Quote:The CPU itself is an engineering sample, coded 2D2701A9UC9F4_32/27_N. Videocardz did a pretty god job on explaining what the nomenclature means, but for now, we do know this sample seems to be running at 2.7 GHz Base, and 3.2 GHz Boost clocks (not too shabby for a 12-core part, but a little on the anemic side when compared to previous reports on a 16-Core chip from AMD that would run at 3.1 GHz Base and 3.6 GHz Boost clocks.) What seems strange is the program's report on the available cache. 8x 8 MB is more than double what we would be expecting, considering that these 12-core parts probably make use of a die with 3 CCX's with 4x cores each, which feature 8 MB per CCX. So, 3 CCX's = 3x 8 MB, not 8x 8 MB, but this can probably be attributed to a software bug, considering the engineering-sample status of the chip.

Quote:These could include 12-core and 16-core parts, and the picture is getting clearer with an exclusive report by Turkish tech publication DonanimHaber. The biggest revelation here that the 12-core and 16-core Ryzen processors will be multi-chip modules (MCMs) of two "Summit Ridge" dies. The 12-core variant will be carved out by disabling 1 core per CCX (3+3+3+3).

Another revelation is that the 12-core and 16-core Ryzen processors will be built in a new LGA package with pin-counts in excess of 4,000 pins. Since it's an MCM of two "Summit Ridge" dies, the memory bus width and PCIe lanes will be doubled. The chip will feature a quad-channel DDR4 memory interface, and will have a total of 58 PCI-Express gen 3.0 lanes (only one of the two dies will put out the PCI-Express 3.0 x4 A-Link chipset bus). The increase in core count isn't coming with a decrease in clock speeds. The 12-core variant will hence likely have its TDP rated at 140W, and the 16-core variant at 180W. AMD is expected to unveil these chips at the 2017 Computex expo in Taipei, this June, with product launches following shortly after.

Quote:Stay tuned for our full review April 11.

(03-30-2017, 07:37 PM)SteelCrysis Wrote: https://www.techpowerup.com/231970/ashes-of-the-singularity-escalation-update-brings-improved-performance-to-ryzen
It's safe to say that AMD is bribing Stardock. There's no other explanation for this.

Quote:One of the lessons learned during the Ryzen 7 launch was that AMD's temperature sensor output has its share of problems. The company configured some of the models to report an additional 20°C on top of the actual temperature in order to guarantee adequate cooling. We’re still puzzled by this choice.

Nevertheless, we're accepting it as reality and going all-out on cooling for Ryzen 5's introduction. Alphacool offers a modified version of an industrial compression cooler under the name "Eiszeit 2000 Chiller." It has a powerful pump, and it’s compatible with the usual 1/4" connectors, thus making the technology accessible to enthusiasts like us.

We’re cooling the processors by connecting an Alphacool Eiszeit XPX CPU block directly to the Eiszeit 2000 Chiller. The compression cooler’s reservoir has a capacity of almost eight liters, serving double duty as a storage tank and compensating reservoir. This provides a large thermal buffer. At 40L per minute, the flow rate is high enough for our purposes as well.
...
At first, we just wanted to see if we could explore how AMD's XFR feature behaved under perfect conditions, and how leakage current would compare to measurements taken with other cooling solutions. However, the first test run delivered a huge surprise:
At just under 17°C, the Ryzen 5 1500X’s reported Tctl temperature values at 19W total power consumption of the CPU package come in a full 3°C below the cooling fluid’s temperature of a constant 20°C!

We don’t really need to explain why these results can’t be correct. Consequently, we dug deep into our toolbox and found a liquid metal pad that was originally developed as a substitute for thermal paste. Our potent cooler kept the temperatures well below the pad’s melting point, so that it didn’t fuse with the processor’s heat spreader or the water block’s heat sink. We placed a thin strip between the heat spreader and the heat sink during our water-cooled measurements, making sure to cut the strip thin enough for it not to result in a measurable change in cooling performance. We used a tiny copper plate milled into a very thin strip for the stock cooler measurements.

The plan was simple: find the exact difference between the temperatures reported by the processor and the temperatures we measured for the processor’s heat spreader with the help of our constant cooling fluid temperature setup and infrared camera. Unfortunately, we found that the resulting differences between the temperature curves weren’t consistent and, consequently, still didn’t make any sense. There were two points at which the temperature curves changed drastically and clearly diverged from a linear pattern.

Based on these findings, we hypothesized that the behavior of the processors’ temperature sensors changed once certain temperature thresholds were reached. If this was the case, we could predict the sensors’ behavior for different temperature ranges and correct the temperature readings.

