Slashdot Mirror
AMD's Athlon-64 Benchmarked With UT2003
Sander Sassen writes "Wondering about the performance of AMD's Athlon-64? Want to how well it runs in 64-bit mode? Hardware Analysis managed to run a few benchmarks on a AMD Athlon-64 demo system using the 64-bit version of Epic' Unreal Tournament 2003. There's also an update with the latest about Athlon-64, Opteron and mobile Athlon-64 including streaming video and pictures of a quad Opteron server."
AMD is soon to be followed by Intel with the introduction of the Unobtainium chip.
~S
how to slashdot an innocent server in the dead of night.
...a 64-bit CPU is totally *pointless* unless I can spawn at least 500 bots on a map designed for 7 at playable speeds with it. Telefraging madness!
"To confine our attention to terrestrial matters would be to limit the human spirit." -Stephen Hawking
As promised we'll give you an update on the performance and other features of the AMD systems that we reported on yesterday. Naturally AMD wasn't very keen on disclosing clockspeed or detailed system configurations of the demo systems they had running, but we took advantage of a few fellow journalists entering the room and keeping the AMD PR people busy to run a few quick benchmarks on the Athlon-64 system. The Athlon-64 demo system we already reported on yesterday had a 2GHz clockspeed and used the SuSe 64-bit Linux operating system and was running the 64-bit version of Unreal Tournament 2003 as a demo.
(pic of monitor playing UT2k3, FPS = 42)
Fig 1. The Athlon-64 system running SuSe 64-bit Linux and the 64-bits version of Unreal Tournament 2003.
Naturally we're intimately familiar with the workings of Unreal Tournament 2003 engine and after a quick look at the display settings, which were set at a 1024x768x32bit resolution with all other features at default, we measured a mere average 42fps and maximum fps around the 55...60fps mark. Considering the fact that this is a 2GHz Athlon-64 processor teamed up with a GeForce Ti 4600 we honestly expected a whole lot better. A 1.6GHz Pentium 4 with that very same GeForce Ti 4600 videocard would have no problems clocking in a similar score while running under Windows XP.
(pic of a white laptop with the terminal showing)
Fig 3. The Athlon-64 notebook running CyberLink's PowerDVD actually showing the first Harry Potter movie.
But there's more, we managed to take a closer look at the notebook too and quickly found out that this indeed is a proof of concept. It plays DVDs very well, mostly courtesy of the ATi M9-series graphics card, and unfortunately all our questions about whether we could do something else with it were answered with a resounding 'no'. We did however manage to find out what was inside in terms of chipset, memory and graphics card. The notebook apparently used a Via K8T400M chipset teamed up with an ATi M9-series graphics adapter and was using PC2100, DDR266, memory. The screen was a standard 14.1 inch running at a 1024x768 resolution and the DVD software they used was none other than CyberLink's PowerDVD.
(naked pizza-box style case)
Fig 2. The quad Opteron server with the top cover removed, the PCI-X slots in the back and a the four CPUs hidden underneath the huge heatsinks.
(two white LCDs next to each other, left showing a web browser and right showing UT2k3.)
Fig 3. The quad Opteron with the SuSe 64-bit Linux operating system running some sort of a database benchmark, right next to the UT2003 demo machine.
We naturally also took a closer look at the quad Opteron as that's definitely something AMD is currently pushing hard. They're putting all their weight behind the launch of their server products and have postponed the launch of the desktop version of the Opteron, the Athlon-64, back to September. The server parts, including completely pre-configured two-way Opteron systems, should be available in late April, right after the April 26th launch of the Opteron server CPU family. Clockspeeds will initially range up to 1.6 or 1.8GHz and performance is expected to be similar to Intel's Xeon offerings. But as always, we'll reserve judgement until we can actually evaluate two similarly configured servers side by side, for now all they have given us are SpecInt-2000 and SpecFP-2000 scores without disclosing the system configurations, so that doesn't tell us anything.
