mini.

Mac Mini: "Now starting at $599" (source: Apple)
Windows 7 Home Premium license: $99.99 (source: Newegg)

PCs including Windows 7 can be had for as little as $300 retail, and something like this beats a mac mini hands down.

No, Apple is far from "free" with their "free" offer. Neither is Microsoft, but much cheaper.

gcc and other free offerings are, of course, free.

Re:Great

[Anonymous Coward]

Newegg only has OEM versions of Windows, the licensing is limited to computers for resale.

The real price is $200USD:
http://www.microsoftstore.com/store/msstore/en_US/pd/Windows-7-Home-Premium/productID.235488300/parentCategoryID.44066700/categoryID.50726100/list.true?Icid=WinCat_032012_Compgrid_boxshot1_Win7HP_PID_235488300&WT.mc_id=WOL_shop

If you buy a Windows 7 computer, you also cannot transfer the license to another computer.

And the link you listed for the $300USD computer doesn't work outside of the USA.

Ehhhhhh.

[Anonymous Coward]

Last I checked FPS over 60 generally doesn't mean much for most people. Very few gamers can even notice a difference between 60-100 let alone over 100.

So there is no real point in posting huge FPS numbers in Windows or Linux as it really doesn't matter.

Re:Ehhhhhh.

[Wain13001]

Huge FPS numbers actually are very useful for a variety of reasons. FPS is highly variable during gameplay, the higher the general FPS, the better your FPS will be when the game gets swamped with moving particles...which means it might not slow to a crawl at the most intense and exciting parts of the game...something that is particularly important in a multiplayer game like L4D.

Not to mention of course the fact that we are able to see a relatively accurate measurement of efficiency between two platforms, which of course is the article's point.

Re:Ehhhhhh.

[Diss Champ]

It seems to me that being faster than necessary on one piece of hardware means that, all else equal, it will be noticably better on some slower piece of hardware.

As long as the benchmark is not done with minimum system that is capable of running the game.

Re:Ehhhhhh.

[bmd256]

From a user standpoint those huge numbers don't really matter. From a developer standpoint it gives them clues on how efficient their code is. The faster the engine runs the better, and developers can start doing more with it (more features,detail, etc...)

Re:Ehhhhhh.

[Anonymous Coward]

Careful with this assumption. Framerate drops have multiple causes, and simply having a high peak FPS unfortunately does not at all guarantee a high minimum FPS.

Re:Ehhhhhh.

[sapphire wyvern]

Since almost all LCD displays only refresh at 60Hz, no matter how frequently the graphics card's framebuffer is being redrawn, the only thing a higher framerate would get you is the ability to turn off vsync and buffering without having any tearing issues. So you might get slightly better latencies maybe?

People with 120 Hz 3d LCDs could benefit from higher frame rates, though. But I don't think it's likely that _anyone_ is gaming on a display that's actually capable of showing >250 FPS. Maybe such displays exist, but I doubt there are many people using them for mere shooters!

Re:Ehhhhhh.

[rwven]

It's just a benchmark. It's not saying "These extra 200 fps are going to make your game sooo much better."

They're talking about the increased performance in opengl rendering in general that you get with Linux.

Re:Ehhhhhh.

[Minwee]

Very few gamers can even notice a difference between 60-100 let alone over 100.

They will, however, notice the difference when a sudden spike in activity causes the fps to drop to half or a quarter of that. Rendering speeds are highly dependant on what is being rendered.

A game running at 240fps can happily suffer a 75% drop without any noticeable impact, while a game which just barely makes 60 fps will not.

Re:Ehhhhhh.

[Anonymous Coward]

Valve is working on their framework in addition to the game. So while FPS rates for this specific game have diminishing playability returns, it will pay dividends for all their other games. Not to mention, its good business and bragging rights if they can boast a free platform which provides a superior gaming experience.

Re:Ehhhhhh.

[jedidiah]

Render rates far above 60fps means that I don't have to treat my machine like an xbox just because I decide to do some gaming with it.

It also means that I can likely get away with a much less powerful system and GPU and still have acceptable performance.

I can use less machine to get the job done.

Re:Ehhhhhh.

[Ignacio]

CRTs can do 250+Hz, but good luck finding one these days outside of a flea market.

Re:Ehhhhhh.

[morgauxo]

If you will pay shipping (or come pick them up) I'll give you a bunch of them!!!!

