Now that Piledriver-based CPUs and APUs are widely available (and the FX-8350 is selling for less than £160), it's a great time to compare value-oriented chips in our favorite titles. We're also breaking out a test that conveys the latency between frames.
At least on the desktop, dual-core processors rarely helped bolster performance when they were first introduced. Most mainstream apps simply hadn't been optimized for multiple cores; that sort of technology was principally enabled in the server and workstation space. You had multi-socket motherboards with single-core chips cranking on complex problems in parallel. But games were almost exclusively written to run on a one core.
Programming with threading in mind isn't easy, and it took developers years to adapt to a world where CPUs seemed destined to improve performance through parallelism rather than then 10 GHz clock rates Intel had foreshadowed back in 2000. Slowly, though, the applications most able to benefit from multiple cores working in concert have been rewritten to utilize modern hardware.
Want proof? Just have a look at our benchmark suite. We test something like two pieces of software that are still single-threaded: Lame and iTunes. Everything else, to one degree or another, is threaded. Content creation, compression, and even productivity apps tax the highest-end four- and six-core CPUs.
Games, on the other hand, have taken longer to "get there." With a primary emphasis on graphics performance, it's not surprising that single-threaded engines still exist. However, spawning additional threads and utilizing a greater number of cores allows ISVs to implement better artificial intelligence or add more rigid bodies that can be affected by physics.
Increasingly, then, we're seeing more examples of games exhibiting better performance when we use quad-core processor. They're still the exception though, rather than the rule. And that's why the great single-threaded performance of Intel's Sandy Bridge architecture (and later Ivy Bridge) dominated most of our processor-bound game testing. Back in the day, dual-core Pentiums went heads-up against quad-core CPUs from AMD, and came out in the lead.
It's now clear that gunning for higher and higher clock rates is not the direction AMD and Intel are going. They're both building desktop-oriented CPUs with as many as four modules (in AMD's case) or six cores (in Intel's). In turn, game developers continue getting better about utilizing available on-die resources. We're clearly at a point where you need at least a dual-core CPU to enjoy today's hottest titles, if for no other reason than sticking with a single-core chip would put you about eight years back in processor technology. But is there a reason to skip over the dual-core models and jump right into the world of gaming on a quad-core CPU?
That's what we're hoping to answer today, and we have a new tool to help us.

Now i wonder how there's practicly no difference between the Zambezi and Vishera CPU's.
No, no, no, no, NO!
All the values (of latency in this case) are set on a range centred on the 50th centile (where you can expect 50% of the population (of data) to be below and 50% above. At the 75th centile value 75% of your population will be below that value and 25% will be above. At the 95th centile 95% of the population will be below and only 5% above. This works for the 5th and 25th centiles likewise. The metric you should be focusing on is the range. i.e. what is the spread of the data. For example, for visible micro-stuttering a graphics setup would manifest with a large range between the 25th and 75th (often called the inter-quartile range, within which half of your data resides) and an even larger range (obviously) between the 5th and 95th centiles. This wide spread would show the large variation in rendering times for consecutive frames.
On a system where there is little or no micro-stuttering you would expect the inter-quartile and 5th-95th ranges to be narrow indicating very regular frame rendering times. Further to this, if the inter-quartile range is small but the 5th-95th is large that would indicate that most frames are rendered well and consistently but there are occasional large out-liers. These out-lier values would be perceived as micro-stuttering as well.
Please Don, when you delve into statistical analysis, be precise with your descriptors. Otherwise it detracts from your, clearly extensive, knowledge and experience in the computing world.
Q.
Edit for grammar.
which is a good thing for the consumer as it will stop intel from becoming complacent. fanboi it all you want, but we need competition in the market or the end user suffers.
What would the results have been if you used a little 6670 with the CPUs? Would the AMD CPUs have performed better because of the ability to Crossfire with their APUs where appropriate?
Disclaimer: I am running an Intel E6750 @ 2.66GHz and an AMD 6670 at home on my gaming PC and I'm happily gaming all my games at maximum detail levels at the maximum resolution my monitor supports which is 1280 x 1024.
This has already been done.
http://www.tomshardware.co.uk/fx-4100-core-i3-2100-gaming-benchmark,review-32384.html
This is a sub $200 comparison and the i7-930 and the i7-920 that it replaced were both around the $300 mark. It's well documented that those CPUs offered further performance gains over the i5 chips but at a diminishing value return creating a lower bang per buck ratio.
That makes no sense. You can't say there was not much difference AND that the 8350 is an improvement while it is still severely lacking against CPUs of the same price and is often matched if not beaten by the much cheaper i3 chips.
They use the GTX680 to remove any possibility of performance bottlenecks associated with a GPU so we get true CPU results.
The only reasons these days to buy an AMD CPU is to upgrade your current AMD system or to build a budget gaming rig or a HTPC using one of their latest APUs without using a dedicated GPU. If you are building a new system with a dedicated GPU then Intel is the obvious choice.
Ok, perhaps something cheaper from the same generation/architecture.