Beta BeOS R5 OpenGL Benchmarks Smoke Linux and Win

BigZaphod writes: "BeNews has a story up that compares GLQuake and Quake2(GL) on BeOS R5 to Windows and Linux (Corel). They compare stats using a Voodoo2, Voodoo3, and a Matrox G200. BeOS wins almost every round -- sometimes by huge margins." Update: 06/19 09:06 by CT : several people pointed out that they did the tests under XF86 3.3 which of course is not an even remotely fast 3D platform. Restesting at the very least under XF86 4 w/ DRI would be necessary to get a fair comparison.

Please Moderate Down

The above poster is either aggressively misinformed, or is deliberately attempting to misinform.

We do not drop triangles. If we do, it is a bug and we'd like to know about it.

Unlike Micros~1, we are not interested in cheating and fudging benchmarks and issuing misleading press releases and pre-announcing non-existent products. We are trying to beat the crap out of Windoze by simply being better than Windoze. With the amazing OpenGL work Jason has done, we're getting closer...

Schwab
Be, Inc.
Currently working on Intel 810 OpenGL driver

Re:BE's lack of stability

Have you ever considered that it is just your hardware? Video works great for loads of other people, and even for me (who does graphics programming and BDirectWindow stuff, apps which tend to crash 50,000 times during their development) R5 hasn't crashed yet.

solution: don't use shit hardware.

Sorry, but people that complain about lack of support for a given OS are usually using the cheapest, nastiest hardware available. No-name NE2K clones, piece-of-garbage motherboards, etc. The money and time invested by Be (or even the Linux developers) to bring support for Bob's Crap Sound Card Model XYZ to the OS probably won't offset the money they'd recoup from increased sales.

Buy good hardware and be happy when everything just works.

- A.P.
--

"One World, one Web, one Program" - Microsoft promotional ad

Not benchmarketing.

Did you guys even read the whole article? How, exactly, is this benchmarketing? The guys put up a test between 3 OSs running an industry standard 3D test, namely Quake. Everyone and their mother has used Quake to test 3D cards and OpenGL. Hell, 3DLinux used Quake to test the nVidia Linux drivers. The tests showed that BeOS was faster in most cases, especially windowed operations. Is it benchmarketing simply because Linux didn't win? BeOS has shown this kind of amazing performance in other fields too, like audio latency around 2 or 3 millseconds opposed to around 50ms in Windows. Is there something wrong with the testing method? True, the machine used was SMP, but if you read the whole article, the BeNews guys said that SMP only lowered the score by about .6fps. Okay, so now it is 72.6 to 77.8 instead of 78.4! It isn't like this is a synthetic benchmark, it was Quake. If Be did something to cheat Quake benchmarks, then good for them. That means that any game that is Quake-like in terms of 3D use (ie. all of them) will run faster. If you're complaining about the narrow scope of the test, remember that Voodoo2's are the most stable cards available on BeOS, and that QuakeII is the only major 3D application. So what's the beef? A lot of people have tossed around the term benchmarketing, but if anyone can give me a specific example, then I'd be glad to listen.

Flawed Benchmark!

Did you notice how the test systems were Dual P3s and the tests were run against Win98 and Linux? the Linux ports of the Quakes are not SMP-optimized, and Win98 can't use the other processor at all. A bit unfair? Definitely!

it is cool that there is no frame loss when switching to windowed mode...

youth

Stuff like this makes me wonder how much of Be's youth has to do with it.

Younger, leaner code, et al ... anyone have any similar thoughts?

I do remember using a preview copy of Be 1 and trying out that terrain demo on a PowerPC. Seeing a mac render so many polygons at one time at full 35fps almost gave me a heart attack. I dont know the details of why Be is so good on the graphics side, but it looks like its still the same way ..

My audio playback comparison of BeOS, Mac and Win

I didn't try Linux when I wrote this comparison of the audio performance of MacOS, Windows and BeOS, but I don't think Linux would have fared much better than Mac or Windows:

The Battle of the Bands [goingware.com]

I was able to play nine uncompressed CD quality audio files simultaneously and independently vary the volume on each on the BeOS. I was never able to play more than one on the other operating systems I tried. I could play up to two tracks simultaneously from an ISO 9660 CD before the seek time of the head broke up the sound.

I've found in the 2.4.0-test1-acX kernels that audio streams will just plain stop until I click on or drag around an XWindow. Posts I've seen on the Linux kernel mailing list suggest that X is failing to yield the PCI bus at times.

I'm not suggesting though that you should all go use BeOS for your sound. What I do suggest is improving the multimedia architecture of Linux until it can match the BeOS.

Tilting at Windmills for a Better Tomorrow.

Flawed Benchmark, seemingly

Although I agree that BeOS has a screaming advantage over Windows or Linux, it's the Windows/Linux difference that disturbs me.
Seeing as they used Corel Linux, they probably did not bother getting any real hardware state-of-the-art Linux drivers. The reason for this is not just Linux-trolling!
The Linux Games article [linuxgames.com] compared Linux and Windows performance, and got much better results for Windows vs. Linux.

Every benchmark is flawed/biased...