Re:Ehhhhhh.

[UnknownSoldier]

When you have TONS of explosions going off, TONS of lighting effects, you want the MINIMUM frame rate to stay ABOVE 60.

BY targeting such a high frame you have a nice SAFETY MARGIN for when the engine needs to start rendering additional (special) effects.

No offense, but you're an idiot.

Go play some BF:BC2 or BF3 on High Quality until you learn why 60+ fps is important.

Re:Ehhhhhh.

[nedlohs]

Because it's measuring how fast the engine is running. That it is passed "fast enough" is meaningless, since it isn't trying to measure anything related to playability.

Plus of course more headroom means you can have more stuff being rendered.

DOS

[Jimpqfly]

Could we get back to DOS 6.22 and benchmark it?

Re:DOS

[TheSkepticalOptimist]

Well the result would be 0 FPS because DOS couldn't even load the game menu, let alone the information to render a single frame of a modern HD game.

Re:DOS

[Hatta]

Sure, why not? HX Dos Extender supports Win32 binaries and OpenGL. Although it's only software rendering, so I don't think you'd get much in the way of FPS.

Episode 3

[Quakeulf]

Nice numbers, but where is Episode 3?

Re:Episode 3

[TheSkepticalOptimist]

If you haven't noticed, Valve stopped making games a while ago and instead became a content whoring system. One game every 5+ years doesn't make you a game developer IMHO.

Re:Episode 3

[Nadaka]

Its what blizzard did, of course I no longer consider them a game developer after diablo 3.

Re:Episode 3

[Quakeulf]

I know, it is just a rhetorical question. Episode 3 is the next vaporware now like Duke Nukem Forever, except this time the developer really DNF.

Re:Episode 3

[BenLeeImp]

http://en.wikipedia.org/wiki/Valve_Corporation#Games

They tend to release at least one game a year.

Re:Episode 3

[demonbug]

Its what blizzard did, of course I no longer consider them a game developer after diablo 3.

I'm not familiar with that game; did you mean Auction House Tycoon?

Re:Episode 3

[VGPowerlord]

If you haven't noticed, Valve stopped making games a while ago and instead became a content whoring system. One game every 5+ years doesn't make you a game developer IMHO.

They released a game last year (Portal 2) and have two more games in private beta right now (DOTA 2 and CS:GO).

Re:Episode 3

[UnknownSoldier]

> One game every 5+ years doesn't make you a game developer IMHO.

And your developer studio is where again with what shipping games??

Oh, you're the THAT guy that complains Quantity is better then Quality. Guess what Gabe said this this philosophy:

"No one will remember if a bad game shipped on time,
No one will remember if a good game shipped late."

Valve has the money (and time) to favor quality over quantity. You can bitch all you want but it won't change their core philosophy.

We developers knew this for a long time..

[Anonymous Coward]

One of the advantages of OpenGL vs DirectX is that it doesn't force the underlying hardware to comply as strictly in areas such as memory management, command batching, shader assembly, etc. This allows implementers more freedom to optimize and usually results in much higher performance. Even if a full backwards compatible OpenGL context is huge.

This approach was proven again very succesful with mobile hardware, where vendors such as Qualcomm, PowerVR or Tegra or ARM (Mali) produce graphics chips that comply with OpenGL but at the same time use the higher level abstaction of the API to their advantage, by supplying very different backends each (Immediate Rendering, Deferred and Tile Based Deferred) as means to improve performance (per watt and silicon space) to levels much higher than the desktop counterparts.

Added to that, programming games under Linux is a joy for those used to it, as the tools are fantastic (command line scripting, gdb with hardware watchpoints, valgrind, strace, etc) and the fact the OS manages the heavy load of games much better. Many companies I worked with, and even big ones such as Naughty Dog (makers of Uncharted) develop their games primarily under Linux, even if the final versions are released for Windows, Mac and Consoles.

Re:We developers knew this for a long time..

[Anonymous Coward]

and even big ones such as Naughty Dog (makers of Uncharted) develop their games primarily under Linux

Uhh their primary target is the ps3. That is a gcc based stack and the sdk is a linux box... So working in linux makes sense. Not because it is somehow better. If you worked with xbox guys you would probably see them working in the MS environments more. But wii and ps3 guys are using gcc based frameworks...

