CPU Charts 2012: 86 Processors From AMD And Intel, Tested

[Nintendo Maniac 64]

I have a 2.8GHz P4 that struggles even with 480p on youtube, but that's with an x1600 Pro graphics card so no youtube acceleration, which is probably why yours can play it.

I have a 2.0GHz Northwood that handles YouTube 480p just fine, and it's only using an old PCI Radeon 7000. (NOT HD 7000) I think your OS may have too much crap running in the background or something.

Heck, if I download YouTube's 720p WebM encodes, I can even play them just fine through VLC. (MP4/h.264 is too much for the CPU)

 

[shaun_shaun]

where is i3 3220/i3 2120 ???

 

[falchard]

I find that AMD actually stepped up a bit this year. Honestly, I thought it would be a much bigger thrashing then it was. Compared to when Bulldozer was first introduced against these very same chips, the performance of the Bulldozer parts seems to have improved. In most workloads you are not looking at too bad a performance hit compared to Intel. Now the PileDriver based chips would have been nice to see in this article since there are now workloads that prefer a Piledriver chip over an Intel chip such as 3D rendering.

 

[jonjonjon]

why is the 2500k ram running at 1333 and the 3570k is at 1600? my ivy bridge defaulted to 1333 and only ran at 1600 if i manually set it or enabled the XMP profile. the same is true for my sandy bridge. seems unfair considering sandy bridge can just as easily run ram at 1600 as ivy bridge can. how about apple to apple not apple to oranges.

 

[Teeroy32]

[citation][nom]tipoo[/nom]I have a 2.8GHz P4 that struggles even with 480p on youtube, but that's with an x1600 Pro graphics card so no youtube acceleration, which is probably why yours can play it. Bioshock 1 was even borderline playable on it, but I would not reccomend the experience.[/citation]
My 3.4 precott with a HD6670 plays Modern Warfare really well, and Plays Doom 3 under wine on ubuntu silky smooth, yet 1080 mkv bring it to its knees

 

[pit_1209]

[citation][nom]dgingeri[/nom]It's funny. People accuse Intel of changing the sockets too often, yet, if you look at the same span of time, AMD has gone through just as many sockets as Intel:Intel: 775, 1156, 1366, 1155, 2011AMD: AM2+, AM3, AM3+, FM1, FM2At least Intel kept the value range server socket and the mainstream desktop socket the same. AMD didn't do that much.[/citation]

C´mon, so you have wasted seconds on you´re life trolling about something that nobody was talking about?

AM2+ and AM3 were retrocompatible like AM3 and AM3+ so it´s not that many sockets at the end and you don´t have to change sockets to go high end like intel as you didn´t know 1366 and 2011 were in the same gen with 1156 and 1155 that´s 4 sockets in 2 generations.

 

[Mckertis]

I have a 2.8GHz P4 that struggles even with 480p on youtube, but that's with an x1600 Pro graphics card so no youtube acceleration, which is probably why yours can play it.

Um, once my GF6600 cooler dropped dead, and while i was buying a replacement i could easily watch Youtube in 480p on my antique GF2MX. Quite sure it doesnt have anything in terms of "youtube acceleration", so it seems to be your particular system problem.

Heck, if I download YouTube's 720p WebM encodes, I can even play them just fine through VLC. (MP4/h.264 is too much for the CPU)

It's not, if you use Core codec and any player that is actually good (KMP, MPC). Unfortunately these days all video is switching from 8-bit to 10-bit, and 10-bit is more than P4 can handle.

 

[JonnyDough]

[citation][nom]amdfangirl[/nom]Sometimes I wish you updated legacy CPUs like the Core 2 Duo or even perhaps the Athlon 64 X2 series, just one or two models so that people upgrading can have an idea how much faster the CPU is in relation to their new purchase.[/citation]

The problem is that an old FX-60 can compete with a low end Core 2 Duo and with the economy going south, the desktop dying, etc. online hardware websites are trying hard to help push sales to stay relevant. Imagine a processor 8 times more powerful than today's Core i7 that uses only fractions of a watt. Would you really need to upgrade your system? It's just a matter of a few years away.

