Part 2: How Many CPU Cores Do You Need?

[cleeve]

In part two of this series, we'll address some of the issues raised in part one, such as the effect of shared cache between CPU cores, how the number of cores affects multitasking performance, and comparing the Phenom II to the Core 2 Quad.

Part 2: How Many CPU Cores Do You Need? : Read more

 

[erdinger]

Very intresting article,now I'm even happyery I bought a Phenom II 720 for my gaming rig!

 

[icepick314]

"In any case, there are two lessons to be learned here: first, try to avoid a virus scan during your gaming sessions."

what kind of PC gamer does virus scanning while running a game?

 

[KyleSTL]

Why no power consumption testing? I was a little curious what disabling cores in the OS would do to power consumption under load. A little let down, but otherwise good article. It's good to see a scaling article at least yearly since people refer to the dual/quad debate so often and often the tests that were run within article that are referenced are out of date and irrelavent.

 

[Onus]

Good article, and very interesting.
Now I really hope I can unlock the 4th core when my 720BE arrives (hopefully later this afternoon), but I won't sweat it.
Did you happen to test if it made a difference what scan priority was set in AVG? I'd really like to see those numbers.

 

[Guest]

So, how did you manage to get an Nvidia-based graphics card (Gigabyte GV-N250ZL-1GI 1 GB DDR3 PCIe) up and running with the ATI Catalyst 9.6 drivers?! ;-)

Besides that bit of confusion, thanks for the benchmarks!

 

[1word]

very happy with my 720 BE. I constantly check with the activity on the cores, and many many apps use all three cores, or multi- tasking uses all the 3 cores. some activities like defrag uses only 2 cores. image editing software, and general applications like browsers, office apps use all three cores, especially when multi tasking.

i'm very happy with the AMD 720BE.

 

[jcknouse]

[citation][nom]KyleSTL[/nom]Why no power consumption testing? I was a little curious what disabling cores in the OS would do to power consumption under load. A little let down, but otherwise good article. It's good to see a scaling article at least yearly since people refer to the dual/quad debate so often and often the tests that were run within article that are referenced are out of date and irrelavent.[/citation]

I liked the article well, but I was too finding myself asking "What was more power efficient? the PII x2 550 BE or the PII x2 955 BE?

Would love to know, even if it was just that you guys just happened to glance at a P3 Kill-a-watt or some other meter you had inline during testing or something.

Thanks for great work, guys

 

[erichlund]

It's true that an application like iTunes does not benefit from multiple cores, when run without any other apps. However, it also doesn't compete for more than one core when multiple apps are running, so single threaded apps also benefit from multiple cores when users are multi-tasking.

What one really needs to know with iTunes and it's competing applications is: Which one competes most efficiently in a multi-processing environment? In other words, which uses the least resources while performing essential tasks, leaving the most resources for the other tasks being performed? To say it in perhaps the clearest way, what applications play well with other types while multi-tasking, and which hog resources, making it more difficult to multi-task?

That's not really the point of this test, but it may lead to some interesting future evaluations.

 

[Onus]

^Yes, that's why it would be interesting to see if (and how much) the impact varied if AVG was set to slow, normal, or fast for its scan priority.

 

[paranoidmage]

You shouldn't test the games at 1024x786 at low details. These benchmarks are supposed to simulate actual usage. No one will actually run games at that resolution and detail unless their computer is a dinosaur. If you want to remove bottlenecks, use a better GPU like a 4890.

How do I know if multiple core will actually help me? I run games at 1920x1200 with med-high details.

 

[erichlund]

[citation][nom]paranoidmage[/nom]You shouldn't test the games at 1024x786 at low details. These benchmarks are supposed to simulate actual usage. No one will actually run games at that resolution and detail unless their computer is a dinosaur. If you want to remove bottlenecks, use a better GPU like a 4890. How do I know if multiple core will actually help me? I run games at 1920x1200 with med-high details.[/citation]
That was not the point of the test. They only really wanted to understand the scalability of multiple cores, so taking the graphics out of the equation helps. We see that multiple cores helps the games.

You can do this same thing with your game of choice. You can set your OS to run the game with different numbers of cores, and see how you games are impacted at your chosen resolution. They've given you the technique, just apply it.