I know I am going to come off as a 'shill' but MS tools rock (I am not talking about their frameworks). It is the one thing that holds me to windows these days. All those tools you mention are available in windows and usually better polished. Valgrind compaired to say using boundschecker. You goto valgrind and bisect issues, boundschecker puts you right on the offending line that they think either overwrote memory or leaked. There are dozens of tools in windows like that. Valgrind is good for what it is (and better than nothing) but needs major work. If I am looking for memory/threading issues I usually port it to windows then debug it there. Most of the the IDE frameworks out there are clumsy wrappers around GDB. They have really improved in the past 5 years. But they still have a long way to go. Dont get me wrong. There is a tool for everything and many that are better than windows. But many are clumsy and tedious to use.

Re:We developers knew this for a long time..

[mellyra]

I know I am going to come off as a 'shill' but MS tools rock (I am not talking about their frameworks). It is the one thing that holds me to windows these days. All those tools you mention are available in windows and usually better polished. Valgrind compaired to say using boundschecker. You goto valgrind and bisect issues, boundschecker puts you right on the offending line that they think either overwrote memory or leaked. There are dozens of tools in windows like that. Valgrind is good for what it is (and better than nothing) but needs major work. If I am looking for memory/threading issues I usually port it to windows then debug it there. Most of the the IDE frameworks out there are clumsy wrappers around GDB. They have really improved in the past 5 years. But they still have a long way to go. Dont get me wrong. There is a tool for everything and many that are better than windows. But many are clumsy and tedious to use.

on a lossely related note - I really liked Shark on OS X when I did some development on that platform.

Re:We developers knew this for a long time..

[Anonymous Coward]

I have previously ported Linux programs over to Windows specifically so that I can use the superior debugging and profiling tools offered by Visual Studio. I am used to both, and frankly it is laughable to suggest that gdb is better.

Re:We developers knew this for a long time..

[TemporalBeing]

I know I am going to come off as a 'shill' but MS tools rock (I am not talking about their frameworks). It is the one thing that holds me to windows these days. All those tools you mention are available in windows and usually better polished. Valgrind compaired to say using boundschecker. You goto valgrind and bisect issues, boundschecker puts you right on the offending line that they think either overwrote memory or leaked. There are dozens of tools in windows like that. Valgrind is good for what it is (and better than nothing) but needs major work. If I am looking for memory/threading issues I usually port it to windows then debug it there. Most of the the IDE frameworks out there are clumsy wrappers around GDB. They have really improved in the past 5 years. But they still have a long way to go. Dont get me wrong. There is a tool for everything and many that are better than windows. But many are clumsy and tedious to use.

I've moved over to entirely Linux development after working in Windows for years. MSVS is quite well refined compared to nearly all the GUIs on Linux, and the MS Debuggers are certainly far better than GDB in many respects. So I very much agree with your sentiment.

That said, many of the times I do break points in GDB I have the same issues with MS Debugger - e.g. looking at STL objects under the debugger. GDB is doing a good job of catching up with their "pretty printers" API in that respect, which can make it just as good on the command-line as in the MSVS GUI. The GUI tools around GDB (DD, etc.) just aren't keeping up in that respect either.

Of course, GDB is designed to be more universal so that other tools can easily plug-in. Where MS Debugger is pretty much limited to just what MS puts out. The Qt Creator guys have done a good job of hooking into both; but they're one of the very few out there that do.

And finally, I'm finding some of the newer MSVS releases to become increasingly flashy while breaking down in functionality. (Yes, I still have to use them to deliver Windows-based versions of my Qt-based applications that I develop nearly solely under Linux.)

Re:We developers knew this for a long time..

[BitwizeGHC]

Naughty Dog I can see, as they are a Sony-exclusive developer.

I'd be interested to know who is developing Windows games on Linux. Everything I've read seems to suggest that most game developers consider Visual Studio an unparalleled toolkit (in particular with a debugger that's leagues ahead of Linux), and Direct3D a superior API in terms of performance, hardware feature support and ease of use.

Re:We developers knew this for a long time..

[SplashMyBandit]

No wonder you are a posting as an AC. IMHO, there are a lot better tools than Visual Studio on Linux and Unix (JVisualVM, DTrace/strace, various GL profilers).

Re:We developers knew this for a long time..