Nevertheless it looks like AMD is indeed trying to get some new and innovative products out of the door. Whether they'll be able to make a lasting impression, both in terms of features and performance, with their new 64-bit products remains to be seen though, we'll be sure to keep a close eye on any future developments.
Sander Sassen.
It's much more entertaining to keep refreshing the page and watch the user count rise
Please register or login. There are 10 registered and 1173 anonymous users currently online. Current bandwidth usage: 2777.75 kbit/s
Looking for hardware (Currently need: Large Etch-a-Sketch) Have one? See my journal!
Ok, wait.
I'm the developer that did the 64-bit port of UT2003 (and the Linux port, and the Mac port...).
You need to keep two things in mind:
1) The OpenGL renderer is not as fast as the Direct3D renderer at this time. This is not the Athlon64's fault. You can see this on 32-bit Windows, since it can use both renderers. Since this is a Linux port of the game, we're using the GL renderer on the Athlon64 at this time.
2) The "stat fps" command isn't really a good benchmarking method.
3) This is a prerelease version of the game running on a prerelease version of SuSE running on prerelease drivers running on prerelease hardware. Please don't consider this "benchmark" to be representative!
--ryan.
Don't say, "don't quote me," because if no one quotes you, you probably haven't said a thing worth saying.
The site, with just a few comments on /., is already showing signs of slashdotting. I'll quote the most important parts about the UT 2003 benchmark, just in case:
"The Athlon-64 demo system we already reported on yesterday had a 2GHz clockspeed and used the SuSe 64-bit Linux operating system and was running the 64-bit version of Unreal Tournament 2003 as a demo."
-snip-
(at 1024x768x32...) "we measured a mere average 42fps and maximum fps around the 55...60fps mark. Considering the fact that this is a 2GHz Athlon-64 processor teamed up with a GeForce Ti 4600 we honestly expected a whole lot better."
Beware: In C++, your friends can see your privates!
You're obviously not a real geek. The fact that the streaming video and pictures are of the quad Opteron server made me click. Unfortunately, it is already /.ed.
I'd like to know if it will run nice and frosty like my AMD 2400+ (plus what? Beats me!)
Any more fans and my computer may levitate. That would be just as cool as a good UT2003 framerate!
I saw that and determined that they were more Windows type people and plain flat didn't know that the OpenGL renderer is much weaker (not due to the API, but due to this being much the first cut of the thing...) than the D3D. What they measured was pretty good considering that detail.
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
The reason you keep losing at UT isn't because your processor is too slow.
I have been pwned because my
i would have to say that this article is about as worthless as the bill gates quotations earlier, int terms of actual usefullness and truthfullness. since everything is prerelease and the details are fairly sketchy, im gonna wait for solid numbers before i decide once and for all who i will be loyal to in the proccesor world.
how well are the drivers for the gefore card working? are they playing nice with that k8t400? are the nvidia drivers 64bit or are they being run in "32-bit" mode? how well is OpenGL playing with the 64-bit OS, 64-bit Chip combo and again, how well are the nvidia drivers playing? is the OS running the AGP in AGP mode or is it PCI mode?
i bet i could easily get a P4 2.7 with this graphics card to product similar numbers, or even worse in linux with some effort to use least optimized drivers and setting the graphics card to PCI.
in fact, my P4 2.4x133@2.7x150 with a GF Ti 4600 doesn't post much better numbers, 55fps by stat fps. and thats on a 32bit "system" with fairly mature drivers and everything work "correctly/fullspeed"
im not an AMD zealot, but i wont make me decision based on a game that is notoriously bad at opengl and on a system that is running all beta software/drivers.
Let's not get over-excited... This is of course interresting information, but it's information of a premature chip on a premature platform.
I doubt that any proper conclusions can be drawn from this, apart from what is already known: The Athlon 64 isn't ready yet. If was the release date wouldn't be set for September.
Much like with Doom III, there is always a cool-factor, but the actual facts at hand are very scarse. One thing is probably for sure though... The Hammer core can't compete with the Barton core on the desktop at this point. Otherwise we'd have the Athlon 64 waiting to be released much sooner.
