AMD hosted its financial & analyst day today, revealing information on Vega, Threadripper, notebook deployments of its CPUs & GPUs, and data center products. Some timelines were loosely laid-out with initial benchmark previews, provided an outline for what to expect from AMD in the remainder of 2017.
Most of our time today will be spent detailing Vega, as it’s been the topic of most interest lately, with some preliminary information on the CPU products.
AMD today made available a power plan update which should change how the Balanced plan impacts Ryzen performance.
Problems with Windows preset power modes have been one of the biggest annoyances with Ryzen, and AMD has officially recommended the High Performance preset in the past in order to avoid subpar performance in benchmarks. This wasn’t a big deal from a testing point of view since High Performance mode effectively avoids all of these issues, but for everyday use, it was: High Performance mode doesn’t allow CPU frequency to drop when idle, and the additional power consumption can really hurt the long-term value of the system (it’s also just wasteful). Balanced mode does drop frequency, but it’s also been overly aggressive with core parking on Ryzen chips specifically, making it sub-optimal for use. We discussed what this looks like from a user’s point of view in our “Just Research” article, where frequency plots offer visualization for the impact of Performance vs. Balanced mode. The same article contains some FPS benchmarks between the two power modes.
AMD has made two major changes in this update. Quoting their statement:
Maintain residency in CPU p0 or p1 to give Zen full control over clocks and volts.
Disable core parking.
They specifically noted that Intel also fully disables core parking in the Balanced power plan. Our tests have always used High Performance mode for Ryzen platforms (except power tests), and our results will not be affected by this update.
This first revisit to Ryzen’s performance comes earlier than most, given the tempestuous environment surrounding AMD’s latest uarch. In the past weeks, we’ve seen claims that Windows updates promise a significant boon to Ryzen performance, as has also been said of memory overclocking, and we were previously instructed that EFI updates alone should bolster performance. Perhaps not unrelated, game updates to major titles could have potentially impacted performance, amounting to a significant number of variables for a revisit.
Today’s content piece aims to isolate each of these items as much as reasonable – not all can be isolated, like game updates – to better determine the performance impact from the individual changes and updates. We’ll then progress cumulatively through charts as updates are applied. Our final set of charts will contain Windows version bxxx.970, version 1002 EFI on the CH6, and memory overclocking efforts.
The AMD Ryzen 5 series is set to continue AMD’s launch of its new Zen architecture, debuting earlier this month with the Ryzen 7 (R7) CPUs. Thus far, we’ve seen the release of the 1800X flagship, 1700X, and 1700 CPUs (the last of which being our option of choice). AMD’s subsequent launches will be focused on the R5 line, announced today, and a later-specified R3 line. We’re looking at a retail release date of April 11 for the R5 1400, R5 1500X, R5 1600, and R5 1600X CPUs; the R3 CPUs, meanwhile, are expected for availability in 2H17.
AMD hasn’t fully revealed all the technical details of these SKUs at this time. We know enough of the basics, but will have to wait for more information on how the CCXs are configured in 6C/12T scenarios.
Here’s a listing of prices, to get started:
AMD yesterday released a community update with interesting assertions regarding thread scheduling, temperature reporting, Windows power plan issues, and SMT challenges.
According to AMD’s Robert Hallock, the company has found no indication that Windows 10 thread scheduling is operating improperly for Zen. This should be the final word in any argument that Microsoft thread scheduling issues are sabotaging Ryzen: they aren’t, as stated by AMD below:
“We have investigated reports alleging incorrect thread scheduling on the AMD Ryzen processor. Based on our findings, AMD believes that the Windows 10 thread scheduler is operating properly for ‘Zen,’ and we do not presently believe there is an issue with the scheduler adversely utilizing the logical and physical configurations of the architecture.
“As an extension of this investigation, we have also reviewed topology logs generated by the Sysinternals Coreinfo utility. We have determined that an outdated version of the application was responsible for originating the incorrect topology data that has been widely reported in the media. Coreinfo v3.31 (or later) will produce the correct results.”
When we first began benchmarking Ryzen CPUs, we already had a suspicion that disabling simultaneous multithreading might give a performance boost in games, mirroring the effects of disabling hyperthreading (Intel’s specific twist on SMT) when it was first introduced. Although hyperthreading now has a generally positive effect in our benchmarks, there was a time when it wasn’t accounted for by developers—presumably partly related to what’s happening to AMD now.
In fact, turning SMT off offered relatively minor gaming performance increases outside of Total War: Warhammer—but any increase at all is notable when turning off a feature that’s designed to be positive. Throughout our testing, the most dramatic change in results we saw were from overclocking, specifically on the low stock frequency R7 1700 ($330). This led many readers to ask questions like “why didn’t you test with an overclock and SMT disabled at the same time?” and “how much is Intel paying you?” to which the answers are “time constraints” and “not enough, apparently,” since we’ve now performed those tests. Testing a CPU takes a lot of time. Now, with more time to work on Ryzen, we’ve finally begun revisiting some EFI, clock behavior, and SMT tests.
