Intel Core Duo: AOpen i975Xa-YDG to the Rescue
by Gary Key on May 4, 2006 8:00 AM EST- Posted in
- Motherboards
Firewire and USB Performance
After looking at many options for Firewire and USB testing, we finally determined that an external USB 2.0, Firewire 400, and Firewire 800 hard disk would be a sensible way to look at USB and Firewire throughput. We utilize a RAM disk as our "server", since memory removes almost all overhead from the serving end. We turn off disk caching on the USB and Firewire side by setting up the drives for "quick disconnect" so our results are consistent.We use 2GB of fast 3-3-3-8 system memory set up as a 450MB RAM disk and 1550MB of system memory. Our standard file is the SPECviewPerf install file, which measures 432,533,504 bytes (412.4961MB). After copying this file to our RAM disk, we measured the time for writing from the RAM disk to our external USB 2.0, Firewire 400, or Firewire 800 drive utilizing our internal Windows based timing program. The copy times in seconds were then converted into Megabits per second (Mb) to provide a convenient means of comparing throughput. Higher Rates therefore mean better performance in this particular test.
Possibly the most interesting finding in our Firewire and USB throughput tests is the outstanding performance of an external hard drive connected to Firewire 800. Our benchmarks show Firewire 800 is up to 48% faster than a drive connected to the more common Firewire 400, and about 32% faster than the fastest USB 2.0 solution.
The AOpen board offers the Agere based IEEE 1394 Firewire option that provides very competitive performance with the more common TI chipset solutions. The Intel USB 2.0 performance offers 22% better performance than the ULi based controller in throughput.
Ethernet Performance
The current motherboard test suite includes LAN performance measurements. All of these boards utilize PCI or PCI Express based controllers with the only difference being the supplier of the core logic.The Windows 2000 Driver Development Kit (DDK) includes a useful LAN testing utility called NTttcp. We used the NTttcp tool to test Ethernet throughput and the CPU utilization of the various Ethernet Controllers used on the Intel motherboards.
We set up one machine as the server; in this test, an Intel system with an Intel CSA Gigabit LAN connection. Intel CSA has a reputation for providing fast throughput and is a logical choice for our Gigabit LAN server.
On the server side, we used the following Command Line as suggested by the VIA whitepaper on LAN testing:
Ntttcpr -m 4,0,‹server IP› -a 4 -l 256000 -n 30000
Ntttcps -m 4,0,‹client IP› -a 4 -l 256000 -n 30000
AOpen's choice of the Marvell 88E8053 10/100/1000Mb/s Ethernet controller is an excellent addition to this board. The performance is basically equal to that of the same Marvell 88E8053 controller utilized on the Asus board. The lower CPU utilization rate on the Marvel 88E8001 PCI controller is directly attributable to the lower throughput numbers generated in our tests.
Facebook
Twitter
RSS
81 Comments
View All Comments
Gary Key - Friday, May 5, 2006 - link
The card was delievered over the weekend, Derek only had a couple of days and nights to test it. He will be expanding upon this article in the near future as more games are launched with support, not to mention all of our test platforms will undergo a radical change here shortly. ;-)
goinginstyle - Thursday, May 4, 2006 - link
Best quote of the day. Ignore that guy as he is just a tool or had his first visit on a computer today.
ShapeGSX - Thursday, May 4, 2006 - link
This is a server chip vs a laptop chip. If anything, the AMD server chip had the upper hand in this review, and the little Intel laptop chip bested it in almost every category.Though, technically the review was a review of the motherboard, not the CPUs, I think it speaks volumes as a comparison of the K8 vs the Yonah architecture given the identical clocking used.
Frumious1 - Thursday, May 4, 2006 - link
"If anything, the AMD server chip had the upper hand in this review."Again with the reading comprehension! Look at the chart for Christ's sake. Notice that a bunch say "lower is better"? At 2.8 GHz, the Core Duo sweeps the tests - only in disk controller performance could you potentially level a complaint. At 1.83 GHz, it's a bit closer, but to say the Opty has the upper hand?
http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...">http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...