For our second round of testing, we compared the Tctl values to actual power consumption. A software tool developed in-house allowed us to generate different loads over long time periods, avoiding any jumps or delays that could skew the results. The tool would do its best to spread the loads equally across all possible threads. The results proved our previous hypothesis:
The reported temperatures are complete nonsense below ~20 to 25W. It’s also plain to see that there are significant differences between the Tctl reporting patterns of AMD’s Ryzen 5 and 7.
In general, we saw that upwards of 20 to 22W, there was a difference of approximately 4 to 6°C between AMD’s Ryzen 5 and 7 when using our cooling solution with its constant fluid temperature. The Ryzen 5 processors’ sensors consistently displayed higher temperature readings than the Ryzen 7 processors at the same power consumption level!

The additional 20°C that AMD builds into some of its Ryzen processor models are already taken into account in this calculation.

One more thing to consider is the potential difference in heat transfer between components for our particular cooling solution. Plugging in numbers based on our experience and then comparing the final figures to our infrared measurements yields the following results for the three wattage ranges:
Across the entire wattage range, the Ryzen 5 produces a difference of approximately 8°C. That's pretty significant. We ran several control tests and got the same results every time.

The next question is whether Ryzen 5's temperature displays too high, or if Ryzen 7 is too low by 4 to 6°C.

The following graph shows the CPU heat spreader’s temperatures over a time period of approximately 30 minutes in degrees Celsius. The Ryzen 7 1800X’s leakage current is approximately 58 to 60W. Comparing that to the curve in the graph above shows that Tctl and heat spreader temperatures should not be the same. The expected difference of approximately 3 to 5°C between the Tctl and heat spreader temperatures is reached by the Ryzen 5 1500X, though. These finding are consistent with what we thought.
The tentative bottom line seems to be that all Ryzen 7 processors report their temperatures as being approximately 4 to 6°C lower compared to the two Ryzen 5 processors at the same leakage current level.

The bend in the curve at approximately 50W becomes more interesting once we get to the gaming power consumption results. The smaller CPUs may stay in this range during gaming, which means that it might be beneficial to take our results into account when setting fan curves. Most people don’t have a compression cooler at home, after all. The lower your cooling performance, the more important it is to change sensor characteristics!

We measured the Ryzen 5 1500X with AMD's stock air cooler to illustrate our point further. Its fan was set to a fixed speed we chose after first running a stress test. The curve's two bends are easy to see in spite of the curves being much steeper. A closer look shows that the total power consumption increased by approximately 5W. This is due to increased leakage currents that accompany the much higher temperatures.

We stay with the temperature results a little while longer and take a look at all of the Ryzen processors’ temperatures. These are corrected for the temperature differences due to the three wattage ranges and the two processor lines.

Here’s a computational example for the Ryzen 7 1800X during the stress test:

67°C (Tctl, Sensor) - 20°C (AMD Ryzen Offset) + 4°C (Additional Offset for Ryzen 7) = 51°C

If you feel confused by all of this, then you’re certainly in good company. Nobody knows what prompted AMD to take the approach it did. If we didn’t use our offsets, then Ryzen 7 would be much cooler than Ryzen 5 at the same level of leakage current, which is physically impossible.

Before we close, we’d like to point out one more time that the temperature differences and offsets only apply to our particular test setup. Different hardware may very well affect behave some other way. The general principle shouldn’t change, though. A final judgement can’t be reached before these – admittedly elaborate – measurements have been repeated with a wide variety of setups.

Doing this just wasn’t possible under the time constraints of a product launch. The measurements took us three days for just the one setup.

Quote:Due to its identical clock rates, AMD's Ryzen 5 1600X demonstrates similar performance as the Ryzen 7 1800X in lightly-threaded content creation and productivity tests. The 1600X also outpaces the Ryzen 7 1700 in a great many scenarios where its higher frequency weighs heavier than its core count deficit. This makes the 1700 a tougher sell.

Intel’s Kaby Lake-based processors beat Ryzen 5 1600X in lightly-threaded applications where they can leverage superior IPC throughput. But the 1600X’s extra cores/threads turn the tables in software well-optimized for multi-core CPUs. Surprisingly, the 1600X even rivals the 1700X in certain scenarios. That paints a pretty convincing picture for a budget workstation chip, especially in light of the incredible price-to-performance ratio compared to Intel’s Broadwell-E line-up.

