Office Performance at 3 GHz

All of our benchmark results can also be found in our benchmark engine, Bench.

Agisoft Photoscan – 2D to 3D Image Manipulation: link

Agisoft Photoscan creates 3D models from 2D images, a process which is very computationally expensive. The algorithm is split into four distinct phases, and different phases of the model reconstruction require either fast memory, fast IPC, more cores, or even OpenCL compute devices to hand. Agisoft supplied us with a special version of the software to script the process, where we take 50 images of a stately home and convert it into a medium quality model. This benchmark typically takes around 15-20 minutes on a high end PC on the CPU alone, with GPUs reducing the time.

Agisoft PhotoScan Benchmark - Total Time

Looking at the subtests for Photoscan, Carrizo's biggest hiccup was in the first stage which loads images into memory and attempts to identify and compare similar points in the photograph and its location relative to the object. This requires a significant amount of memory management, and Carrizo loses 17% in the first stage to Kaveri. Stage 4 gets a bump with Carrizo over Kaveri, enough to bring it back over Trinity and Richland, but overall the Carrizo part is still 6% down.

Cinebench R15

Cinebench is a benchmark based around Cinema 4D, and is fairly well known among enthusiasts for stressing the CPU for a provided workload. Results are given as a score, where higher is better.

Cinebench R15 - Single Threaded

Cinebench R15 - Multi-Threaded

Similar to Dolphin and 3DPMv2, Carrizo's microarchitectural changes help with CineBench 15 as well. Kaveri appearing on the same level as Trinity/Richland for single threaded is as expected, with the better multithreading handling coming through in the second test.

HandBrake v0.9.9: link

For HandBrake, we take two videos (a 2h20 640x266 DVD rip and a 10min double UHD 3840x4320 animation short) and convert them to x264 format in an MP4 container.  Results are given in terms of the frames per second processed, and HandBrake uses as many threads as possible.

HandBrake v0.9.9 LQ Film

HandBrake v0.9.9 2x4K

Typically we expect something like video conversion to be memory intensive, especially with codec decone/encode, however the base microarchitecture gives Carrizo an 8% lead and a 9% lead over Kaveri in our low quality and double UHD tests.

Hybrid x265

Hybrid is a new benchmark, where we take a 4K 1500 frame video and convert it into an x265 format without audio. Results are given in frames per second.

Hybrid x265, 4K Video

However, with x265. both Kaveri and Carrizo perform similarly.

Performance at 3 GHz: Real World Performance at 3 GHz: Linux
Comments Locked

131 Comments

View All Comments

  • aryonoco - Friday, July 15, 2016 - link

    Oh great, another AMD Bulldozer CPU, on 28nm, in 2016.

    I can't even begin to pretend to be enthusiastic about this.

    The only people who will buy this are internet cafe/game centres in developing countries; none of whom care about AT says. I wonder why Ian thought he should spend so much time thoroughly reviewing a part that not one of AT's readers will ever buy.
  • Cryio - Friday, July 15, 2016 - link

    To be fair this is an architecture from 2015 brought to desktops half-hearted and it wasn't released for any good reason IMO.

    Given the improvements Kaveri and Carizzo pose over the previous generations most of the time, if AMD would have released FX CPUs based on Stream Roller and Excavator, we would've got some interesting CPUs.
  • kondor999 - Friday, July 15, 2016 - link

    Oh look, another shitty, hot, antiquated AMD CPU that only (gosh, I can't actually think of anyone) would buy.

    I really wish they'd get their act together. Not necessarily because I'd buy their products, but just to force Intel into giving us more than a 3% (or so) IPC iterative improvement.
  • TheinsanegamerN - Friday, July 15, 2016 - link

    So, if you only want AMD to be competitive so that intel is more competitive, how do you expect them to do that when nobody buys their stuff? R and D needs money.
  • silverblue - Friday, July 15, 2016 - link

    Though temperature readings appear to be a bit hard to come by, the 845 appears to idle at 26 degrees and full load on AIDA64 at 40 degrees which puts it below anything else AMD has, making it comparable to the Pentium G3220 (though the former does have a better cooling solution).

    Source: http://www.eteknix.com/amd-athlon-x4-845-carrizo-p...

    It's up to you whether you want to believe that or not.
  • Meteor2 - Saturday, July 16, 2016 - link

    I don't think there's more than 3℅ IPC increase annually available. Apple, ARM, Qualcomm and Intel all seem to be converging at any given power.
  • zodiacfml - Friday, July 15, 2016 - link

    Waste of time. Yes, it competes with the Pentium, but the Intel chip has integrated graphics which could be useful whether the user has a graphics card or not. Paying more for less.

    Talking about Zen, it will just compete with Haswell generation chips. Intel knew this which is why their tick-tock strategy has slowed down.
  • cocochanel - Friday, July 15, 2016 - link

    Intel didn't slow down because of AMD. It's because of x86. It's harder and harder even for mighty Intel engineers to squeeze more performance from an antiquated ISA. The fact that Zen took so long to get here it's a clear indication of the same.
  • TheinsanegamerN - Monday, July 18, 2016 - link

    There is no clear indication

    AMD has far fewer resources then intel does, so it makes sense it would take them much longer to make a new CPU arch then intel.

    Intel isnt making any advancements because they have no competition. There may be more performance sitting in their arch they are not using, since there really is no reason to.
  • Lolimaster - Saturday, July 16, 2016 - link

    Carrizo totally gimped by l2 cache.

    The problem is that Bristol Ridge comes with the same pathetic 2MB for their 4 cores APU.

Log in

Don't have an account? Sign up now