Moderate to substantial wins by Core Duo across the benches.
http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...">http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...
More ties or wins for Core Duo.
http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...">http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...
Opty wins in Nero Digital Audio... which I have never even used or seen as a benchmark. One win for Opty 165! Woohoo! AMD Rules!
http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...">http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...
More substantial losses in file compression.
http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...">http://www.anandtech.com/mb/showdoc.aspx?i=2750&am...
It loses every gaming test by a reasonable margin, with or without CrossFire.
ONE win for AMD at 1.8 GHz, and that's in Nero Digital Audio. A few ties, but otherwise AMD loses. Boy, I can't imagine why AnandTech would do an article like this. I mean, Intel managing to win almost every benchmark is old news! I remember Pentium II/III spanking K6/K6-2/K6-3 ages ago. Looks like nothing has changed... except for the whole K8 vs. NetBurst era where Intel got the shit kicked out of it!
Intel looks primed to take back the performance lead. I've been running lots of AMD K8 systems for the past 3 years, but it looks like I will now have a serious reason to consider Intel again. (Before K8, I ran mostly K7 setups. Before that I was always running Intel because they were better. Notice the pattern? Buy the BEST chip, regardless of who makes it!)
Calin - Friday, May 5, 2006 - link
What is sad for AMD is the fact that the Intel chip is advantaged by frequency increase. At 1.8GHz, they are more or less at a tie (with few not-so-great exceptions), and the increase in frequency to 2.8GHz favors Intel much more than AMD.So, overclockers would choose the chip that will give them a bigger increase in performance for the same increase in MHz (Intel). The situation changed from the Athlon64 versus Prescott days.
Gary Key - Thursday, May 4, 2006 - link
His context was in the positive, meaning the AMD had the upper hand going into the review but was outperformed in several areas.
:)
ShapeGSX - Thursday, May 4, 2006 - link
Indeed it was meant to be positive. Thanks!Laptop cpus beating server cpus, dogs and cats living together!
Frumious1 - Thursday, May 4, 2006 - link
I'm not sure what you read, but I saw a review in which a ~$590 Core Duo + AOpen board was able to basically match a ~$550 Opteron 170 + ASUS board in performance. What exactly don't you like - the fact that Core Duo overclocks more than 50% on the board? Or the fact that Athlon X2/Opteron doesn't win every benchmark?Thanks for being such a retard. FYI, Intel doesn't want this type of review, because AT is basically promoting buying their $240 CPU and overclocking rather than buying their $600 CPU. Let me rephrase your post:
ANANDTECH
In this review, we'd like to show your how an "Athlon FX-62" compares to a 2.8Ghz Core Duo.
..."As you can see, the Core Duo actually beats the FX-62 equivalent on just about every fucking benchmark. AMD's former monster has been humbled, and it looks like the stupid ass AMD fanboys like snorre need to stop snoring and brush up on their god damned reading comprehension! If that's not enough, Core Duo will add another 25-40% performance clock for clock over Core Duo (see Johan's article). Needless to say, even the best AMD is prepared to offer looks to be in serious trouble."
Don't let the door hit your ass on the way out. Comparing this to a THG article is an insult to both parties. THG would have used more hyperbole and run fewer benchmarks, while AT wouldn't accept large cash payment to do an article. Go back to whatever black ole you crawled out of. PLEASE!
BTW thanks for proving that no matter how good an article is some stupid shit will wander in and bitch about the results. "OMG my eyes! I can't look at a graph and stand to see AMD lose!" The Greater Internet Fuckwad Theory has been proved yet again. (Google that if you don't get the reference.)
JarredWalton - Thursday, May 4, 2006 - link
I think I see the problem, frumious. You used some odd text in your post and it killed the colors.JarredWalton - Thursday, May 4, 2006 - link
Note to others: don't use the abbreviation for HardOCP. LOL