Ryzen Release Thread

[SteelCrysis]

https://www.extremetech.com/computing/26...used-to-be

When Ryzen 7 launched, AMD promised that the performance we saw on launch day would continue to improve over time thanks to future UEFI updates and app optimizations. Companies often make these claims, but the long-term improvement rate historically isn’t very good. It’s not surprising that AMD had ground to make up back in 2017 — after six years in the metaphorical wilderness, and with its desktop share in full retreat, plenty of vendors hadn’t had to think about AMD performance optimizations for years. Even vendors that did invest in Bulldozer or Piledriver-specific performance improvements would’ve found those gains ill-suited to Ryzen’s architecture. But at the same time, there was no initial way to know what kind of improvements to expect over Ryzen’s lifetime.

These performance gains won’t rewrite the book on the Ryzen 7 1800X, but they’re proof that AMD was telling the truth when it told users to expect better performance figures over time. While the gains are application-specific, the Ryzen 7 1800X is faster today than it was in 2017.

[SteelCrysis]

https://www.gamersnexus.net/hwreviews/32...der/page-3

We’re getting to a point where Ryzen’s generational improvements, from a pure FPS and performance perspective, are going to look an awful lot like Intel’s. If you bought Ryzen 1, there’s really no reason to replace it with Ryzen 2. AMD’s improvements are primarily in unseen places to the end user, like the reduced minimum voltage at a given frequency, which we previously highlighted here. These are less flashy than gaining, say, 20% in framerate, so will undoubtedly be largely overlooked by the general userbase. Such improvements are critical, and we think AMD has done well to reduce voltage at a given frequency over first generation.

As for the rest, it’s rather lateral from Ryzen 1. If you already own a 1600X, don’t buy a 2600X to replace it. If you don’t own a current CPU, strongly consider a 2600X. The 2600 and 2600X get all the same accolades as the 1600 and 1600X did. They’re the same price, they perform a bit better, they reduce given-frequency voltage requirements, and they fit in the same boards. Where we’d previously recommend the R5 1600, we now recommend the R5 2600. It’s as simple as that.

The R5 2600(X) is an all-arounder CPU, and if it weren’t for Intel’s 6-core i5s, it’d be the only option. The 6-core i5s still have some viability, but AMD’s R5 line remains our go-to at the price point, in general, with Intel remaining the go-to at the $330 for gaming, or AMD at $330 for production/streaming-type applications (although the 8700K does well to compete there – better than the 8600K does to compete with the R5s).

[SteelCrysis]

Lower power consumption Ryzen APUs announced: https://www.techpowerup.com/243581/amd-o...cient-apus

[SteelCrysis]

Several first-gen Ryzen CPUs reach end-of-life: https://www.techpowerup.com/243596/six-f...-reach-eol

[SteelCrysis]

https://www.gamersnexus.net/guides/3290-...ency-curve

As stated in our review, the Ryzen 2 processors primarily shine in lower voltage requirements at a given frequency, which is doubly illustrated here, against the 1700. Both CPUs do have somewhat non-linear, pseudo-exponential curves as frequency increases. This also happens with Intel, mind you, but that’s well-documented. The Ryzen CPUs are still relatively new. Without BCLK and other tuning, we hit a wall at 4.2GHz and can’t push voltage high enough to stabilize >4.2GHz (multiplier only). We could with exotic cooling, probably, but temperature begins to dictate boosting functionality -- just like modern GPUs.

The more interesting point is that our 2700X can hold its base frequency of 3.7GHz on all cores at 1.0v, while our 1700 needs at least 1.1625V for the same clock speed.

