Sign in with
Sign up | Sign in

Results: Synthetics

The Core i7-4770K Review: Haswell Is Faster; Desktop Enthusiasts Yawn
By

This one’s mainly interesting for the Physics benchmark, a pure measure of CPU performance. Core i7-3930K asserts itself early on, with quad-core Haswell, Ivy Bridge, and Sandy Bridge chips taking the next three spots. In the case of FX-8350 and, to a greater extent, A10-5800K, lower host processor performance brings down the Overall suite score, represented by the red bar.

Back when I published Core i7-4770K: Haswell's Performance, Previewed, a lot of folks (even some from within Intel) tried claiming our sample wasn’t representative of final performance. A quick comparison shows that, while we’re seeing very similar SSE3-based numbers, the Core i7-4770K’s advantage over -3770K is actually smaller in this latest version of Sandra, despite Haswell benefiting from AVX2 support.

Sandra’s Multimedia benchmark generates an image of the Mandelbrot Set fractal using 255 iterations for each pixel, representing vectorised code that runs as close to perfectly parallel as possible.

Our AVX2-based results from Core i7-4770K almost match the preview piece’s exactly, while the AVX-accelerated Ivy and Sandy Bridge numbers are close too. We now see that AVX2 helps a four-core Haswell part outperform Sandy Bridge-E’s six cores in AVX-optimized code.

When it comes to floating-point performance, the AVX code that runs on Core i7-3770K and -2700K matches pretty closely. But -4770K’s FMA3 path isn’t as consistent. Floats are faster, while doubles slow down quite a bit. The scaling is much closer to what we were expecting when we were testing for the preview.

Sandy Bridge-E takes off like a bat out of hell in the Encryption/Decryption module due to the extreme memory bandwidth you’ll see in the next chart, which feeds the CPU cores data quickly.

The low memory bandwidth issues affecting our early motherboard are mostly worked out, and we now see Core i7-4770K delivering competitive throughput in the AES-NI-accelerated workload, along with strong hashing performance.

These bars could have been predicted based on the AES256 results above, but here they are.

L1D, L2, and L3 cache bandwidth are all up on our Core i7-4770K compared to my preview piece. However, only the L1D yields the doubling of bandwidth we were expecting. Given 64 bytes/cycle (compared to 32 for Ivy Bridge), this number should still be much higher than it is. There’s still no good explanation for that outcome.

Ask a Category Expert

Create a new thread in the UK Article comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 20 comments.
This thread is closed for comments
  • 1 Hide
    phil_livesey , 1 June 2013 16:56
    Yet another competition not open to UK residents, despite this being on the UK site. Thankyou Tomshardware!!!!
  • 0 Hide
    Pailin , 1 June 2013 18:06
    Sounds... disappointing after hearing about the Uber high LN OCs.
    Also thought I recently heard somewhere of others getting Nice 5GHz+ OC's on water and Very low vcore's - perhaps you guys have a Poor batch?
  • 0 Hide
    das_stig , 1 June 2013 19:46
    So once again, the UK site publicises a competition as part of a review and once again, it's limited to the US readers. Come on TH tis just sucks and shows that you don't give a toss about the rest of the world. The internet is a global community, it doesn't stop at the US border. Surely your programming monkies can code a conditional statement to blockout content if outside the U.S of Ass or are they to busy coding up more ad boxes ????
  • 0 Hide
    daglesj , 1 June 2013 21:01
    Guys, you are posting this on the useless and superfluous UK version.
    No one will be reading it.
  • 0 Hide
    mi1ez , 2 June 2013 11:01
    Quote:
    the only arrangement emerging today is the quad-core SoC

    SoC is pushing it a bit given it doesn't contain RAM, USB, network, etc.
  • 0 Hide
    bwrlane , 2 June 2013 14:08
    Great article, but what a let down. Ivy Bridge was already a disappointment, but Haswell is far more disappointing even than that. My 2700k based PC can manage 4.84GHz under full load, while remaining stable. My 3770k PC can do 4.74GHz. The IPC improvements are almost exactly compensated for by the lower attainable clock speed.
    But with Haswell, the world has gone backwards. Apparently, a 4770k can be pushed to 4.4GHz and that's it. That's a 7% reduction in clock speed. Since most benchmarks don't show a 7% improvement at stock, Haswell is slower than the Ivy Bridge that it replaced.
    For years we've been hearing that the answer to all our tech questions is "you have to wait for Haswell for that". But as this article shows, that was a lot of hot air.
  • 0 Hide
    doive1231 , 2 June 2013 19:33
    What a load of baloney we were spun about Haswell graphics. I read it would be a real step change and we should expect something special. Well, the comparisons between HD4000 and HD4600 are a real kick in the teeth. Nothing more that a few frames per second in most tests. Bah, Intel. And not I won't be getting Crystalwell.
  • 0 Hide
    selfmade_exe , 2 June 2013 21:45
    amd trinity gets bottlenecked by 1600 ram!!!
  • 0 Hide
    JRAtk94 , 2 June 2013 23:42
    Yay for AMD.
    Finally, AMD wins on price AND performance.
  • 0 Hide
    selfmade_exe , 3 June 2013 22:22
    TO ALL THOSE RESPONSIBLE FOR THIS TEST: I DARE YOU "CLEVER" TESTERS TO TEST AMD TRINITY WITH 1866 RAM OR ABOVE, AND ALSO COMPARE WITH RICHLANT FLAGSHIP. ALSO, PLZ RUN GPU_CAPS_VIEWER -> THE JULIA SET WITH GPU OPENCL AND COMPARE WITH TRINITY! PERIOD!
  • 0 Hide
    Tedfoo25 , 5 June 2013 08:34
    I'm quite impressed actually. With a decent overclock, you are essentially getting a 3930k on the cheap with a socket that will have CPUs developed for it in the future.
  • 0 Hide
    mateau , 11 June 2013 15:22
    Haswell is old tech already. AMD just rekeased FX-9590 a 5 ghz 8 core MONSTER that is sure to stomp all over Hasbeen. And the FX-950 is unlocked for overclocking!!!!!
  • 0 Hide
    Alps , 11 June 2013 19:26
    "several strides behind"?? I'll but an AMD FX-8350 and overclock it, thanks for the usual biased review!!
  • 0 Hide
    Alps , 11 June 2013 19:26
    "several strides behind"?? I'll but an AMD FX-8350 and overclock it, thanks for the usual biased review!!
  • 0 Hide
    Alps , 11 June 2013 19:26
    "several strides behind"?? I'll but an AMD FX-8350 and overclock it, thanks for the usual biased review!!
  • 0 Hide
    Alps , 11 June 2013 19:26
    "several strides behind"?? I'll but an AMD FX-8350 and overclock it, thanks for the usual biased review!!
  • 0 Hide
    Alps , 11 June 2013 19:29
    "several strides behind"?? I'll but an AMD FX-8350 and overclock it, THANKS for the usual BIASED review!!
  • -1 Hide
    genz , 16 June 2013 19:49
    Why are we comparing a dual module CPU to a quad core/8 thread one? Shouldn't you be using the quad module/8 core AMD chip?
  • 0 Hide
    genz , 16 June 2013 19:49
    Why are we comparing a dual module CPU to a quad core/8 thread one? Shouldn't you be using the quad module/8 core AMD chip?
  • -1 Hide
    genz , 16 June 2013 19:51
    Why are we comparing a dual module CPU to a quad core/8 thread one? Shouldn't you be using the quad module/8 core AMD chip?