Dual Core Intel Platform Shootout - NVIDIA nForce4 vs. Intel 955X
by Anand Lal Shimpi on April 14, 2005 1:01 PM EST- Posted in
- CPUs
SATA Controller Performance
Both NVIDIA and Intel offer support for NCQ in their SATA controllers, and given our recently renewed interest in NCQ performance, we decided to find out if there were any performance differences between the two SATA controllers. However, as we've found in the past, coming up with tests that stress NCQ is quite difficult. Luckily, there is a tool that works perfectly for controlling the type of disk accesses that you want to test: iometer.
An Intel developed tool, iometer allows you to control the size, randomness and frequency, among other things, of disk accesses, and measure performance using data generated according to these specifications. Given that NCQ truly optimizes performance when disk accesses are random in nature, we decided to look at how performance varied according to what percentage of the disk accesses were random. At the same time, we wanted the tests to be modeled on a multitasking desktop system, so we did some investigation by setting up a computer and running through some of our multitasking scenarios on it.
What we found is that on modern day hard drives, the number of outstanding IOs (IO Queue Depth) is rarely above 10 on even a moderately taxed system. Only when you approach extremely heavy multitasking loads (heavier than anything that we've ever tested) do you break into queue depths beyond 32. So, we put together two scenarios, one with a queue depth of 8 and one with a queue depth of 32 - the latter being more of an extreme condition.
In each scenario, we sent the drives a series of 64KB requests, 75% of which were reads, 25% were writes; once again, derived from monitoring our own desktop usage patterns.
We then varied the randomness of disk accesses from 0% (e.g. 100% sequential) up to 100% (0% sequential reads/writes). In theory, the stronger NCQ controllers will show better performance as the percentage of random accesses increases. We reported both Average IOs per Second and average IO response time (how long accesses took to complete on average):
With a queue depth of 8, the two SATA controllers offer virtually identical performance.
Looking at latency, Intel actually offers a very slight performance advantage here - nothing huge, but it's definitely there.
The results get much more interesting as we increase the queue depth to 32:
Here, NVIDIA starts to pull away offering close to a 20% increase in average IOs per second as the access patterns get more random (e.g. as more applications running at the same time start loading down the hard disk).
What's truly impressive, however, is the reduction in average response time - up to a 90ms decrease in response time, thanks to NVIDIA's superior NCQ implementation.
But stepping back into reality, how big of a difference NVIDIA's NCQ implementation makes depends greatly on your usage patterns. Heavy multitaskers that are very IO bound will notice a performance difference, while more casual multitaskers would be hard pressed to find any difference. For example, Intel was actually faster than NVIDIA in our gaming multitasking scenarios from our dual core investigation.
Facebook
Twitter
RSS
96 Comments
View All Comments
Questar - Thursday, April 14, 2005 - link
Oh! A link to Tom's....Can I link to Tom's article about Athlons that burn themselves up?
Please, no respecting Anandtech reader should be siting Tom.
Questar - Thursday, April 14, 2005 - link
#21,Let me explain it to you:
Intel get's a cut of the money from every chipset nVidia sells. What part of that don't you get?
But that's okay that you don't understand that, after all your post implies that AMD is going to put Intel out of business: "You are pretty stupid if you think Intel has a chance against AMD".
Ummm...yeah right, go right on thinking that.
overclockingoodness - Thursday, April 14, 2005 - link
#23 segagenesis: I don't it's worth to post links here. It's quite apparent that Questar is an idiot himself, so why bother. :)We are only wasting our energy on ignorant people like him.
segagenesis - Thursday, April 14, 2005 - link
#20 - Sorry to rain on your parade but http://www.tomshardware.com/cpu/20041114/index.htm...Quit calling people idiots when you dont even keep up on current events.
overclockingoodness - Thursday, April 14, 2005 - link
#20 QUESTAR: "Too Hot" is merely a figurative comment. Don't try and be a smart ass. We all can clearly see through your Intel favoritism. You are definitely not as knowledgable as Anand or some of the people here, so get lost.If you don't like what AnandTech has to say, stop reading the site. People like you only waste valuable bandwidth, plus, it will be one less troll on the Internet.
overclockingoodness - Thursday, April 14, 2005 - link
Questar: Please, you're comments are quite stupid. You are not the only one with Power, Intel and AMD CPUs, you know.Just to let you know, AnandTech has a reputation of being the best of the best, and Anand is the pioneer of reviewing hardware, so he has been in this business for a long time. Therefore, it means that he has seen quite a bit of hardware in these 8 years running AnandTech. It's pretty ignorant of you to question him.
I agree with everyone. You are pretty stupid if you think Intel has a chance against AMD. Prescotts are illogical and rather poorly designed CPUs.
"Intel probably makes as much net profit off the licensing of the nVidia chipset as they do selling thier own - after all thay don't have to design, build, ship or sell anything. So why would they be worried?"
Once again, your opinion. Can you please get Intel to leak these numbers to you, so we can have a reason to believe you?
Licensing prices are fixed. Sure, Intel could be making more money from licensing their technologies to NVIDIA, but what will happen in the future when PCIe and DDR2 will start to pick up the pace. Then, Intel would want to sell as much of their chipsets as possible for maximum revenue and when you have a strong chipset maker like NVIDIA, it would be pretty hard, don't you think? In the future, NVIDIA's licensing fee wouldn't cut it.
It's pretty logical: If company A makes chipsets and company B makes the chipset with same technologies, the market will surely divide between the two.
I guess your brain is sealed somewhere, which is why you probably can't think straight. I hate stupid people and I think you are one of them.
Questar - Thursday, April 14, 2005 - link
Yes Prescott CPU's are hotter than some other CPU's. Stating they are "too hot" is opinion. Please provide your quanitative proof that they are "too hot".segagenesis - Thursday, April 14, 2005 - link
So are you avoiding having to make sense? Quit making up stories when you call for proof yet provide none yourself. I gave you proof, now its your turn.Here, I'll help you decipher it because you seem to be ignoring posts in favor of your own flawed logic. Here is a snippet of one of your own.
> "Honestly, Intel processors and even the platform haven’t been interesting since the introduction of Prescott. They have been too hot and poor performers, not to mention that the latest Intel platforms forced a transition to technologies that basically offered no performance benefits (DDR2, PCI Express)."
> Your opinion only, don't make this out to be fact.
The link I provided shows that in *fact* there is more heat output by modern Intel processors. Yes, this is a quantitative analysis. If it was qualitative you could have called it opinion, but its not eh? Try again.
Questar - Thursday, April 14, 2005 - link
#16,So are you changing the argument? I never argued about heat. Please have somebody that is capable of actual cognitive thought explain to you what post #8 says.
Questar - Thursday, April 14, 2005 - link
#14,Actually I'm rather agnostic between CPU makers. I own systems based upon Power, Intel and AMD cpu's.