As with any new technology, the early days of Ryzen have been filled with a number of quirks as manufacturers and developers scramble to support AMD’s new architecture.
For optimal performance, AMD has asked reviewers to update to the latest BIOS version and to set Windows to “high performance” mode, which raises the minimum processor state to its base frequency (normally, the CPU would downclock when idle). These are both reasonable allowances to make for new hardware, although high-performance mode should only be a temporary fix. More on that later, though we’ve already explained it in the R7 1700 review.
This is quick-and-dirty testing. This is the kind of information we normally keep internal for research as we build a test platform, as it's never polished enough to publish and primarily informs our reviewing efforts. Given the young age of Ryzen, we're publishing our findings just to add data to a growing pool. More data points should hopefully assist other reviewers and manufacturers in researching performance “anomalies” or differences.
The below is comprised of early numbers we ran on performance vs. balanced mode, Gigabyte BIOS revisions, ASUS' board, and clock behavior when under various boost states. Methodology won't be discussed here, as it's really not any different from our 1700 and 1800X review, other than toggling of the various A/B test states defined in headers below.
AMD’s R7 1700 CPU ($330) immediately positions itself in a more advantaged segment than its $500 1800X companion, which proved poor value for pure gaming machines in our tests. Of course, as we said previously (page 5, 8), the 1800X makes more sense for our tested production tasks than the $1000 6900K when considering price:performance. For gaming, both are poor choices; the 1800X performs on par with i5 CPUs in game benchmarks, and the 6900K is $1000. It’s about value, not raw performance: Multiplicative increments in price to achieve performance equivalence (gaming) to cheaper chips is not good value. Before venturing into the 1440p/4K argument, we’d encourage you to read this review. The R7 1700 – by nature of that very argument, but also by nature of a trivial overclock – effectively invalidates the 1800X for gaming machines, finally granting AMD its champion for Ryzen.
We are also restricting this review to one page, as a significant portion of readers had unfortunately skipped straight to the gaming results page without context. It’s not as good for formatting or page load times, but it’ll hopefully ensure the other content is at least scrolled past, even if still ignored altogether.
Enough of that.
In this AMD R7 1700 review, we look at the price-to-performance of AMD’s new $330 CPU, which was explicitly marketed as an i7-7700K counter in price/performance when presented at AMD’s tech day. We’re benchmarking the R7 1700 in our usual suite of gaming, synthetic, and render tasks, quickly validating average auto voltages and temperatures along the way. Overclocks and SMT toggling further complicate testing, but provide a look at how the R7 1700 is capable of eliminating the gap between AMD’s own flagship and its more affordable SKU.
We already explained this amply in our AMD Ryzen R7 1800X review, primarily on pages 2 and 3 (but also throughout the article), but it's worth highlighting in video form for folks who prefer not to read articles. It's unfortunate that the test methodology and logistical pages were largely overlooked in the review -- most folks just jumped straight to the conclusion or gaming results, sadly -- so we are highlighting again, in video format, some of the things discussed on those pages.
As stated several times in this new video, we strongly encourage checking out the article. We are delaying our R7 1700 review by a day because of the addition of this video to our release schedule. There's not much more to say here, so we'll just embed that below:
Intel has enjoyed relatively unchallenged occupancy of the enthusiast CPU market for several years now. If you mark the FX-8350 as the last major play prior to subsequent refreshes (like the FX-8370), that marks the last major AMD CPU launch as 2012. Of course, later launches in the FX-9000 series and FX-8000 series updates have been made, but there has not been an architectural push since the Bulldozer/Piledriver/Steamroller series.
AMD Ryzen, then, has understandably generated an impregnable wall of excitement from the enthusiast community. This is AMD’s chance to recover a market it once dominated, back in the Athlon x64 days, and reestablish itself in a position that minimally targets parity in price to performance. That’s all AMD needs: Parity. Or close to it, anyway, while maintaining comparable pricing to Intel. With Intel’s stranglehold lasting as long as it has, builders are ready to support an alternative in the market. It’s nice to claim “best” on some charts, like AMD has done with Cinebench, but AMD doesn’t have to win: they have to tie. The momentum to shift is there.
Even RTG competitor nVidia will benefit from this upgrade cycle. That’s not something you hear a lot – nVidia wanting AMD to do well with a launch – but here, it makes sense. A dump of new systems into the ecosystem means everyone experiences revenue growth. People need to buy new GPUs, new cases, new coolers, and new RAM to accompany any moves to Ryzen. Misalignment of Vega and Ryzen make sense in the sense of not smothering one announcement with the other, but does mean that AMD is now rapidly moving toward Vega’s launch. Those R7 CPUs don’t necessarily fit best with an RX 480; it’s a fine card, just not something you stick with a $400-$500 CPU. Two major launches in short order, then, one of which potentially drives system refreshes.
AMD must feel the weight borne by Atlas at this moment.
In this ~11,000 word review of AMD’s Ryzen R7 1800X, we’ll look at FPS benchmarking, Premiere & Blender workloads, thermals and voltage, and logistical challenges. (Update: 1700 review here).