We now also understand explicitly (what we already implicitly understood), that we should run games by themselves, because background processes will impact our performance (there's a duh in there somewhere).

The work they've left on the table is for application users. Picking the right set of applications that work well together in a multiprocessing environment could be interesting. For instance, does Sun's Open Office, originally developed as Star Office in a Unix multiprocessing environment, scale better than MS Office? Which is more heavily impacted by background processes? The fact that's it's free (assuming you have the ability to download it) is an added benefit, but not directly the issue.

 

[2shea]

if you go and use a high res then the results will depend far more on the gpu rather than cpu, which is what is the core of this article... If they do it at low level then it will be done at high level because high level details depend on the gpu far more.

 

[masop]

Great article. The benchmark results with avg running alongside wic are interesting. Although I don't have virus scans running while I do my gaming, I do have a crap load of apps and utilities running on my system at any given time, most idle, but still occupying resources. I think it is about time to upgrade my dual core to a quad setup. Aside from trying to justify the upgrade costs to my wife (LOL) and not get murdered in the process, I'm considering it all but done, hehe.

 

[chaohsiangchen]

For gaming performance, Tom's should also use games that support quads such as GTA4, HAWK, FarCry2 or Empire:Total War. Other review sites has shown that games optimized for quads perform better with more cores than higher clocks.

 

[volks1470]

yea, my 720BE @ 3.6GHz + a GTX 295 works pretty well with Crysis. I just wish the damn processor would overclock to 4GHz, but I just can't on air

 

[cleeve]

[citation][nom]paranoidmage[/nom]You shouldn't test the games at 1024x786 at low details. These benchmarks are supposed to simulate actual usage. [/citation]

The problem with that is, the graphics card will bottleneck the results far more than the CPU cores, so we won't learn as much by high-res testing.

With low-res, we know where the number of cores becomes a limitation.

 

[cleeve]

[citation][nom]bogan[/nom]So, how did you manage to get an Nvidia-based graphics card (Gigabyte GV-N250ZL-1GI 1 GB DDR3 PCIe) up and running with the ATI Catalyst 9.6 drivers?! ;-)[/citation]

Ouch! Thanks for pointing that out, fixed!

 

[cleeve]

[citation][nom]jcknouse[/nom]I liked the article well, but I was too finding myself asking "What was more power efficient? the PII x2 550 BE or the PII x2 955 BE?Would love to know, even if it was just that you guys just happened to glance at a P3 Kill-a-watt or some other meter you had inline during testing or something.Thanks for great work, guys[/citation]

From the glances I took, the Phenom X4 used more power than the X3 at the same frequencey, even if the OS was only using 2 cores. So in my mind there wasn't much of a story there, but it was shortsighted of me not to share that in the article.

Good suggestion though, next time I look into this I'll definitely supply some formal power usage charts...

 

[Proximon]

Hopefully we can do the same for Windows 7. Great article.

 

[Aerobernardo]

Amazing article.

I would just like to point out that one of my concerns about having less then 4 cores is the scalability/performance with Xfire or SLI. That point is missing on this article, but I do believe it's just a minimal result that is not hte focus fere.

 

[bk420]

1

 

[rambo117]

[citation][nom]erdinger[/nom]Very intresting article,now I'm even happyery I bought a Phenom II 720 for my gaming rig![/citation]
hehe, same here

 

[apache_lives]

again with the prehistoric benchmarks - styled in 1999, not for 2009

if i look down at my task bar as i write this i have 2 tv shows recording, msn messenger, torrents, antivirus, video playing, and god knows whatever else i am running IN A NORMAL SYSTEM NOT A CLEAN FRESH INSTALL

i also love playing gta iv online which in its self loads up all the cores

THERE IS NO WAY A DUAL CORE IS CLOSE TO A QUAD IN THE REAL WORLD AS I CAME FROM AN E6600 TO THIS Q6600

 

[xaira]

really interesting article, i second the power usage thing it would have been informative to see the differences, as i always say, get more shaders and get more cores, i dont understand y the test with the crippled 955 vs the x2 550 was done tho, the x2 550 is a native crippled quad, if it was a native dual id understand, thx tom as always for the interesting read.