AMD Carrizo Part 2: A Generational Deep Dive into the Athlon X4 845 at $70
by Ian Cutress on July 14, 2016 9:00 AM ESTOffice 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.
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.
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.
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.
However, with x265. both Kaveri and Carrizo perform similarly.
131 Comments
View All Comments
Meteor2 - Saturday, July 16, 2016 - link
Thanks. It's a little more complex than i3/5/7-nxxx, where n increments by one each generation...TheinsanegamerN - Monday, July 18, 2016 - link
cariizo cores use the excavator design. excavator is core design, carrizo is the product line.Mokona512 - Saturday, July 16, 2016 - link
Please do this test with the Phenom II series in order to understand the generational IPC changes, and also providing a better point of reference for the Zen CPUs. The Zen claims are based on IPC changes from a CPU series where there was a drop in IPC.Ian Cutress - Monday, July 18, 2016 - link
I recently redid a Phenom X6 in Bench, though that's absolute chip perf and not exactly what you're looking for, but it's there :)Elizabeth king - Sunday, July 17, 2016 - link
Love spell came out tremendously, I highly recommending robinson.buckler @ yahoo . com for whatever problems you are experiencing in your relationship. He is the real deal. his love spell is absolutely wonderful.lwatcdr - Sunday, July 17, 2016 - link
I find AMDs low cost offerings really interesting but this just doesn't work for me. The Carrizo on the desktop just seems too limiting. I wish that AMD would update the AM1 line. It is so inexpensive and can support a good number of PCIe lots. For things like a NAS, media pc, or even a Chrome box/low end pc they seem like a really good choice except that they have not been updated in years.silverblue - Monday, July 18, 2016 - link
Puma+ is kind of like what Atom did with their earlier Atoms, a more efficient version of Jaguar. I'm surprised that they didn't just lock the turbo and produce these in Jaguar's place, unless it's not cost-effective to do so.The cat cores are dead now, which is a shame as we never got to see how a dual channel memory interface would improve their performance.
Eris_Floralia - Monday, July 18, 2016 - link
Another great review. I've translated all articles about Bulldozer architecture into Chinese in order to let people know why it didn't success. I believe that an eight-core Steamroller or Excavator would be competitive, but that never comes out.TheinsanegamerN - Monday, July 18, 2016 - link
it wont be competitive, excavator is still far behind intel is performance and in TDP, and cant overclock at all. And it is still 28nm and cache limited.AMD really needs to kill the bulldozer line. It is AMD's netburst.
Eris_Floralia - Monday, July 18, 2016 - link
well, with some adjustments, steamroller can still reach high frequency. with additional L3 and larger L2, the problem with excavator may get solved. I mean that latest bulldozer architecture can do better than present piledriver, but the improvement obiviously doesnt worth a try.