[SplashMyBandit]

Shame you aren' working in Java. The debugger is first rate in any of the IDEs (you have a choice) and JVisualVM is unparalled - you don' have to do anything special, just attach to any program running in your VM and away you go. Plus, with Java you can use JoGL for write one run anywhere (apart from a few AMD vs Nvidia gotchas) high-performance OpenGL programs, that's what I'm doing and it is awesome to use the same code between Mac, Linux and Windows without having to any porting effort (since the OpenJK, jogamp and jinput guys smooth over all of that away for you).

Re:We developers knew this for a long time..

[Clueless Moron]

I know I am going to come off as a 'shill' but MS tools rock (I am not talking about their frameworks). It is the one thing that holds me to windows these days. All those tools you mention are available in windows and usually better polished. Valgrind compaired to say using boundschecker. You goto valgrind and bisect issues, boundschecker puts you right on the offending line that they think either overwrote memory or leaked. .

You're going to come off as an MS shill because you are flat out lying. Valgrind tells you exactly what line is offending.

char *p = malloc(8);
p[10] = 42;

Let valgrind run that code and it will immediately tell you that the source filename and linenumber that p[10]=42; is on along with a callstack backtrace. Ditto with leaked memory, reads of uninitialized memory etc.

Re:We developers knew this for a long time..

[TemporalBeing]

Shame you aren' working in Java. The debugger is first rate in any of the IDEs (you have a choice) and JVisualVM is unparalled - you don' have to do anything special, just attach to any program running in your VM and away you go. Plus, with Java you can use JoGL for write one run anywhere (apart from a few AMD vs Nvidia gotchas) high-performance OpenGL programs, that's what I'm doing and it is awesome to use the same code between Mac, Linux and Windows without having to any porting effort (since the OpenJK, jogamp and jinput guys smooth over all of that away for you).

No, that's just an advantage of a byte-encoded language since you effectively have the source all the time as a result. And while Eclipse is in many ways superior to even MSVS, it also faulters in many ways as well - e.g. handling C/C++ projects is a PITA to get setup.

Re:We developers knew this for a long time..

[Anonymous Coward]

Oh so instead of IN the editor at the same time and it breaks right into the debugger? Or is it a separate step? Oh and do you have the right plugin's? Oh ding 'bad parameters passed in?', oh 'that line caused a memory overwrite', and 'thread race conditions'? Oh and a suggestion on what to fix? Oh and as you are debugging it? Oh I use these tools. Use both for *many* years. Like most linux tools they are 'almost there' but what I call 'scatter brained'.

Properly *configured* valgrind can do very well. But oh the configuration part... Most windows tools you are ready to rock after install. By default does valgrind do what you are talking about? No. Oh you need 4 different command line settings. Then it *starts* to produce useful information. Oh you need this other thing oh you run again and hopefully you have the right plugin.

Lying? I think you are being over protective. Its a free tool. Its good for what it does. But honestly (and this opinion is shared by many I talk to), they are a pain to use. So you end up bisecting.

This is typical of most linux debugging. It is done with debug logs. Which is good if you dont have the code in front of you. However, if you have the code and the editor and the tool there why are you debugging with logs?

Re:We developers knew this for a long time..

[Clueless Moron]

Oh so instead of IN the editor at the same time and it breaks right into the debugger? Or is it a separate step? Oh and do you have the right plugin's? Oh ding 'bad parameters passed in?', oh 'that line caused a memory overwrite', and 'thread race conditions'? Oh and a suggestion on what to fix? Oh and as you are debugging it? Oh I use these tools. Use both for *many* years. Like most linux tools they are 'almost there' but what I call 'scatter brained'.

Properly *configured* valgrind can do very well. But oh the configuration part... Most windows tools you are ready to rock after install. By default does valgrind do what you are talking about? No. Oh you need 4 different command line settings. Then it *starts* to produce useful information. Oh you need this other thing oh you run again and hopefully you have the right plugin.

By default (no arguments, out of the box) valgrind gives a description of the problem and a stack backtrace with the names of the procedures, files and linenumbers. You can tell it to also drop you into the debugger of your choice at the offending line.

The point is, your accusation "You goto valgrind and bisect issues" (exact quote) would make unfamiliar readers think that valgrind does not give the exact source code location where the errors occur and that, sir, is a flat out lie. I'm sorry that there are no weaker words for it.

Re:We developers knew this for a long time..