The Ryzen 5 1600X also makes a compelling argument against purchasing the 1700 for your next gaming PC. Six nimble cores regularly match or beat AMD's budget-oriented eight-core model. It'd be easy to speculate that, due to the 1600X’s lower core count, less inter-CCX traffic unburdens the Infinity Fabric and provides more competitive performance. We'll explore this in more depth later. For now, we think it's safe to say there's little reason for enthusiasts to splurge on the higher-end Ryzens for gaming, especially when the dual-CCX die overclocks similarly, regardless of configuration.

But don't forget the Core i5-7600K. It's a capable gaming processor. And although the 1600X challenges it in much of our benchmark suite, the Core i5 still comes out on top at stock settings. Further, overclocking Kaby Lake opens up a sizeable advantage that AMD cannot overcome, given limited frequency headroom. We expect Ryzen's overclocking potential to improve as GlobalFoundries' 14nm process evolves, but Intel's isn't sitting by idly, either.

AMD is still working on improving the utilization of Ryzen 5's resource-rich architecture, and a few gaming titles make it apparent that this is still a work in progress. We’ve seen several developers come forward with Ryzen-specific patches, and if AMD's gets its wish, more will follow suit. In the meantime, AMD developed its own Windows power profile to combat the performance issues we observed back when Ryzen 7 launched. Unfortunately, some of the other workarounds we played with may not apply as universally to Ryzen 5. For instance, disabling SMT on the 1600X resulted in choppy frame delivery, apparent in our frame time charts.

At least overclocking is allowed on inexpensive B350-based motherboards. This makes Ryzen 5 a much better value than Broadwell-E, arguably superior to Ryzen 7 for mainstream gamers, and at least competitive with Kaby Lake. As shown in our Infinity Fabric-oriented tests, though, you'll want a fast memory kit to achieve the best gaming performance.

Ryzen 5 1600X provides a tremendous price-to-performance ratio for budget workstations, rivaling Core i7-6800K. It also facilitates playable performance in games (though it still lags Kaby Lake-based Core i5s more often than not). Considering what Intel charges for its Core i5-7600K, we'd certainly like Ryzen 5 1600X a lot more for gaming if it debuted at a lower price. Much of the Ryzen tapestry is woven using value as its thread. But it's hard to keep that story together when Ryzen 5 1600X sells for $249 and Core i5-7600K goes for $240. With that said, professionals on a budget are far more likely to jump on a potent six-core chip like the 1600X when it's able to beat the $450 Core i7-6800K.

Quote:AMD reported its financial results for the first quarter of 2017 today. Despite an 18% year-over-year increase in revenue to $984 million, the company still reported an operating loss of $29 million. AMD further recorded a net loss of $73 million for the quarter.

The Computing and Graphics division hauled in $593 million of that revenue after the launch of the Ryzen 7 and Ryzen 5 CPU lineups. That figure is up an impressive 29% year-over-year, and the company said it led to a better-than-expected sequential seasonal decline. Client-product average selling prices were up both year-on-year and sequentially, a change the company attributes to higher desktop processor ASPs. Still, the division lost $15 million, although that figure is down from $70 million a year ago.

Quote:As we and other sites have demonstrated, using fast memory can result in a significant performance boost on systems with AMD's Ryzen processors. It can be a trying task to actually get your Ryzen rig to make use of fancy overclocker memory, though. AMD's 1.0.0.4 AGESA microcode update seemed to help matters somewhat, and some users reported improved memory speeds after that update. Now we have word—by way of a Gigabyte employee's forum posts—that AGESA update 1.0.0.5 is on its way with further improvements for memory compatibility.

Gigabyte rep Matt posted in the company's forums last Wednesday confirming that the company is working hard to use the new AGESA microcode in its BIOS updates to help improve its AM4-platform motherboards. The new updates should include fixes to IOMMU support, "soft brick" issues, and the ability for the CPU to enter lower power states when it's overclocked. Matt stated that the firmware updates would include AMD's latest AGESA microcode version 1.0.0.6. However, last Friday he posted in the same thread correcting himself and stating that the AGESA update would be version 1.0.0.5.

Either way, he says that the microcode update should enable access to "20+ memory registers" in the Ryzen CPUs' memory controller, and that the change should improve memory compatibility. It seems unlikely to us that these registers were actually disabled, but perhaps they were locked down and unavailable for modification. Matt says the BIOS updates for Gigabyte's boards should be available this week.

It's fairly safe to assume that Gigabyte isn't the only company readying BIOS updates with the new AGESA microcode. Asus, MSI, and all the other manufacturers are certainly working hard on new firmware as we speak.

(05-03-2017, 03:06 AM)SteelCrysis Wrote: https://www.techpowerup.com/232955/amd-increases-its-market-share-on-the-back-of-strong-ryzen-sales
I don't know if this will be long-term, but it's an encouraging sign so far.