[SteelCrysis]

Lower-end Ryzen 2 models on the way: https://www.techpowerup.com/243757/amd-l...ryzen-skus

[SteelCrysis]

Zen 2 will be coming next year: https://www.techpowerup.com/243818/amd-t...019-launch

[SteelCrysis]

https://www.techpowerup.com/reviews/AMD/...00/20.html

One area where the Ryzen 7 2700 managed to surprise us is energy efficiency. Its multi-threaded power-draw is over 50 W lower than that of the 2700X (141 W vs 199 W), while offering not that much lower performance. The underlying reason is that 2700X boosts very high, into a region where the processor is faster, but not operating so efficiently anymore. Ryzen 7 2700 on other other hand runs at lower clock and lower voltage in this scenario, resulting in higher efficiency. Our new energy efficiency testing, which doesn't just measure power, but also takes into account how quickly tasks complete due to higher performance, shows the amazing lead of the processor. This makes the Ryzen 7 2700 the most energy-efficient processor we ever tested.

[SteelCrysis]

https://www.techpowerup.com/243877/bluec...s-for-2018

The information gleaned is a confirmation, of sorts, of AMD's planned launch of their Z490 platform in June; the B450 chipset coming a little bit later, in July (an expected product, in every sense); and AMD's second-gen Threadripper, a known-quantity already, which should accompany a X399 platform refresh.

[SteelCrysis]

http://www.legitreviews.com/ddr4-memory-...orm_205154

Our quick look at DDR4 memory scaling on the AMD Ryzen 7 2700X 8-core, 16-thread processor on the AMD X470 platform with a G.Skill Sniper X DDR4-3400 memory kit was pretty fun. We were able to push this 3400MHz kit up to 3733MHz with the stock timings of CL16-16-16-36 by just bumping up the DRAM voltage. We were never able to get this high on the Ryzen 1000 series on the AMD X370 chipset, so the Ryzen 2000 series is allowing AMD to get higher DDR4 clock speeds than before. For a hot second it looked like we could get 3800MHz up and running, but we had to throw in the towel for the time being.

[SteelCrysis]

B450 chipset detailed: https://www.techpowerup.com/244415/amd-b...t-detailed

[SteelCrysis]

AMD has canceled the Z490 chipset because they had trouble finding the PLX chips that they needed:

[SteelCrysis]

More Ryzen models: https://www.techpowerup.com/245235/amd-r...-geekbench

[SteelCrysis]

https://www.techpowerup.com/245676/first...0x-surface

The overall value proposition of this 2300X CPU is somewhat marred, though, by the existence of the 2200G APU - a quad-core solution as well, which also packs in integrated Vega 8 graphics - for $99. And it's unlikely the Ryzen 3 2300X will be priced much lower than the $125 AMD was asking for its 1300X processor, which already looked bad compared to the 2200G. It seems AMD has a missed opportunity in its hands to further differentiate these two product lines by enabling SMT on one of them. If anything, I'd suggest doing so on the APU side of the equation - thus strengthening AMD's performance compared to Intel's i3 CPUs, which all pack integrated graphics, but lack any sort of SMT support.

[SteelCrysis]

https://www.techpowerup.com/245735/techp...ct-is-real

Despite being newer, fewer respondents use 6th generation "Skylake" and 7th generation "Kaby Lake" processors than older generations, because those on something like 4th generation "Haswell" or even "Ivy Bridge," don't see the value in upgrading. But then something changed in 2017 - AMD became competitive again, and forced an increase in CPU core counts across the segment. AMD's Ryzen processor family, including both its 1st and 2nd generations, are better received in the market than Intel's competing 8th generation "Coffee Lake" and 7th generation "Kaby Lake." The data stands to validate the "Ryzen effect," the idea that the introduction of Ryzen disrupted Intel's near-monopoly, increased core-counts, and brought innovation back to the segment.

More of our readers use AMD Ryzen processors than Intel Core "Coffee Lake" and "Kaby Lake." So in the period following Intel's launch of 7th generation "Kaby Lake" (slightly before the launch of Ryzen), more AMD processors were installed among our readers. This of course doesn't mean that there are more AMD users, since we're not counting pre-Ryzen Intel generations such as "Skylake" and "Haswell." This seems to suggest that the "Ryzen effect" is not a myth.