[SplashMyBandit]

> No, that's just an advantage of a byte-encoded language since you effectively have the source all the time as a result. And while Eclipse is in many ways superior to even MSVS, it also faulters in many ways as well - e.g. handling C/C++ projects is a PITA to get setup.

Fortunately I've moved on from C++ and only have to use it when interfacing hardware. Given a choice between developing in Java or C++ there is no contest anymore IMHO, unless you are on a severely memory constrained device (getting increasingly rare these days) Java wins for development speed, ease, cross-platform portability, tools, and breadth of modern libraries.

Re:We developers knew this for a long time..

[TemporalBeing]

> No, that's just an advantage of a byte-encoded language since you effectively have the source all the time as a result. And while Eclipse is in many ways superior to even MSVS, it also faulters in many ways as well - e.g. handling C/C++ projects is a PITA to get setup.

Fortunately I've moved on from C++ and only have to use it when interfacing hardware. Given a choice between developing in Java or C++ there is no contest anymore IMHO, unless you are on a severely memory constrained device (getting increasingly rare these days) Java wins for development speed, ease, cross-platform portability, tools, and breadth of modern libraries.

I'm glad to be in C++ still; and I would quite well beg to differ. But it'd be pointless to argue.

Re:We developers knew this for a long time..

[SplashMyBandit]

Well, it depends on what you are building doesn't it? and who is funding your development project (business-as-usual has a different pace than project work, in my experience).

Re:We developers knew this for a long time..

[Anonymous Coward]

Well, it depends on what you are building doesn't it? and who is funding your development project (business-as-usual has a different pace than project work, in my experience).

Well, I work on near real-time systems. Memory is not so much an issue, but performance is. And VM'd tech (even .NET's C# and C++/CLI) just don't cut it.

- TemporalBeing (not logged int)

Re:We developers knew this for a long time..

[SplashMyBandit]

> Well, I work on near real-time systems. Memory is not so much an issue, but performance is. And VM'd tech (even .NET's C# and C++/CLI) just don't cut it.

"real-time" can mean a lot of things. Meanwhile, Java has RTS (Real-Time Specification for Java) for both soft and hard real-time. Here's another starting point for you: http://en.wikipedia.org/wiki/Real_time_Java
Take a look, you'll see that Java can do hard realtime if you need it (especially since you are not memory constrained). That way you get the goodness of Java (eg. easy XML, expanding collections etc) while still being deterministic and meeting hard real-time requirements. Cool, eh?

Which versions?

[Anonymous Coward]

One key factor omitted is which version of OpenGL they are using. I certainly find the advantage highly interesting, but the comparison may unfortunately not be completely fair if they used OpenGL 4.x. Bear in mind the source engine only supports D3D9, which is 10 years old in a few months.

We developers knew this for a long time..

[goruka]

One of the advantages of OpenGL vs DirectX is that it doesn't force the underlying hardware to comply as strictly in areas such as memory management, command batching, shader assembly, etc. This allows implementers more freedom to optimize and usually results in much higher performance. Even if a full backwards compatible OpenGL context is huge.

This approach was proven again very succesful with mobile hardware, where vendors such as Qualcomm, PowerVR or Tegra or ARM (Mali) produce graphics chips that comply with OpenGL but at the same time use the higher level abstaction of the API to their advantage, by supplying very different backends each (Immediate Rendering, Deferred and Tile Based Deferred) as means to improve performance (per watt and silicon space) to levels much higher than the desktop counterparts.

Added to that, programming games under Linux is a joy for those used to it, as the tools are fantastic (command line scripting, gdb with hardware watchpoints, valgrind, strace, etc) and the fact the OS manages the heavy load of games much better. Many companies I worked with, and even big ones such as Naughty Dog (makers of Uncharted) develop their games primarily under Linux, even if the final versions are released for Windows, Mac and Consoles.

Whats the point of 300+ FPS

[TheSkepticalOptimist]

Rather then going for broke and getting as much FPS as possible, why don't game developers focus on optimizing the experience for a SOLID 60 FPS, that is, instead of peaking at 300 FPS in one scene, and then dropping to 45 FPS in another, strive for a constant frame rate.

If 60 fps does not tax a rendering system, then focus on MORE content in the scene, such as more particles or physics effects which enhance the gaming experience.

Maybe this is just some raw development figure, but there is absolutely no point in dumping as many frames as possible to a screen which only refreshes a given number of times per second.