Quote:Today is an eventful day in the tech world, with two high-impact leaks already offering themselves up to our scrutiny. We had previously covered AMD's upcoming HEDT platform, based on the company's new X399 chipset, as having a quite distinctive lineup of processors, with not only 16 and 12-core offerings hot on foundries presses', but also some 14-core, 28-thread chips as well. Now, a leak has apparently revealed the entire Ryzen HEDT platform, whose processor marketing name, Ryzen 9, sounds really close to Intel's Core i9.

AMD's offerings look to offer an edge at least on core-count, with the Red team's top offerings, the Ryzen 9 1998X and Ryzen 9 1998, bringing in a game-changer 16 cores and 32 threads to the table. Perhaps even more importantly, we have to mention that the 1998X (these names, if true, are quite a mouthful, though) achieves a 3.5 GHz base, 3.9 GHz boost clock, which owes nothing to AMD's Ryzen 7 1800X consumer flagship CPUs. Rumors of AMD's frequency demise on higher core-count Ryzen CPUs have been greatly exaggerated, it would seem. And did I mention that these chips are coming with a TDP of 155 W - 5 W lower than Intel's purported 12-core, i9-7920X offering? Consider that for a moment.

Quote:To start the show, AMD CEO Lisa Su unveiled the new EPYC data center processor. EPYC appears to be the new branding for the Naples data center processors that AMD announced last year. These 32-core/64-thread processors support 128 PCIe 3.0 lanes, and AMD has designed them specifically to interact well with its Vega GPUs.

Su noted that the copious PCIe connectivity will allow the dual-socket server platforms to support more GPUs than Intel's competing products. The processors support eight memory channels per socket, for a total of 16 DDR4 channels and 32 DIMMs in a two-socket server (up to 4TB of memory).

AMD also displayed EPYC with the heatspreader removed, and we spotted four 8-core Zeppelin die on a single package (MCP). These four die are glued together with AMD's Infinity Fabric.

AMD also announced its Threadripper CPUs for the high-end desktop segment. These 16-core/32-thread processors will debut this summer, but the company didn't provide any further details. Pricing will play an important role, but frankly, 16 cores is plenty exciting. Intel's 10-core/20-thread i7-6950X Broadwell-E weighs in at $1,700, and considering AMD's Ryzen pricing history, we can expect a much lower price point than that to get more cores.
...
AMD announced that the Ryzen Mobile products will feature on-die Vega graphics, which is a great indicator that we can expect Vega integrated graphics on the forthcoming APUs, as well.

AMD also provided a few rough details, such as 50% more CPU performance than the 7th-generation (A-Series) APUs and 40% more GPU performance. AMD also claimed that they'll consume 50% less power, which is attractive for the mobile segment.

Finally, as encouraging as many of the announcements are, AMD has more to come. The company divulged that it will debut its Zen 3 products on Global Foundries' second-generation 7nm+ process in 2020. The first generation 7nm process will obviously debut in the interim, but the company did not provide a firm timeline for it.

Quote:ASUS, through its ROG (Republic of Gamers)brand, has started teasing what is to be one of the first Ryzen-powered gaming laptops. Other than Ryzen's circular orange logo and the ROG brand, the video doesn't offer any specifics of what hardware rests under the hood. The clip includes the words "something has awakened," and the post is accompanied by the hashtag #Computex2017.

There are a number of things this teaser could mean. Ryzen and Vega-powered APUs aren't expected for some time, but this could be a teaser for an advance announcement for such a product taking place in Computex. The other, more likely scenario, is that ASUS will be making use of AMD's recently-announced Ryzen mobile solutions, though this would mean launch of this laptop will take place closer to the end of the year.

Quote:The 1400 is the first Ryzen part that doesn't do justice to the innovation AMD put into these chips, which leaves a bad taste in the mouth. The unlocked multiplier which lets you pull it up to 1500X performance is its only redeeming feature.

Quote:At its Computex 2017 event, AMD announced the market availability of EPYC enterprise processors based on the "Zen" architecture. The chips will be available from June 20, 2017. The company didn't get into the specifics, however we know that EPYC consists of chips with up to 32 cores, up to 64 threads, up to 32 MB of L3 cache, and a vast 112-lane PCIe root complex. There could be 1P and 2P models.

Quote:AMD demonstrated ThreadRipper's performance in a few scenarios, including the now-famous Blender rendering test of the Ryzen CPU and logo. The render required a mere 13.04 seconds, which is much faster than the 36 seconds AMD demonstrated with the Ryzen 7 1800X prior to its launch.

