The Clarkdale Experiment: Mea Culpa
by Loyd Case on May 4, 2010 1:57 PM ESTSeveral months ago, I wrote about my little experiment with Clarkdale, where I built a small form factor system, based around a Core i5 661 CPU, an Asus H55 motherboard and a Radeon HD 5850 graphics card. That system also used a pricey, 250GB SSD, which was a little over the top for an otherwise modest system, but the idea was to make it small, quiet and low power.
Quite a few people critiqued the article, and most of the critiques revolved around the lack of performance data. After all, the general feeling went, how do we know this is really a good system? It’s pricey, to be sure, but we also have no way of judging performance.
As it happens, I have another small form factor system in the basement lab, which happens to be running a Core i5 750. Ignoring hard drive performance for the moment, all I really needed to do was swap out the graphics card, since the Lynnfield system was running an older Radon HD 4870. So I dropped in a Radeon HD 5850 and took both systems for a spin.
Price versus Performance
There’s a time and place for integrated graphics – but not for PC gaming. Dropping in a Radeon HD 5850 likely means the system will be used for PC gaming – which was my intent all along. The Core i5 661 is priced nearly identically with the Core i5 750. Beyond price, the differences are pretty substantial:
· Clock speed: 3.33GHz (Clarkdale) versus 2.66GHz (Lynnfield)
· Maximum Turbo Frequency: 3.6GHz (Clarkdale) versus 3.2GHz (Lynnfield)
· Dual core with hyper-threading (Clarkdale) versus quad core without hyper-threading (Lynnfield)
· 4MB shared L3 cache (Clarkdale) versus 8MB shared L3 cache (Lynnfield)
· TDP: 87W versus 95W (Lynnfield)
So Clarkdale has a 25% raw clock speed advantage and a 12.5% maximum turbo boost performance. Lynnfield has the edge in cache – while both have a shared L3 cache, the sheer cache size gives Lynnfield an edge over Clarkdale in cache sensitive apps. Lynnfield also has four actual, physical cores, rather than two physical plus two SMT (virtual) cores. Interestingly, Clardale only has a modest TDP advantage over Lynnfield.
With these thoughts in mind, I ran a number of game benchmarks, plus a few other performance tests. Given the clock speed disparity versus cache size and number of cores, I didn’t expect big performance disparities. As it turns out, I was in for a few surprises.
System Configuration
These systems weren’t identically configured, but were similar.
|
Component |
Clarkdale System |
Lynnfield System |
|
CPU |
Core i5 661 @ 3.33GHZ base |
Core i7 750 @ 2.67GHz base |
|
Motherboard |
Asus P7H55-M EVO |
Gigabyte GA-P55M-UD4 |
|
Memory |
2 x 2GB OCZ DDR3-1600 @ 1333 |
2 x 2GB OCZ DDR3-1600 @ 1333 |
|
GPU |
XFX Radeon HD 5850 |
XFX Radeon HD 5850 |
|
Hard Drive |
OCZ 250GB SSD |
WD Caviar Blue 640GB |
|
Optical Drive |
Asus BD-ROM / DVD+/-RW |
Lite-On DVD+/-RW |
|
PSU |
Cooler Master 500W |
Cooler Master 500W |
The key differences in components were motherboards (Asus H55 versus Gigabyte P55) and hard drive (an SSD versus a standard rotating media drive.) None of the tests I ran were particularly storage intensive, and any power advantage due to drive differences were pretty minimal.
Facebook
Twitter
RSS
45 Comments
View All Comments
zappb - Wednesday, May 5, 2010 - link
Loyd,I've been a regular reader of anandtech for over 10 years. This is the first time I've registered.
Just to say I enjoyed this review (and the previous Clarkdale experiment) and was perfect for my level. The previous Clarkdale experiment article I printed out and brought it to the bathroom to read in work! Ok I'm sure you didn't want to know that.
Sometimes you just need to throw in a 256 gigabyte SSD, just because you can!,, I would have done this myself.
Good job and keep up the good work.
Peace
mathew7 - Wednesday, May 5, 2010 - link
You say that the power difference is negligible. But I can tell you it's not. I'm not talking about money savings, I'm talking about noise. Especially in a SFF. I bet you did open-bench testing.I have a MicroATX cube with E8400 (soon to be replaced with i5), ASrock G43TwinS, ATI 5850 and Corsair HX450W. The PSU is right above the 5850, with the fan covered by the 2-slot profile of the card.
I play with desktop speakers as I wear glasses and was unable to find a good headphone set.