It doesn't matter what you're testing. There's no cross-platform benchmark that will be accepted by EVERYONE. The loser will invariably (and sometimes justifiably, sometimes not) whine about how the benchmark is unfair, or unoptomised or whatever.

Witness:

ByteMark is biased in favor of PPC
SpecMark is biased in favor of intel
Unreal FPS is biased in favor of Glide
Quake FPS is biased in favor of OpenGL
Office benchmarks are biased in favor of windoze
Photoshop benchmarks are biased in favor of MacOS

Mindcraft benchmarks are.... well, I won't even go there.

But you get the point?

All of the above have been used, at one point or another (again, sometimes justifiably, sometimes not), by the side that came out behind in an attempt to invalidate the test that showed their OS / CPU / video card / FPS game / etc. to be inferior.

john
Resistance is NOT futile!!!

Haiku:
I am not a drone.
Remove the collective if

An un-NetBSD-related comment

Posted by 11223:

Because the site is slashdotted, I'll tell you about the other neat issue with BeOS performance. BeTips [betips.net] has been running under BONE (the BeOS Networking Environment) for the past few days, and has seen great performance increases in networking. This machine also serves as Scot Hacker's personal computer and is used for playing MP3's, rendering personalStudio videos, etc. The performance of BONE is in the same league as Linux and NetBSD now, and it's design was modeled on the excellent BSD TCP/IP stack.

Quake benchmark, not OpenGL benchmark

Just a nit-pick:

If they are only comparing Quake speeds, you can hardly call it a comparison of OpenGL implementations. Quake and Quake-based games use a relatively TINY portion of the OpenGL API.

If run at high resolutions, the tests are basically testing the hardware's fill rate, and at lower resolutions, they are testing the drivers' geometry rates--definitely nowhere near a comprehensive "OpenGL benchmark".

Perhaps the headline, "Quake faster on BeOS" would be more appropriate...

My thoughts on BeOS

On a whim, I downloaded the personal edition of BeOS over this last weekend. Mostly on a whim - I'd heard about it but had never had the time to give it a runthrough before. My initial impression of it from about 3 or 4 hours of use is this: BeOS is everything that Windows 98 should be and that Linux, regardless of what you might think, doesn't want to be. By that, I mean that it is suitable for the 99% of the world's population that wants a lightweight, easy to use desktop operating system that is allows them to access the "new technologies" they've heard of (you know, this email stuff and the world wide wait?) with a minimum of intervention or knowledge on their part. Those of us who enjoy the power and configurability of linux should really stop trying to jam it down the throats of the average consumer - we like our operating system powerful and configurable, but the average consumer does not agree with us. Oh sure, we say "look, it's got a GUI, and an installer, it's waay easy, check out RPM, a monkey could install Linux". But we forget that the average monkey, apart from eating bananas and scratching their butt, has very little else to do and could spend the next week or so finding the proper drivers and poking through configuration scripts. Joe Consumer sees his computer and thinks TeeVee and he wants to be able to boot up his computer and have everything set up and clean with no "do not touch this" buttons that could potentially ruin everything. This is why the "non user serviceable" parts to a TeeVee (ie the syncs and other such adjustments)are hidden away from view. As they are with BeOS. Linux, OTOH, is like a TeeVee with the back cover removed - it's suitable for those who like to fiddle and know enough not to touch the toroid. Anyway, just thought I'd mention that - I do hope that linux continues to gain support and popularity, but if you're looking for something that's going to replace Windows for the majority of home users, I'd say look no further than Be.

Re:An objective benchmark.

They used XF86 3.3.6

Re:BeOS..

The BeOS filesystem is a hideosly fast journaling implementation called bfs. It IS faster than anything available on Linux. Messaging speed is also top notch. I don't know how you benchmark mem. management, but BeOS has good memory protection and uses about half the memory of Linux/GNOME on the same machine.

Re:what do you expect?

Actually, its designed as a good OS. Multimedia OS is just something the marketing droids call it because it sounds cool, and its a good niche for a wonderful operating system.
BeOS is from the ground up, a beautifully thought out and put-together operating system. I don't believe an open source development process could have created such a result. I also don't believe Micro$oft will ever catch up without completely ditching their entire codebase and starting fresh. Hey they tried to do just that thing with NT (in collaboration with IBM), but we can all see how well that turned out.
I don't mean to this to sound like flamebait, so I apologize. I just want to make the point that BeOS is designed as a single user desktop OS with a first rate filing system, process and memory management and a clean interface to boot. That makes it great for writing music or working with graphics or whatever simply because they tax the parts of an OS which BeOS is particularly well designed. BeOS capabilities with different types of media is a result of its design, not the other way around.

Spyky

Re:My thoughts on BeOS

I do think, however, that you're analogy is a bit wrong. Be has most of the cool UNIX stuff you like in Linux, but its hidden away under the / directory. (For you non Be users, the GUI has /boot as the reference so stuff like /bin, /lib, and /etc are never seen by the user.) It's more like a TV with an easily removed cover, wheras Win98 is like a TV with the cover welded shut. (For good reason, people would puke if they ever say what was in there!)