I mean if someone came out with a car that outputs 300hp when driving 60 mph, I wouldn't be impressed, so why do I care how many frames are rendered between screen refreshes.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

Hard to say why they use a case like this. Max FPS is commonly ignored in gaming circles because its common knowledge that a system which manages a good top FPS, may not at all cope so well when the system is under heavy load (e.g. 30-60 fps). Not only that, but its also far more important. It doesn't, as you say, help anyone if a system can manage 150 FPS over 100 FPS in idle situations, but dumps down to 25 FPS rather than 30 when theres actual load.

In terms of experience, the second system with 30 FPS under load is superior. I dearly hope Valve's results are consistent.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

Older hardware that wasn't playable before now may be and better overall performance will give you avstable experience even on normally slow parts like many many particles appearing...

Re:Whats the point of 300+ FPS

[Anonymous Coward]

RAGE did this. Everyone hated it.

Re:Whats the point of 300+ FPS

[gman003]

This was 300fps on an *extremely* powerful system. The GeForce 680 is the most powerful single-GPU card on the market, and the 3930K is one of Intel's top consumer chips. I myself have a 660 and a 3610, weaker versions of the ones they used, and I can max out every game I have.

Getting 300fps on that means that, on a system a tenth as powerful, you get 30fps without dropping the graphics settings. Valve just doesn't chase the extreme high-end hardware - they don't bother adding more particles and such that make it look better only on a fraction of a percent of systems. Valve has perhaps the best knowledge of what real-world users are gaming on, thanks to the Steam Hardware Survey. So they can make an informed decision as to whether it is worth it to have the artists come in and add yet another layer of detail that will only be seen on the newest and most expensive computers.

That said, 120hz monitors seem to be rising in popularity, making rendering at 120fps a worthwhile goal. And it's often good to have a buffer of 10-20fps or so, because the amount of stuff you have to render isn't constant. In combat (with all the particle effects, explosions, flying debris, etc.), it often drops by 10%-20%, which can put you below 60fps if you're running at 70-80 normally.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

RAGE had the problem of being a console port, completely ignoring for example the far greater amounts of memory a computer has available. Or any configuration options whatsoever.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

You should consider the platform used for testing, which is really far from the average : the best breed of intel CPU coupled with the best of each GPU manufacturers, it's not what most people have. So yeah, 303+ fps for the elite PCs, and perhaps ~60 fps for us mere mortals.
Anyway, Valve is doing an awesome job. It will perhaps help other game companies making the leap to Linux development. I also hope that Valve will provide support to small studios wishing to follow that road as well.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

Buffers in games are a bad idea due to the inherent addition of input lag. Even the tiniest amount of input lag will often render a FPS unplayable, a buffer of 20 frames is completely suicidal. Some games may allow a buffer of 3 frames if explicitly enabled, but most will use at most two - if any.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

Because that's not how things work.
Your argument only makes sense if you know nothing *at all* about rendering and level design.

One does not simply add effects or polygons because there are "too many fps".
There are areas that have more detail and more things, and others have less. This is based on the story and experience you want to tell. Not on some pointless goal like "Screw making a good game! We must have X fps at all times ... AT ALL TIMES!"

The 300 fps are exactly so you have more headroom for the key scenes where everything goes crazy.
In the scenes where it doesn't, there's no point in artificially lowering the fps through additional detail.

If you want to limit your fps, use the fps limiter, like normal people. (If there isn't one, vsync does about the same thing. But beware, since if the game can render 59 fps, and you have set the vsync to 60 Hz, you will get only 30 fps because the game is always too late for the first time window and so has to wait for the second one... halving the fps.)

Re:Whats the point of 300+ FPS

[DeadCatX2]

You misread GP. When he said buffer, he means having an extra 10-20 FPS more than necessary. Headroom would have been a better word IMO.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

and I can max out every game I have.

That's because the games were designed to run on consoles too. But e.g. that top notch card has EIGHTY(80!) times the graphics performance of a XBox360.
If it doesn't run at 300 fps, then there's something seriously wrong with the game engine/drivers. ^^

Re:Whats the point of 300+ FPS

[TriezGamer]

Wrong kind of buffer. He's referring to a buffer as in a safe-zone -- A buffer against performance hiccups. If your frame rate drops below 60 FPS even briefly, any system with VSYNC enabled instantly drops to 30 FPS to prevent screen-tearing until the FPS goes above 60 again.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

Unfortunately, as said that safezone inherently results in a delay between game action / user input and displayed output.