 

[army_ant7]

On the "Sandy Bridge" page, I believe the i7-3960X has 15MB of L3 cache. On the "Ivy Bridge" page, I believe that "tick" and "tock" are switched in all the statements there. Also, it would've been nice to see at least one i3 on either the Sandy or Ivy pages. Not that I blame you guys or anything.

I see that you've already integrated a Piledriver CPU (FX-8350). Now I think an update/edit to the text parts on multiple pages of the article are in order. Also, I didn't see the FX-8170 in the power consumption charts.

It's peculiar how the G630T performed that badly in the "Creativity" test of PCMark, especially when it's Pentium brethren are a little ways up. I was expecting the hexa-core Sandy Bridge-E to top at least one chart in PCMark... I wonder why it didn't.
Were the Nehalem/Westmere IMC's that bad that in triple-channel (I assume), they lose to the FX's in memory bandwidth (SiSoftware Sandra)? Unless it's the RAM modules' fault...

I'm not sure, but Crysis 2 seems to want a more powerful GPU even for those low settings (a GPU bottleneck maybe). What's up with those low Mafia II (FPS) numbers? Is PhysX turned on or something? Because if it is, it's not indicated on the "Benchmark Hardware and Software" page. But, if I'm not mistaken, isn't the CPU PhysX implementation their bad (single-threaded)? Maybe you could've tried Metro 2033's which I heard is multi-threaded.

The raw data makes it pretty clear that Intel's Sandy Bridge, Sandy Bridge-E, Ivy Bridge, and Nehalem micro-architectures are clearly faster than AMD's highest-end models.

This statement seems too misleading to me. Maybe because it doesn't mention whether the author meant in terms of per clock per core, or what. It also seems to forget how AMD's FX's can trump Nehalems at some tests, sometimes even Ivy and Sandy Bridges (though something about this is mentioned somewhere below in the same page).

 

[thegreatsquare]

Please add a few C2D & C2Q processors. I really would like to get an idea of what my Q9450 @ 3.2GHz is going to do with a newer GPU

Maybe add 4 older Intels:
-Q9770 or Q9650
-E8600 or E8700
-Q6600 or Q6700
-E6750 or E6850

 

[jtenorj]

Is the decimal point in the wrong place for ALL the Mafia II results?

Edit: should there even be a decimal point with those numbers?

 

[dcointin]

I appreciate the work that went into benchmarking and compiling this information, and realize that they were open about not having the opportunity to include certain product lines. Without the inclusion of AMD's Piledriver based chips and Intel's i3, however, I how question valuable the information really is.

 

[Outlander_04]

Why benchmark games at such low resolution and detail settings? If you want to indicate actual user experience try 1080p resolution and high detail settings .Thats where we want to play

 

[NuclearShadow]

I have a 920 and a 950 rig in my house and I am quite amazed by this to see that they are still preforming quite well compared to newer CPU's. There is simply no reason for me to upgrade even after all this time. The 920 is 4 years old now and I cannot think of a single instance in the past that I waited over 4 years to make a new build.

 

[butremor]

Really looking forward to seeing missing CPUs benches.
Also, have you measured A10 5800k graphics performance with dedicated gpu or purely integrated performance?

 

[edlivian]

I feel like toms was paid to leave out i3 ivy and sandy.

They are the best dual cores out there, and they were left out??

 

[Outlander_04]

[citation][nom]cheryljansen8[/nom]the power consumption going from AMD 955 to a Core i5 3570K. ... http://www.HappyHolidaysWithGoogle [...] O2m5pYEaSY[/citation]

My antivirus says thats a malicious link . Id urge people not to click it

 

[JebbyC]

[citation][nom]Outlander_04[/nom]Why benchmark games at such low resolution and detail settings? If you want to indicate actual user experience try 1080p resolution and high detail settings .Thats where we want to play[/citation]

So that there's no GPU bottleneck. The comparison should remain valid; if you want to see if a particular chip will run the game you want at the settings you want, read the review of the chip.

