
Sandy Bridge-E has little trouble jumping to the top of the Arithmetic test, ahead of Intel’s outgoing -990X. Given Sandra 2011’s synthetic nature, it’s no surprise to see it exploiting all aspects of these eight processors.


Using SSE 4.1 (integer) and 2 (floating-point), Core i7-3960X slides right past Core i7-990X for the number one spot. Those figures improve dramatically with the implementation of AVX, though.

One of the things I noticed in Intel Core i7-3960X (Sandy Bridge-E) And X79 Platform Preview was that Sandy Bridge-E enabled significantly better AES256 bandwidth than Gulftown or Sandy Bridge. That advantage persists in the C1 stepping, nearly doubling Core i7-2600K’s result in the Cryptography benchmark. Intel confirms that it made changes to enhance AES throughput, but doesn’t expound on what it did.
Not satisfied, I did a little digging and started pulling memory modules. With three channels of memory, Sandy Bridge-E achieves 8 GB/s AES256 bandwidth. Two channels facilitate 5.43 GB/s. And a single channel of memory installed yields 2.72 GB/s. It seems that AES-NI is very much constrained by throughput (given that it's accelerated in hardware, and consequently very easy to execute), so it looks like the changes Intel suggested are tied to its memory controller, rather than its AES-NI implementation.
Hoping for some correlation to real-world performance (and a reason to get more excited about four 64-bit channels on the desktop), I ran a few tests in the latest stable build of TrueCrypt using the built-in benchmark and a 1 GB buffer. Despite a mean result of 3.8 GB/s in single-channel mode and 5.2 GB/s using two channels, performance fails to scale beyond that, indicating a bottleneck other than the speed at which the processor can encrypt and decrypt data.

The memory bandwidth advantage of a quad-channel DDR3-1600 bus is incredibly evident in Sandra 2011, which manages to realize around 37 GB/s from a potential 51.2 GB/s theoretical maximum.
Impressive though that number is, keep it in context. Sandy Bridge, with its dual-channel DDR3 memory controller, already showed that it wasn’t particularly starved for memory bandwidth in most desktop software. Practically, there won’t be many apps able to exploit those big throughput numbers. Perhaps that’ll change in the first quarter of next year when Sandy Bridge-E turns into Xeon E5 for dual-socket servers.


- Say Hello To The PC Hardware Trophy Wife
- Quad-Channel Memory And PCI Express 3.0
- X79 Express: P67, Is That You?
- Cooling And Overclocking Core i7-3960X
- Test Setup And Benchmarks
- Benchmark Results: PCMark 7
- Benchmark Results: 3DMark 11
- Benchmark Results: Sandra 2011
- Benchmark Results: Content Creation
- Benchmark Results: Productivity
- Benchmark Results: Media Encoding
- Benchmark Results: Crysis 2
- Benchmark Results: DiRT 3
- Benchmark Results: World Of Warcraft
- Crysis 2 In SLI
- DiRT 3 In SLI
- World Of Warcraft In SLI
- Battlefield 3 In SLI
- Power Consumption
- Core i7-3960X Versus Core i7-990X
- Core i7-3960X Versus Core i7-2600K/Core i5-2500K
- Core i7-3960X Versus FX-8150
- A Symbolic King In A Crowd Full Of Value
The funny thing is that cores don't scale well. They do, but it's far from ideal as the percentages from the 2600K show (and the FX-8150 but that's a different story).
But the takeaway:
-If you're playing games the i5-2500K is the best purchase you can make and it's enough for Tri-580 SLI. Only WoW shows any difference, but most games ignore it.
-X79 is Intel being just plain lazy. No matter how you slice it- the X79 should have been called X67 and left like that. It's also a wildcat platform that will only support at most 6 CPUs that aren't terribly crippled.
-A Phenom II 955BE (or unlocked 960T, or a 1090T/1100T) is still a fine CPU to have unless you're gaming with dual graphics cards or doing time-intensive tasks.
What we have today is simply a platform for bragging rights not a serious contender to the X38, X48, X58 family.
I would LOVE to see them pick up their game and provide me with a worthy upgrade over my 4GHz i7 2600 (Non-K). I would swoop it up.
Look, BD had 4 modules with two "cores" each, each module is equivalent to a Sandy Bridge core.
They should just combine both of those cores or make them a single core, so we get 4 threads.
Then create 4-6-8 core versions of those CPU's..
Think about it.. the FX8150 is more of a 4-core CPU where the resources are halved pretty much so you get two threads per core, it would have been MUCH MUCH better if they just kept 4 strong cores.
Not sure why either but I always seem to start an AMD related comment :\
The labels are wrong on the graphs on this page the last ones should read DDR2-2133 on the last two shouldn't it?
JeanLuc
The only use for the 3820 really seems to be a cheap placeholder processor if you need a new PC now, but want to wait for a likely full 8c/16t version to come out around the time Ivy Bridge is released. The 3930k should prove to be a very good high end gaming/ mid range workstation part though for people who invest close to $1k in graphics cards.
The funny thing is that cores don't scale well. They do, but it's far from ideal as the percentages from the 2600K show (and the FX-8150 but that's a different story).
But the takeaway:
-If you're playing games the i5-2500K is the best purchase you can make and it's enough for Tri-580 SLI. Only WoW shows any difference, but most games ignore it.
-X79 is Intel being just plain lazy. No matter how you slice it- the X79 should have been called X67 and left like that. It's also a wildcat platform that will only support at most 6 CPUs that aren't terribly crippled.
-A Phenom II 955BE (or unlocked 960T, or a 1090T/1100T) is still a fine CPU to have unless you're gaming with dual graphics cards or doing time-intensive tasks.
Yessir! Working on it now!
Yes. Its expensive. In other news the Earth orbits the Sun. I wish I had enough $$$ that the costs of this CPU was inconsequential to me.