[SteelCrysis]

https://www.extremetech.com/computing/27...en-3-2300x

The positioning of AMD’s Raven Ridge APUs makes the value proposition of the Ryzen 5 2500X and Ryzen 3 2300X rather dubious. There are, of course, those who see no value in an integrated GPU because they never intend to use one. But even these buyers are practically acquainted with buying one anyway, since every mainstream Intel desktop CPU ships with an integrated GPU. The only thing you give up by opting for the 2400G as opposed to the 2500X is 4MB of L3 and 100MHz of maximum clock speed — about 2.5 percent of clock, in other words. That L3 cache could be good for a couple more percent, but that’s about it in the tests we’ve seen. Obviously, there can always be specific workloads where the larger L3 is useful, but AMD’s Ryzen CPUs haven’t shown themselves to be particularly L3-limited in desktop benchmarks.

The 2300X has an even tougher position to carve out because the Ryzen 3 2200G is a tough chip to beat at its $100 price point. Budget builders typically fight for every dollar and the combination of a capable 512-core GPU and AMD’s much-improved quad-core performance relative to Carrizo means AMD has a lock on budget gaming with this APU. But that doesn’t just apply to the Core i3-8100, Intel’s competition — it also applies to other AMD solutions. And while the Ryzen 3 2300X does offer an eight percent boost clock improvement, I’d honestly rather have the APU’s onboard GPU at the $100 price point than a touch more clock.

The most practical explanation, of course, is that AMD is simply engaging in a bit of die recovery and selling hardware it otherwise would’ve had to toss in the bin. But assuming this rumor is accurate, nobody who bought into Ryzen’s lower-end chips last year need feel like they’ve been shafted. AMD isn’t making any major product changes this cycle.

[SteelCrysis]

https://www.neowin.net/news/low-power-am...-watts-tdp

It's worth noting that this is not the first time that the low-power processor surfaced online as a recent listing on the website of motherboard manufacturer ASRock also revealed the Ryzen 7 2700E along with another energy-efficient CPU, the Ryzen 5 2600E. The listing is for the AB350M Pro4 model, which includes support for both CPUs.

The ASRock listing indicates that the Ryzen 7 2700E will have a base clock of 2.8 GHz and a 4MB cache, which coincides with the 3DMark data result. Meanwhile, the Ryzen 5 2600E seems to have a higher clock speed of 3.1GHz based on the ASRock listing.

While AMD has yet to officially confirm the SKU for the Ryzen 7 2700E, the processor will likely be integrated into small form factor PCs and devices, Guru3D noted.

[SteelCrysis]

ASRock adds the Ryzen 7 2700E and Ryzen 5 2600E to its supported CPUs list: https://www.tomshardware.com/news/amd-ry...37427.html

[SteelCrysis]

MSI states that its B450 motherboards will support CPUs with more than 8 cores: https://www.techpowerup.com/246100/msi-d...ore-counts

[SteelCrysis]

Ryzen 5 2500X on the way: https://www.tomshardware.com/news/acer-n...37504.html

[SteelCrysis]

https://www.tomshardware.com/news/amd-ea...37513.html

AMD CEO Lisa Su stated at Computex 2018 that the company had sold a total of five million Ryzen processors since they were released, which is a seemingly minor amount compared to the ~150 million PCs sold per year. Nevertheless, AMD's income from Ryzen processors jumped 64% year-over-year, which isn't surprising given that Ryzen processors currently hold four of the top five spots on Amazon's CPU best sellers list. Ryzen processor sales only increased slightly over last quarters' 60% of AMD's CPU sales, but AMD forecasts that number to grow as more customers upgrade from older processors over the coming months.

AMD lumps its sales of CPUs and GPUs together into its Computing and Graphics segment, which jumped a whopping 64% year over year but reflected a loss of 3% compared to the prior quarter. Su attributed the lower revenue to the reduced average selling price of its flagship Ryzen 2000-series desktop CPUs, which debuted at much lower prices than the first-gen Ryzen models, and increased sales of less-expensive Raven Ridge APUs. That has contributed to a lower average selling price (ASP), which impacts revenue. Threadripper 2 will arrive to the market in the coming months, and the higher-priced models should help improve the company's declining ASPs.

https://www.techpowerup.com/246282/no-16...aunch-2019

AMD in its Q2-2018 investors conference call dropped more hints at when it plans to launch its 3rd generation Ryzen processors, based on its "Zen2" architecture. CEO Lisa Su stated in the Q&A session that rollout of 7 nm Ryzen processors will only follow that of 7 nm EPYC (unlike 1st generation Ryzen preceding 1st generation EPYC). What this effectively means is that the fabled 16-core die with 8 cores per CCX won't make it to the desktop platform any time soon (at least not in the next three quarters, certainly not within 2018).