Re:Whats the point of 300+ FPS

[Tallfeather]

I'm pretty sure the author wasn't actually referring to queued rendered frames, but rather "headroom" of rendering performance before the FPS dips to perceptible choppiness in times of high loading.

Re:Whats the point of 300+ FPS

[Ziggitz]

In the current example where they are comparing the same game on the same hardware with a different operating system and driver set, framerate comparisons are valid. If they were just touting a high framerate on a new release without any context yeah it means nothing, but when you are doing a side by side comparison with all variables the same except the ones you want to benchmark, it's a useful measurement.

Re:Whats the point of 300+ FPS

[epyT-R]

because, for some types of games, 60fps isn't enough. also, you can't have a constant framerate.. as the scene changes, the framerate will change, so most games are targeted somewhere above it and capped later.. quake4 was capped at 60fps and it bombed because of it (and other reasons too).. by the time they uncapped it, the community had already died.

Re:Whats the point of 300+ FPS

[VGPowerlord]

and I can max out every game I have.

That's because the games were designed to run on consoles too. But e.g. that top notch card has EIGHTY(80!) times the graphics performance of a XBox360.
If it doesn't run at 300 fps, then there's something seriously wrong with the game engine/drivers. ^^

Arkham City was technically designed for consoles, but on a PC with an i7 2600k CPU and dual nVidia 570s in an SLI configuration on a 1920x1200 display, enabling DX11 Tessellation in the Arkham City graphics configuration screen (everything else is maxed) results in an instant drop in framerate to around 40FPS.

Re:Whats the point of 300+ FPS

[blahplusplus]

"Rather then going for broke and getting as much FPS as possible, why don't game developers focus on optimizing the experience for a SOLID 60 FPS, that is"

You just missed the point, the high FPS is just an average, you want a minimum FPS the game never falls under given heavy load (lots of objects, effects onscreen).

IF you ever played Quake 3 back in the day it was all about having many things going on onscreen at the same time.

Re:Whats the point of 300+ FPS

[Narishma]

Rage locked the framerate at 60fps and automatically adjusted image quality on the fly to keep it smooth. Everybody hated it.

Re:Whats the point of 300+ FPS

[gman003]

I don't say this often, but you're a moron.

Buffer has several meanings, as most words in most languages do.

It could be a render buffer, what you insistently seem to think I'm saying. That's where it renders frames to a queue before displaying them. This buffer is desirable, of at least one frame, as it prevents the frame from being copied to the display while it is being updated, resulting in "tearing". Additional pre-rendered frames are mostly undesirable. However, this argument is pointless because THAT'S NOT WHAT I'M FUCKING TALKING ABOUT.

What I said, and what anyone with a three-digit IQ could have figured out from context, is that having an extra few milliseconds where the renderer doesn't actually have to do any work means that if the workload suddenly spikes, you maintain a sufficiently-high framerate. Say some big particle effect gets used somewhat-often, but not constantly, and it cuts your framerate by 20%, and let's also assume you're on a 60Hz display. If you run at only 60fps on average, you will normally be updating the screen on every cycle - 60fps. But when that particle effect comes on, it drops to 48fps, which means you start missing display refresh cycles. Not good. But if you have a "buffer", say 80fps, when that particle effect comes on you still render at 64fps, and thus never miss a refresh. It's a safety buffer. Like how, when driving, you don't drive 5cm behind the car in front of you - you drive several meters behind, so if the car in front slams on his brakes for some reason, you have time to do so as well before you slam into him.

Finally, I'll note that "buffer" can also mean "one, or something, that buffs", but that's not important.

Re:Whats the point of 300+ FPS

[Anonymous Coward]

they're trying to stress the system as much as possible. that's like saying, oh my code works fine when i give it a small problem. when i give it a real problem, all the bottlenecks are found. if there are no hardware bottlenecks, and the software still runs at 6 fps, then it's the drivers or the code that's the problem.

the only way to get an average 30/60/whatever frames per second is to bump up the max (you can cap it, but that's not going to improve your frame rate)

Why is this news? BOYCOTT

[Anonymous Coward]

The summary mentions 'Windows'. Therefore it has no place on /.

You're starting to slip!

Well if Linux is getting game support!

