This site may earn affiliate commissions from the links on this folio. Terms of use.

When AMD launched its Carrizo CPU refresh last year, information technology fabricated it clear that the chip would focus almost entirely on notebooks rather than desktops. The company just one depression-cost function for the desktop infinite — the Athlon X4 845. This flake doesn't utilise Carrizo's updated integrated GPU, just packs in four cores in ii CPU modules with a base clock at 3.5GHz and a 3.8GHz Turbo Style. Based on AMD's disclosures regarding Carrizo, the new chip should be faster and more efficient than the Kaveri cores information technology ostensibly replaced — but the truth turns out to be a chip more complicated.

Over at Anandtech, they've taken AMD'south latest core and matched it confronting previous parts based on Kaveri, Richland, and Trinity. The consequence is a 26-page magnum opus that compares the diverse chips in a huge range of scenarios and tests, from gaming to full general-purpose compute in Windows. Linux benchmarks are included, equally are results on a number of Intel products. At that place'southward a great bargain of information packed into the article and I highly recommend information technology.

The big-picture takeaway on Carrizo versus Kaveri at the CPU level is this: At that place'southward a solid grouping of tests where Carrizo shows existent efficiency improvements over Kaveri. The graph below compares each APU to the previous generation and gives the improvement in pct terms. A negative percentage means that the APU in question is slower than its predecessor, a positive percentage ways the newer chip is faster.

This is only ane of the overall graphs in the review. Anandtech's entire not-gaming Windows test suite measured the Athlon X4 845 equally being 7.3% faster than Kaveri when all of the APUs were locked to 3GHz and tested at that clock speed. That'due south non bad for a generational improvement, particularly considering that the X4 845 is a 65W office compared with a 95W X4 860K.

Situational boosts

Unfortunately, Carrizo is dogged by 2 issues. Kickoff, its gains are situational. In some workloads, Carrizo is as much every bit 32% faster than Kaveri. In others, it'southward eight-12% slower. Gaming takes a particular hit — Kaveri is approximately half dozen% faster than Carrizo when gaming, well-nigh beyond the board.

2nd, AMD was forced to pull clock speeds downwards when it shifted to Carrizo, merely every bit it was when Kaveri debuted. The 65W Carrizo tops out at iii.8GHz with a 3.5GHz base, while the X4 860K is a 3.7GHz / 4GHz CPU. Anandtech reports that the overclocking headroom with their item sample is pocket-size, at roughly 10%. Users would need to push the chip'due south clock at to the lowest degree that high to count on matching Kaveri'southward operation in the worst-case scenarios, though an OC'd X4 845 could too be substantially faster than the X4 860K.

Workloads that fit comfortably within Carrizo's larger L1 cache (128K L1-D, compared to 64K for Carrizo) or do good from its increased enshroud associativity (viii-way, upwards from iv-mode) show the largest improvements. Other tests show Kaveri winning past its newer cousin, presumably thanks to a combination of college clocks and a larger L2 cache. This core was originally designed for laptops and information technology shows — the smaller L2 cache and eight lanes of PCIe 3.0 may have been smart tradeoffs in the 15-25W space, but this is a 15W chip competing against desktop processors. Just pushing the TDP up to 65W doesn't mean that Carrizo was actually designed to compete in these power envelopes (as we discussed concluding year, Carrizo is really optimized to outstrip Kaveri at lower power envelopes, but may not compete well against it in the 65W+ infinite).

Those of you lot who have followed AMD'due south designs over the past few years are probable aware that nosotros saw a very like blueprint when Kaveri launched. Back in 2014, Kaveri proved information technology was an extremely stiff replacement for Richland at the 45W TDP envelope simply less persuasive at the 65W and 95W targets. Fries clocked above their sweet spot tend to require more voltage, which in plow generates more heat, which then requires more voltage… you get the picture.

On a more positive note, the competitive price ($lxx) and its quad-threaded blueprint makes the X4 845 a very potent competitor against some of Intel'southward dual-core CPUs similar the 20th Anniversary Pentium it replaced last year. In the near 4 years since AMD'southward offset Piledriver-based APU launched, the company has managed to better IPC (instructions per clock cycle, a mensurate of efficiency) past between 10% and twenty% while simultaneously reducing power consumption. That'southward a significant accomplishment, peculiarly for a visitor every bit cash-strapped as AMD, but it'due south going to take Zen to really motility the bar on the visitor's overall performance-per-watt story.

BristolRidge-Features

When AMD appear Bristol Ridge earlier this twelvemonth, nosotros idea the chips and chipsets would be debuting already — only Computex has come and gone with no sign of the refresh. If Bristol Ridge doesn't debut soon, information technology's possible that AMD will hold the wheel for a CES announcement, presumably aslope Zen. Sunnyvale continues to insist that its adjacent-generation CPU volition sample late in Q4 for a Q1 2017 launch, but Zen's get-go iteration is CPU-but. AMD volition still need an APU to pair it with, which means 28nm Bristol Ridge APUs based on Carrizo will share space with 14nm Zen cores based on AMD's new architecture. AMD hasn't said when it'll push Zen into APUs, merely it'south safe to assume the visitor will make that transition as speedily as it tin can. Even if Zen-based APUs stick with DDR4 as their memory standard, the additional CPU operation and superior 14nm process make information technology a much more attractive role — assuming information technology hits its power and operation targets.

Bristol Ridge is unlikely to shake upwards the overall roadmap or Carrizo's performance very much. While the chip will have some improvements and tweaks (and should support dual-aqueduct retentiveness in laptops) the typical gain for this kind of refresh is in the 3-5% range. Another 5% sure wouldn't injure the core's CPU performance, but Zen's 40% is what people are going to be watching for. If you're looking to build an entry-level AMD gaming rig, Kaveri is probably the better choice. If, on the other hand, you lot want a general-purpose organisation, Carrizo and the X4 845 may be the better core.

At present read: How L1 and L2 CPU caches work, and why they're an essential function of modern chips