[SteelCrysis]

Ryzen DRAM Calculator released: https://www.techpowerup.com/246355/ryzen...s-released

[SteelCrysis]

Lenovo reveals more info on Ryzen 5 2500X and Ryzen 3 2300X: https://www.tomshardware.com/news/lenovo...37557.html

[SteelCrysis]

Ryzen 5 2500X and Ryzen 3 2300X are officially announced, appear to be exclusive to OEMs: https://techreport.com/news/34080/amd-de...ebuilt-pcs

[SteelCrysis]

https://www.techpowerup.com/247689/amd-r...e-i9-9900k

The Forbes article that cites the Elchapuzas Informatico, however, is skeptical that AMD could make such a short-sighted product investment. It believes that development of a 10-core die on existing "Zen+" architecture could warrant a massive redesign of the CCX (Zen Compute Complex), and AMD would only get an opportunity to do so when working on "Zen 2," which AMD still expects to debut by late-2018 on its EPYC product line. We, however, don't discount the possibility of a 10-core "Zen+" silicon just yet. GlobalFoundries, AMD's principal foundry partner for CPUs, has given up on 7 nm, making the company fall back to TSMC to meet its 7 nm roadmap commitments. TSMC already has a long list of clientele for 7 nm, including high-volume contracts from Apple, Qualcomm, and NVIDIA. This could force AMD to bolster its existing lineup as a contingency for delays in 7 nm volume production.

[SteelCrysis]

https://www.techpowerup.com/247942/amd-m...-digitimes

Intel's woes with the 14 nm process node has been reported before, let alone the troubled transition to the 10 nm process, which further helps AMD's case. The report further claims that board partners including ASUS, MSI, and Gigabyte "have ramped up production and shipments of devices fitted with AMD processors, driving up the chipmaker's share of the desktop processor market to over 20% in the third quarter. The company is very likely to see the figure further rebound to the level of 30% again." This, in addition to the recently announced AMD EPYC implementation contracts with Cisco and HPE, can make that 30% number more plausible as well. It remains to be seen what the following months hold for Intel and AMD in the desktop processor market, but there is no denying that things are competitive in multiple segments at the same time, and that is always a good thing for the consumer.

[SteelCrysis]

https://www.gamersnexus.net/news-pc/3369...-30-launch

While this problem seems to be mostly limited to China, it’s noteworthy that the demand for AMD’s entry level boards are so strong. ASUS, Gigabyte, and MSI have all been affected with tight supply of A320, B450, and to a lesser extent, A68 series of motherboards. The increase in demand for entry level boards seems to stem from bundling deals from the above mentioned vendors.

To keep prices stable and mitigate shortages, affected vendors have canceled current bundling promotions, and ASUS has plans for production expansion meet help meet demand.

[SteelCrysis]

https://www.tomshardware.com/news/amd-in...37864.html

We won't have desktop market share numbers for Q3 for another month, but we spoke with Mercury Research's Dean McCarron about AMD's chances of attaining 30 percent of the market by the end of Q4. McCarron cautioned that he doesn't forecast future market share, but he did provide some interesting perspective:

"[...] for a 30 percent AMD desktop share to happen in Q4, AMD's volumes would need to either be 250 percent higher in the quarter than they were last year, or Intel's shipments would need to decline 65 percent, or some combination of the two. Since 2000, the largest on-year decline Intel has experienced for any quarter was -24 percent, and the largest on-year increase AMD has experienced was +57 percent.