[Anonymous Coward]

The this means only one thing for me. Time to build out a full native Linux install again. I switched to Windows only and with Sun( yes sun not Oracle, well it Oracle now, but I miss Sun) Virtual Box Linux VM's both Servers and Desktops for my Linux needs. But, I can build out a new BOX with some new kick ass hardware and start tweaking out a desktop. Hmm, things are looking up for Linux on the desktop. Sweet!

Eric

Left4Dead is DirectX9

[Dunge]

DirectX10+11 improved performance a lot. This study don't talk about it.

Re:Left4Dead is DirectX9

[Anonymous Coward]

This is not a study, and it's not about OpenGL vs DirectX performance either. No who knows a shit about gamedev really cares if L4D2 would run at 340 fps on one API and 325 on another, okay?

OpenGL vs DX

[SebNukem]

I've noticed the different in speed of OpenGL versus DirectX in the past when the first versions of Google Earth came out. You could choose between OpenGL and DX to run the program. OpenGL was very smooth whereas DX's FPS could be counted on your fingers.

Irony

[oneandoneis2]

It'd be highly amusing to me if, in a few years time, Windows users are keeping a copy of Linux around because "I need it for the games" :)

Re:Irony

[foradoxium]

I hope so..

I'm trying to switch over to linux, but only if I can run lotro. I'm hoping eventually swtor will play in wine...but right now I'll be happy just playing lotro.

I really like windows 7...but I'd prefer to run linux even though I haven't run linux for my desktop in over 6-8 years.

Re:Irony

[Luminous Coward]

I'm trying to switch over to linux, but only if I can run lotro. I'm hoping eventually swtor will play in wine...but right now I'll be happy just playing lotro.

I was under the impression that "Lord of the Rings Online" worked well in Wine 1.4, is it not the case?
http://appdb.winehq.org/appview.php?iAppId=4891
What GPU do you use? and what distro are you running?

Re:Irony

[foradoxium]

I was able to get it to work, but I wouldn't call it "working well." I'm using an ati 6870, which under windows hardly ever blips below 50+ with high settings...in linux it goes from 15-30 with low settings. I was going to try it again, also using lotro's regular client, not the HD.

cheap jerseys

[ximeng]

Very happy to see your article, I very much to like and agree with your point of view. Thank you for sharing. At the same time,i love best pram very much .Welcome to look at my website and blog articles.Hope we can become good friends, and exchange and to help each other! Thanks!! It is a good blog,I like: http://www.nflmlbncaacheapjerseys.com/

Porting to multiple platforms is good for code.

[Anonymous Coward]

When you port to several systems you find, even if they have the same userland and syscalls, that your code becomes better and more robust by your efforts to get it to compile cleanly on all platforms.

No, you were right the first time.

[Anonymous Coward]

Which makes more sense?

      Yellow paint looks better than black on a house.

or?

        Yellow paint on a house looks better than black paint on a house.

The "on a house" part is orthogonal to the comparison of yellow and black paint. No need to be redundant.

Don't forget the FS

[phorm]

This depends on the operation before, but one of the big areas I noticed improvements seemed to be in loading times due to disk access. As it the same game on the same machine/disk (though different partitions), I'd put that down to filesystem efficiency.

Drivers

[phorm]

Was it a laptop?
I found that often where the windows driver hadn't been updated in awhile, the Linux driver was actually more fully featured at the time.
My old HP laptop ran some games (I believe it was Battlefield 2) that wouldn't work at all in windows due to driver issues.

Re:Drivers

[Urza9814]

Yup. My old Dell Vostro 1000 laptop. That could be it I suppose...

Silly question

[JDG1980]

Silly question from someone who doesn't keep up with cutting-edge 3D stuff: is there any reason for the industry not to standardize on OpenGL ES 2.0? That already seems to be the de facto standard on portable devices. What features does it lack that make it unsuitable for desktop and console gaming?

Re:Silly question

[Anonymous Coward]

Probably because OpenGL ES is a trimmed down version of OpenGL designed to be usable on weaker devices. Desktop PCs fall on the opposite end of the spectrum.

You're a fucking moron

[Anonymous Coward]

One of those people who says... whats the point of 32 GiB of RAM.... so short sighted. because 4GB will always be enough for EVER. It wont run 300FPS on YOUR computer. The point is to improve efficiency for all computers, even the low end crud you own so that you get 60+ FPS instead of 2