When I played Mass Effect 2 (I don't recall if it was before or after underclocking the 5850), I noticed after 30min of gameplay a noisier than usual computer. But how I noticed it? Because every time there was saving or loading involved (including level change) I would hear an rpm lowering. The rpm was being lowered less than 1s after save/load triggering and would get high again less than 1s after it finished the action. The CPU has the big heatsink that would not permit such fast reaction. The GFX card fan was not, as I would place my finger to stop it from spinning and no noise change was heard. So the only remaining "stressed" component was the PSU (do NOT tell me that I have a too weak PSU, because I have NEVER had a system that would consume more than 350W at the UPS).
So dismissing the Clarkdale because it's slower and the power difference is not justifiable by savings is not a good reccomandation. If you test SFF components, PUT THEM IN A SFF CASE. The combined effect of restricted airflow and higher power draw could put the PSU in a "crank up the fan" situation. So maybe the 661's power advantage is much better seen(or heard) then 750's performance advantage IN A SFF.
PS: in the system comparison table you wrote "core i7 750".
7Enigma - Wednesday, May 5, 2010 - link
Saw your post and wanted to comment (hopefully you see it):I, like you, wear glasses and headphones were always an annoyance. But with a kid and another on the way the only time I get to play games is at night when said kid is asleep and a booming sub and loud speakers would not be "allowed". So I did a bunch of searching and came up with the highly recommended Seinheiser HD280. I got mine for $80 on a sale and they frequently go on sale at the major e-taliers. Fantastic sound quality and surprisingly comfortable even after extended gaming/listening to music/movies with glasses.
Check them out, and I hope this helps a fellow 4-eye. :)
Finraziel - Friday, May 7, 2010 - link
While you have a point, I think you'd be better off improving your airflow or for instance choosing a case where the PSU draws its air from outside, or just get a better PSU, rather than choosing a slower yet more expensive part just in case your PSU fan might get a bit antsy. I think your example mostly serves to not underestimate the impact of choosing the correct PSU if you want a silent system.Also, don't overlook that the difference in power draw can not possibly be only or even largely from the CPU difference. There's only an 8 watt difference in TDP and while I know TDP does not equate to how much a CPU actually uses, I think it's fair to say that the 750 itself isn't actually consuming over 60 watts more than the 661. Since P55 mainboards don't draw much power either (I have to admit I have no real numbers, but how else would they be able to survive without bigger heatsinks), the biggest difference I would say has to come from the videocard that is allowed to work harder because the 750 can feed it better. So if it really bothers you on a hot day or something, underclock your CPU and videocard and you should get the same effect.
On a sidenote, I agree on the sennheiser :). I also wear glasses and also have one of the sennheiser models that go completely over your ear and with a soft cushion... Unfortunately I don't know the model nr and am not at home, but it's survived for about 10 years now already I guess, including dragging it along to quite a few lanparties. I absolutely love it, investing in good comfortable headphones is definately worth it I'd say.
basket687 - Wednesday, May 5, 2010 - link
Please clarify how you measured power consumption, did you run a game, a stress test, or something else?The difference in power consumption under load is about 65W and this is really more difference than what I would expect between an i5 750 and an i5 661.
If you measured it by running a game the result can be biased towards the Clarkdale because it is slower and thus imposes less load on the GPU.
mathew7 - Thursday, May 6, 2010 - link
You have to remember that he's comparing SYSTEM load. The 661has 87W TDP with GFX while 750 has 95W TDP without GFX. Since the integrated GFX is not used, the difference is even higher. Also, since the MBs are different, they also have different power consumptions.zipzoomflyhigh - Wednesday, May 5, 2010 - link
Why use a cpu with integrated video and a H55 and then drop in a HD5850??Makes no sense whatsoever. Your paying a premium for a cpu with built in graphics for what?
ClagMaster - Wednesday, May 5, 2010 - link
The Clarksdale is really meant for HTPC and Office PC applications. Here the on-die GPU will be used.However for HTPC or Office PC, I believe you would get much better value and good performance with AMD Athlon X2 and 890G motherboard .
I agree I would have preferred a dual core processor without the GPU. Power consumption would be better.
strikeback03 - Thursday, May 6, 2010 - link
If the GPU isn't being used it won't consume power. And given that these are the only dual-core 32nm processors around I imagine they are still the most power-efficient ones around (short of Atom/CULV, which are far less powerful).BernardP - Wednesday, May 5, 2010 - link
Intel could reduce prices to a more competitive level and reach users not interested in its integrated graphics by offering the Clarkdale architecture in a lineup of CPU-only chips.Give us a CPU-only version of the i5-680 @ 3.6 GHz with 4,0 GHz Turbo and I will be glad to add my own graphics card to build a nice all-around machine for multimedia and casual gaming.