If somehow moves at these historic extremes simultaneously happened in Q4, AMD's share would fall in the low 20 percent range, well short of 30 percent. Should a 30 percent AMD share materialize next quarter, it would be an unprecedented statistical outlier within the history of the processor market and carry with it far-reaching consequences."

McCarron's statements are telling. In order to reach just 20 percent market share, either AMD would have to more than double its year-on-year volume, or Intel shipments would have to decline 65 percent in a very short amount of time.

But the market is fluid. According to Intel, its shortage of 14nm processors comes courtesy of record demand, but the general consensus is that the 10nm delay has wreaked havoc on Intel's production scheduling. In either case, Intel is selling every piece of silicon it spins in its fabs. That makes it unlikely that the company would lose such a large amount of share in a short amount of time, even after accounting for the first growth we've seen in the desktop PC market since 2011.

It's natural to expect AMD to pick up some sales from Intel's production missteps, though, particularly as the companies face off during the yearly holiday battle. However, a report from industry soothsayer IDC last week says that AMD hasn't materially gained from Intel's production challenges, at least not yet:
...
OEMs can source high-end Core models, but we've tracked a steady increase in pricing over the last several weeks in the retail market and scarce availability of Intel's premium K-series models, particularly in foreign markets. That makes it likely that AMD will make at least some progress against Intel in the DIY market during the 14nm shortage.

It's natural to expect AMD's growth rate to accelerate if it can roll out 7nm CPUs for the desktop before Intel can release 10nm products. AMD has also brought a new lineup of low-cost Athlon processors with integrated graphics to market, which is a key to securing more OEM orders in the high-volume budget PC market.

For now, industry watchers haven't seen a massive uptick in AMD's market share in relation to Intel's shortages, and we've yet to see AMD's pricing increase, but that could change quickly as we move into the holiday season. AMD had a bang-up holiday season last year, and we expect it will have similar, if not better, success this year as Intel suffers from high pricing and scarce availability. We'll have the numbers for AMD's Q3 market share in a month, but we'll have to wait until the beginning of next year to see how the holiday season pans out.

[SteelCrysis]

https://www.extremetech.com/computing/27...00k-launch

The sharp-eyed folks at Tech Radar caught a 10 percent cut to the 2700X that went live today, taking the chip from $329 to $294. The move puts AMD at a sharp price advantage over Intel, whose Core i9-9900K will debut at $488. The 9900K is widely expected to outperform the 2700X overall, thanks to a combination of higher single-threaded clocks and superior single-threaded efficiency, but the significant price gap will work to AMD’s favor. Over the past 18 months we’ve seen the performance crown switch back and forth between the two companies. The Ryzen 7 1800X took the crown from the 7700K, the Core i7-8700K reclaimed it on the basis of stellar six-core scaling and high single-threaded clocks, and the Ryzen 7 2700X took it back again with improved single-thread perf and a clock speed boost of its own. Now, with the Core i9-9900K, Intel may indeed reclaim the overall performance crown. But at the same time, AMD isn’t going to let that pass without a fight.

[SteelCrysis]

https://www.techpowerup.com/248642/amd-z...over-zen-1

According to the Italian tech publication, we could expect Zen 2 IPC gains of 13 percent over Zen+, which in turn posted 2-5% IPC gains over the original Zen. Bits n Chips notes that these IPC gains were tested in scientific tasks, and not in gaming. There is no gaming performance data at the moment. AMD is expected to debut Zen 2 with its 2nd generation EPYC enterprise processors by the end of the year, built on the 7 nm silicon fabrication process. This roughly 16 percent IPC gain versus the original Zen, coupled with higher clocks, and possibly more cores, could complete the value proposition of 2nd gen EPYC. Zen 2-based client-segment products can be expected only in 2019.

[SteelCrysis]

https://www.tomshardware.com/news/amd-ne...38029.html
I think they meant to say "tick equivalent".

AMD is already sampling its 7nm Rome processors, which mark the debut of the Zen 2 microarchitecture, to customers. The firm also has its Zen 3 processors under development. This third-gen microarchitecture will debut on the 7nm+ process, with the "+" indicating this will be a second generation of the 7nm node (a "tock" equivalent).
...
AMD also announced that it is already working on the Zen 4 microarchitecture, but didn't reveal significant details.