.: Max Romantschuk
If one compares the claimed 42fps with other cpu:s, it seems it is at a level of a Celeron 500 MHz...
There is something fishy here as the UT2k3-makers themselves claimed there is a 15% increase in 64-bit mode (on Windows). Normally Quake3Arena for Linux is on par with the Windows version, so it should not be the OS' fault either.
the real benchmarks I'm interested in:
How many *C does the CPU run at?
What size PSU does it need?
I curious... how do the extra bits per clock cycle supposed to increase performance? I mean the number of instructions per second don't increase...
Yuioup
I've worked inside a handful of medium-sized machines, including a couple Microway dual Alpha "rugged racks" and a Compaq ES40 Model II. The 8u (or 9?) ES40 is nicely laid out, but removing the motherboard or messing with drive cables is a pain. The 4u "rugged racks" are a disaster of fans and wires. We've had 4 or 5 fans go out on those, and it takes approximately 45 screws (I counted) and about 60 minutes of fast work to remove and replace a midboard fan.
That opteron case, on the other hand, appears to have plenty of cooling that is easy to reach. I don't see any wires permanently attached to the case. It looks very clean and easy to service, except possibly getting the motherboard out.
-Paul Komarek
The article said that there will be benchmarks. And there are none. A screen shot of a game does not qualify. I want to see the whole spec or at least the basic ones. And after that I can look at the game snapshot.
Moderator seriously why this posted with such a misleading title?
>> Sure it will run faster, but do you seriously need that speed?
Sure, just like I needed more than 512K of RAM.
I recently got into video editing. Until now, I've never needed anything faster than a single CPU system.
Now I understand, completely, what those who do rendering gripe about when it comes to CPU speed never being fast enough.
2:57 of video takes my 1GHz w/ 1GB RAM machine nearly 2 HOURS to render. Just for 3 stinking minutes of video!
The fastest current single CPU would only decrease that number to about 40 minutes, which is still too slow.
A dual CPU solution would bring it down to 20 minutes, but again, if I ever wanted to render even 15 minutes of video, that would be 1 hr 40 mins of CPU time.
And forget doing anything else with the computer while it's rendering. It will start dropping frames like mad, and you have to start over.
Now a 4-way workstation is something that would work. With a 4-way 3GHz Opteron system, I could render in near real-time, and a regular sized MB, if not slightly oversized, could handle 4 procs.
SuperMicro is the only MB mfg. I know of that makes a 4-way board, but it's for Xeons and is insanely expensive ($1800 +/- $100) and that's before you add the overpriced CPUs.
If AMD came out with a moderately prices 4-way workstation, they could get the CAD/CAM, video editing, 3D modeling, rendering and compiling crowd all at once, in addition to the freak gamers and Gotta Have The Best Even Though I'll Never Use It crowds.
The 4-way system is a neglected niche. AMD should fill it.
Talisman
"Study your math, kids. Key to the universe." -The Archangel Gabriel
"You've made your point Hans, let them pull back!"
"I'll take it under advisement, Mr. Cowboy, now HIT REFRESH AGAIN"
(followed by an explosion, either of a police RV, or of a webserver).
"Nothing strengthens authority so much as silence." - Charles de Gaulle
Oh my god,The server is toast!!!
And forget doing anything else with the computer while it's rendering. It will start dropping frames like mad, and you have to start over.
Unless you're talking about capturing, which you *will* have to be able able to do in real-time to avoid losing frames, how exactly do you manage to lose frames during rendering? The only way I can think of is by working on a preview while the "real" render is made from tape, but that'll require you do to the actual render in real-time too. Frankly, you're not making any sense to me.
Kjella
Live today, because you never know what tomorrow brings
I call BS on that. There was a noticeable improvement when moving from AGP 1x to 2x. The difference was nonexistant when moving from 2x to 4x. Same thing when moving from 4x to 8x. AGP is definitely NOT the bottleneck!