 

[Outlander_04]

[citation][nom]JebbyC[/nom]So that there's no GPU bottleneck. The comparison should remain valid; if you want to see if a particular chip will run the game you want at the settings you want, read the review of the chip.[/citation]


The comparison is meaningless for most people . Monitors refresh at 60 Hz , which is identical to 60 FPS .
A benchmark showing 180 fps compared to another processors 120 FPS doesnt show that processor to be 50 % more powerful at gaming since the results dont scale linearly , and it doesnt show anything about the user experience at all since both would be monitor limited .

 

[chaostheory_22]

Great benchmarks.
But i want some processors which were legendary overclockers, and representatives of their generation of CPU's, included with a nominal OC :

intel C2D E7300 : 2.66- > 3.33
Intel C2Q Q6600 : 2.4- > 3.0ghz
Intel i5-750 : 2.66 - >3.33

Its highly likely that a person has owned at least one of these CPU's. I want to know how well these compare to modern processors.

Hello,

Just what I thought. Look what polish PCLab did. They've completely destroyed toms in terms of everything tbh:

http://pclab.pl/art50000.html

Enjoy

 

[army_ant7]

Really looking forward to seeing missing CPUs benches.
Also, have you measured A10 5800k graphics performance with dedicated gpu or purely integrated performance?

As I see it, they did it using the dedicated GPU listed, so as to compare pure CPU performance.

The comparison is meaningless for most people . Monitors refresh at 60 Hz , which is identical to 60 FPS .
A benchmark showing 180 fps compared to another processors 120 FPS doesnt show that processor to be 50 % more powerful at gaming since the results dont scale linearly , and it doesnt show anything about the user experience at all since both would be monitor limited .

For one thing, showing that a certain CPU has the ability to churn out more FPS than others shows that it would probably be able to do so with more CPU-intensive games. For example, if "CPU A" put out 120FPS and "CPU B" put out 60FPS in a certain game, it could mean that "CPU A" would put out 60 FPS in a more CPU-intensive game, and "CPU B" would only be able to put out 30 FPS. The point of this article is to compare CPU's--rate them relative to each other. That info is pretty valuable for system builders who'd want the best CPU they can get, I would think.

Also, I think it's safe to say that "CPU A" above is more powerful. A person who can lift 100Kg is more powerful than a person who can only lift 50Kg, wouldn't you say? When you watch them lift up a dinner fork, you'd see that there're no practical difference between them (because their "powers"/strengths are more than enough for that task), but when it comes to more demanding tasks, you may see who's more powerful.

 

[The_Trutherizer]

The more the better. Keep it up TH!

 

[klewlis1]

It would be nice to see the 960T from AMD tested once in awhile.

 

[Outlander_04]

For one thing, showing that a certain CPU has the ability to churn out more FPS than others shows that it would probably be able to do so with more CPU-intensive games. For example, if "CPU A" put out 120FPS and "CPU B" put out 60FPS in a certain game, it could mean that "CPU A" would put out 60 FPS in a more CPU-intensive game, and "CPU B" would only be able to put out 30 FPS. The point of this article is to compare CPU's--rate them relative to each other. That info is pretty valuable for system builders who'd want the best CPU they can get, I would think.

What you have failed to account for is that the relationship is not linear . Depending on the game engine a difference in cpu power of less than 5% could double fps . Worse still is that running higher resolution and image settings loads the processor in a different way .
And the real world is more complicated . Computers are never just running a game benchmark . If you are playing online you will have firewall ,antivirus and network services wanting cpu cycles . But the way these benchmarks are run makes no account of that .
Anyone who believes a dual core i3 can match an i5 at the same clock speed is going to wind up very disappointed ... despite the results of many of these benches

 

[mayankleoboy1]

Hello,

Just what I thought. Look what polish PCLab did. They've completely destroyed toms in terms of everything tbh:

http://pclab.pl/art50000.html

Enjoy

Thanks. This is freaking awesome tests done.