[SteelCrysis]

https://www.techpowerup.com/249450/amd-z...r-than-zen

According to Expreview, AMD conducted DKERN + RSA test for integer and floating point units, to arrive at a performance index of 4.53, compared to 3.5 of first-generation Zen, which is a 29.4 percent IPC uplift (loosely interchangeable with single-core performance). "Zen 2" goes a step beyond "Zen+," with its designers turning their attention to critical components that contribute significantly toward IPC - the core's front-end, and the number-crunching machinery, FPU. The front-end of "Zen" and "Zen+" cores are believed to be refinements of previous-generation architectures such as "Excavator." Zen 2 gets a brand-new front-end that's better optimized to distribute and collect workloads between the various on-die components of the core. The number-crunching machinery gets bolstered by 256-bit FPUs, and generally wider execution pipelines and windows. These come together yielding the IPC uplift. "Zen 2" will get its first commercial outing with AMD's 2nd generation EPYC "Rome" 64-core enterprise processors.

[SteelCrysis]

https://www.techpowerup.com/249876/amd-r...-the-issue

To me this explanation sounds like bs.

OEMs don't buy customized APU chips from AMD, they all use the same physical chip, with the same capabilities. All the "driver tailoring" usually is just a bunch of logos and adding or removing features, which quite often is actually harming the user experience. While of course other components in the laptop might differ (networking, storage, audio), and the connected displays might run various refresh rates and resolutions, it's not like such differences have any significant effect on traditional desktop PCs. Imagine having to wait for your monitor vendor to approve and release a graphics driver update.

This somehow reminds me of the Android ecosystem, where phone makers were responsible for validating and releasing updates to the Android OS. Of course they already had your money, so why would they invest time and resources into improving something that yields no return and can possibly lead to support calls for issues with the upgrade (they'll happily sell their new phone model though). Just like AMD is trying now, Google has then started forcing OEMs to increase the update frequency, which never really worked out. An alternative approach is what NVIDIA does. Besides the vendor-supplied drivers, they offer a generic notebook driver on their website, that is updated with every new driver release and that you are free to use, and that as far as I know, works with nearly no issues.

Many users had success using the "force install" option in Windows Device Manager, and report that they're actually having fewer issues with that approach than when using the official driver. I think we can all agree that business users and casuals don't need a lot of driver updates, but the tech enthusiasts are a significant driver of AMD's business and should be kept happy (and they'll beta test the drivers, too, for free). Enthusiasts will tell their relatives and friends (who might not even know of AMD as a tech brand), what products to buy or to avoid, which is very important for a company like AMD that wants to establish a foothold in the highly competitive laptop market.

[SteelCrysis]

https://www.techpowerup.com/250183/amd-3...putex-2019

In a development that could explain why Intel is frantically stitching together 10 cores with the "Comet Lake" silicon, a slide leaked from a private event hosted by motherboard major GIGABYTE reveals that AMD's third generation Ryzen desktop platform could launch as early as Computex 2019 (June). The platform will include AMD's first client-segment processor based on its "Zen 2" microarchitecture, codenamed "Matisse," and its companion chipset, the AMD X570.

3rd generation Ryzen with X570 is expected to be the world's first mainstream desktop platform to feature PCI-Express gen 4.0. AMD could maintain the processor's backwards compatibility with older 300-series and 400-series chipset motherboards by shaping its PCI-Express implementation to use external re-drivers based on the motherboard, according to Taiwan-based industry observer Lars Nilsson. This could make 500-series motherboards slightly pricier than current AM4 motherboards. Backwards compatibility could mean unless you really need PCIe gen 4.0, you should be able to save by opting for older motherboards.