Lesbian Nazi Hookers Abducted by UFOs and Forced Into Weight Loss Programs - -all next week on Town Talk.
There was nothing to this artice. Here is what I learned from reading this article: 1)there's a 64-bit linux port of UT 2003 2)amd likes secrecy 3)the people who were showing off the laptop like Harry Potter wow. now wasn't that informative?
"Some kind of database benchmark." Thanks for that insightful analysis of the 4-way, and pimping your own site on Slashdot. Tasteless!
...which you would expect if you'd never actually looked at any of the UT2K3 benchmarks on the net that show frame rate scaling linearly with CPU speed.
3 &p =13
for example - with a Radeon 9700:
botmatch:
Intel Pentium 4 1.5GHz = 35.5 FPS
Intel Pentium 4 3.06GHz = 69.6 FPS
flyby:
Intel Pentium 4 1.5GHz = 114.5 FPS
Intel Pentium 4 3.06GHz = 205.5 FPS
http://www.anandtech.com/cpu/showdoc.html?i=178
I'm not saying the original article made a fair comparison, but the game really does rely on the CPU a lot more than you seem to think.
Wow! 2756 anonymous users online?! I wonder where they came from.
Send us your Linux Sysadmin articles.
Geeky modern art T-shirts
Good work on the port(s).
Before the Intel/AMD fanboys go crazy. I wanted to get a few questions in:
1. Can you tell us what specific optimizations you have done/are planning to do for the 64 bit architecture?
2. What optimization benefit do get from a straight "re-compile" of the UT codebase in 64 bit mode?
cheers,
j.
"My cat's breath smells like cat food." - The Tao of Ralph Wiggum.
Anyone that has played the pathetic UT2003 port to Linux will know that it is many times slower than the Windows version. The game was coded for DirectX. It uses a wrapper in Linux to convert Direct3D calls to OpenGL in real-time, and it *REALLY* puts a tax on the host CPU.
In my experience, properly ported OpenGL games on Linux (like RTCW) were faster than under Windows, but UT2003 is definately not the case. For that reason, these comparisons are way too early. I can't speak for the port to the Athlon64 architecture, but when it comes to the 32 bit version of UT2003 for Linux, it's very slow in comparison to the Windows release. This is what happens when you code a game for one platform, one API, and then try to port it to other operating systems.
Yippee-ki-yay, motherfucker!
The world moves for love. It kneels before it in awe.
Epic releases UT2003 text mode.
This has the advantage of being playable on all kinds of hardware specifications, from a measly 8086 to AMD's flagship AMD 64...
You hop Alice-in-Wonderland-like in a room full of bots. What do you do?
>
/. Where the truth
There have been some disappointed posters, wondering why it isn't faster. Stop and think about it: Why would a 64-bit CPU be faster than a 32-bit CPU? It's not bus width, because Pentiums have always had 64-bit busses. It's not FPU width, because x86 FPUs have always been 80 bits internally. It's not 64-bit integer registers, because it's very rare indeed to need to do 64-bit integer math. It's not 64-bit pointers, because this is a machine with less than 4GB of memory. What it comes down to is that this processor is using slightly newer tech than AMD's previous chips, including a larger cache. But it has nothing whatsoever to do with being 64 bits, and hence the results are not mindblowing.
There's a persistant myth that a 64-bit processor is twice as fast as a 32-bit processor, which is completely incorrect.
Which might indicate that vsync is enabled, effectively capping the the max fps while lowering the average. Whenever I run a benchmark and it tops out at 60 fps and I suspect, as these guys did, that the machine should be faster, I always double check the refresh rate settings and vsync.
-dameron
Did anyone notice that the location of the video, keyboard, mouse, floppy drive, cd-rom,, and power cable are opposite from the pci slots? I have worked with quite a few different rack servers myself in the past, but this sure is one strange looking monster. Not to mention that it really does have some massive heat sinks and the RAM slots appear to be staggered around the motherboard, somewhat like the Sun's motherboards.
Perhaps someone has some insight into other types of rack mounted systems and motherboard configuratioins they can share?