[SteelCrysis]

https://www.techpowerup.com/250412/amd-h...en-5-3600x

The tease in question was posted by an AMD-contracted Sales agency in South Korea, which launched a campaign inviting users to guess Cinebench scores for upcoming AMD processors: namely, the Ryzen 7 3700X and Ryzen 5 3600X - thus confirming the nomenclature for AMD's upcoming CPUs. The contest finishes on December 14th, and is basically asking users to take a gander on scores for unreleased CPUs - promising prizes of said CPUs when they launch.

[SteelCrysis]

https://www.tomshardware.com/news/amd-ry...38310.html

Russian price tracking website e-Katalog has listed AMD's complete upcoming Ryzen 3000-series processor lineup. It’s notable that E-Katalog doesn't sell products. Instead, it lists pricing for a range of items, like household and computer equipment, electronics, home, and office products, much like PCPartPicker.

While the specifications fall in line with the recent AdoredTV leak, these listings could be placeholders and should be taken with a grain of salt. However, if there is any truth to the listings, AMD could really shake up the processor market with the new Ryzen chips.
...
According to the listings, the Ryzen 3 3300 runs at 3.2GHz with a 4GHz boost clock and 50W TDP (thermal design power), while the Ryzen 3 3300X variant boasts a 3.5GHz base clock and 4.3GHz boost clock with a slightly higher 65W TDP. They are expected to cost no more than $130, which is almost unthinkable for a hexa-core chip.

The Russian website lists the Ryzen 5 3600 at 3.6GHz with a boost clock that reaches 4.4GHz. As for the Ryzen 5 3600X, the chip reportedly has a 4GHz base clock and 4.8GHz boost clock. The non-X variant comes with 65W TDP and the X variant with a 95W TDP.

Going up the Ryzen 3000-series ladder, the Ryzen 7 3700 is listed with a 3.8GHz base clock, 4.6GHz boost clock, and 95W TDP. The higher-end Ryzen 7 3700X flaunts a 4.2GHz base clock, 5GHz boost clock, and 105W TDP.

Lastly, the Ryzen 3800X ticks at 3.9GHz with a 4.7GHz boost clock and is listed with a 125W TDP. On the other hand, the Ryzen 9 3850X is listed with a 4.3GHz base clock and a shockingly-high 5.1GHz boost clock with a 135W TDP rating. However, e-Katalog only listed the first, which could reinforce the rumor that AMD will release the Ryzen 9 3850X at a later date (May 2019).

[SteelCrysis]

https://www.tomshardware.com/news/amd-ry...38399.html

Possibly matching Intel's single-threaded performance is a watershed moment for AMD, as that type of workload has long been one of the few areas where the Ryzen processors lagged behind Intel's models. But that isn't all. Su also pointed out that the denser 7nm node allowed the Ryzen processor to consume less power, which ultimately equates to heat, than the Core i9-9900K.

The AMD processor pulled ~30% less power than the Core i9-9900K. That's a tremendous advantage over Intel's processors, which have gained a reputation for requiring high-end motherboards, power supplies, and coolers to extract the optimum level of performance. The Ryzen processors' impressive achievement means that it could be much cheaper to build full systems with AMD's Ryzen 3000-series processors, thus offering a similar level of performance while maintaining the value advantage.
...
Third-gen Ryzen is coming to market in the "middle of 2019," and Su said that the company would share more details as it comes closer to launch.

https://www.techpowerup.com/251379/amd-3...e-chiplets

On close inspection of the substrate, we find that while the I/O controller die is somewhat centrally to the side of the package, the sole 8-core CPU chiplet is not located at a similar position (think Intel "Clarkdale" MCMs). On zooming in further, we find that just south of the 8-core CPU chiplet die, there appear to be blank bumps protruding over an area similar to that of a chiplet covered up by the outer layers of the substrate, leading us to conclude that the AM4 package is capable of three dies, an I/O controller, and two 8-core CPU chiplets. There very much will be a 16-core/32-thread Ryzen for the AM4 platform, and it's only a question of when.

The 16-core Ryzen AM4 MCM will be similar in concept to the larger 64-core SP3r2 EPYC/Threadripper MCMs: the CPU dies only pack the CPU cores and an InfinityFabric interface, while the I/O controller die is wired to multiple CPU dies, and manages the memory, PCIe, and SoC connectivity of the processor.
...
AMD's engineering bravado with "Matisse" also unlocks the possibility of the Ryzen "Raven Ridge" APU successor being an MCM with one 8-core chiplet, and an oversized I/O controller die that packs a "Vega" or "Navi" based iGPU, in addition to memory, PCIe, SoC, and the works. Dies on that package could be arranged differently from this.

https://www.tomshardware.com/news/amd-ry...38401.html

But after speaking with several motherboard vendors here at CES 2019, we've learned that many of them have successfully tested PCIe 4.0 on 300- and 400-series AMD motherboards, meaning that the feature could be enabled with a simple BIOS update, at least partially.

Our sources tell us that after unlocking the feature via a BIOS update, the older motherboards supply a PCIe 4.0 x16 connection to the first slot on the motherboard, but the remainder of the slots revert to PCIe 3.0 signaling rates. That's because any trace routing on the motherboard that exceeds six inches requires newer redrivers and retimers that support PCIe 4.0's faster signaling rates. That means the PCIe slot nearest to the CPU will easily support PCIe 4.0, while the other slots, including M.2 ports, will run at a PCIe 3.0 signaling rate.

The 500-Series chipsets will consume more power than the 28nm chipsets used on current AM4 motherboards, but that's because the 500-series chipsets also support PCIe 4.0. We weren't told the specific lane allocations of the new chipset, but those faster lanes will be useful for numerous types of secondary I/O devices.

Even though multiple board partners have tested PCIe 4.0 on previous-gen chipsets, it remains to be seen if AMD will allow them to expose that functionality via BIOS updates. Our sources tell us that AMD can simply lock out that feature and that the fate of PCIe 4.0 support on 300- and 400-series motherboards haven't been communicated to them yet.

[SteelCrysis]

https://www.tomshardware.com/news/amd-ry...38401.html

Update: We spoke with AMD representatives, who confirmed that 300- and 400-series AM4 motherboards can support PCIe 4.0. AMD will not lock the out feature, instead it will be up to motherboard vendors to validate and qualify the faster standard on its motherboards on a case-by-case basis. Motherboard vendors that do support the feature will enable it through BIOS updates, but those updates will come at the discretion of the vendor.

[SteelCrysis]

https://www.gamersnexus.net/news-pc/3428...ng-asmedia

Thus far, the information we have obtained regarding Ryzen 3000 points toward a likely June launch month, probably right around Computex, with multiple manufacturers confirming the target. AMD is officially stating “mid-year” launch, allowing some leniency for changes in scheduling, but either way, Ryzen 3000 will launch in about 5 months.

The biggest point of consideration for launch has been whether AMD wants to align its new CPUs with an X570 release, which is presently the bigger hold-up of the two. It seems likely that AMD would want to launch both X570 motherboards and Ryzen 3000 CPUs simultaneously, despite the fact that the new CPUs will work with existing motherboards provided they’ve received a BIOS update.

https://www.gamersnexus.net/news-pc/3429...ram-prices

A quick update on our previous X570/PCIe story: First off, as pointed out previously, "chipset" wasn't really the right language to use when referring to the X570's induction of "parts" of Epyc -- it's just PCIe 4.0, more or less, that's moving over. We had a few people reach out to us and confirm that the chipset will almost certainly be running PCIe 4.0, responsible for the power requirement increase and for potential logistical challenges.

Separately, on core counts, our engineering contacts within the industry have informed us that we should expect 16C and 12C CPUs with Ryzen 3000, in addition to the usual 8-core parts. It's just a question of if those launch altogether or independently.

[SteelCrysis]

https://www.tomshardware.com/news/ryzen-...38474.html

It's encouraging to hear that AMD's next round of Ryzen chips will not need specific new software enhancements to accommodate the design, as that was a key concern when the first-gen chips arrived. AMD has said that it will release the new third-gen Ryzen processors in mid-2019, which lines up nicely with Computex. As with AMD's previous big launches, we expect more information to come to light slowly in the intervening months as the company builds the hype for its newest round of processors.