ThreadRipper's copious PCIe lane allocation means it can support up to four GPUs with no restrictions, such as bridge chips, which is a real win for the workstation segment, not to mention enthusiasts. To highlight AMD's newfound PCIe lane advantage, the company employed four Vega Frontier Edition GPUs during a real-time Blender render. The workload spread between the GPUs and the CPU using AMD's Radeon Pro plugin, and we could see that several of ThreadRipper's cores went unused, while the remainder hovered in the 20-25% range.

Quote:The company didn't disclose cache structure, or pricing. We expect the 10-core Threadripper models to take on Intel's 4-thru-6-core "Kaby Lake-X" and "Skylake-X" parts; the 12-core Threadripper on Intel's 8-thru-10 core parts, the 14-core Threadripper on Intel's 10-thru-12 core parts; and the 16-core Threadripper, bolstered by its high clock speeds, on Intel's 16-thru-18-core parts.

Quote:AMD recently cut the price of its current flagship desktop processor Ryzen 7 1800X from its USD $499 launch price to $469. At the time, it left prices of the Ryzen 7 1700 and Ryzen 7 1700X untouched. It looks like the two received small price-cuts as well. The Ryzen 7 1700X is now priced at $349 in leading online stores, down from its launch price of $399. The Ryzen 7 1700 (non-X), on the other hand, is now selling for $319, down from its launch price of $329. The two cuts may seem minor, but could help AMD turn up the heat against Intel's Core i7-7700K and its upcoming "Kaby Lake-X" Core i7-7740X and i5-7640X.

Quote:All told, the shortcomings of AMD's Ryzen CPUs mostly fade away when one isn't paying $300 or more for a chip. The Ryzen 5 1600X delivers Ryzen 7 1800X-class gaming performance for half the money, and its productivity performance is unprecedented for its price. The Ryzen 5 1600 and Ryzen 5 1500X also bring new levels of multithreaded performance to their price points. PC builders have real choice again in sub-$300 CPUs with the Ryzen 5 family, and for those who lean hard on their systems, our tests show their choice should be AMD.

Quote:First of all, we have a report suggesting a price of $849 USD. This is the cheapest 16 core 32 thread model from AMD. Furthermore, we can expect faster models with full-fledged features to cost a bit more. In addition, the fact that this is the entry level 16 core model also means there will be more than one 16 core Threadripper coming out. Secondly, to go along with Threadripper, we have the new X399 platform.
...
Due to the MCM design, AMD is able to reduce their costs quite a bit. At $849, that is just a tad over half of what a 16 core Intel CPU will set you back. While AMD is dramatically cheaper, the chips are unlikely to match Intel in performance. This is due to the MCM design, as well as the remaining gap in architecture performance. At half the price though, it will be interesting to see how many users opt to save an extra $800 to spend on other components.

Quote:The rest of the lineup is detailed as well, but as this is a leak that is blatantly admitting to violating an NDA, it goes without saying this could be nothing more than a fabrication. Take it with your usual dose of healthy skepticism.

For the full details of the leak (including the rest of the expected lineup), you can view the source link below.

Quote:AMD is readying a new stepping of its 14 nm "Summit Ridge" eight-core CPU silicon, which powers its socket AM4 Ryzen processors, according to Canard PC. The new B2 stepping reportedly addresses a lot of hardware-level errata which cannot be fixed merely by AGESA updates. According to Canard PC, the changes seem to be focused on the uncore components of "Summit Ridge." Typically, uncore refers to the integrated northbridge, which includes components such as the memory controllers, PCI-Express root complex, etc.

Quote:AMD has identified the single-socket server as a potential high-growth market. Roughly 25% of two-socket servers have only one socket populated, so offering a capable single-socket platform reduces unnecessary redundancies (such as sockets, networking components, and power supplies), thus saving on up-front cost, power, space, and cooling expenses. AMD feels it has an advantage in this burgeoning space due to EPYC's hefty allotment of I/O, memory bandwidth, and memory capacity. As such, the company created three single-socket SKUs that feature the same connectivity options and a TDP range of 155W-180W. We also noticed that the highest frequency of the lineup was at 2.4GHz on the 7351P.

Again, these value comparisons are derived from a data set that AMD adjusted to remove what it considers an unfair advantage.

...
What use is all of the fancy technology if it isn't secure? Well...none. To that end, AMD has a robust set of security features that are all controlled by a sandboxed ARM processor on the SoC package. This separates the security apparatus from the host operating system/hypervisor and provides hardware-based memory encryption, which is useful in multi-tenancy environments, among other features.

AMD's re-entrance into the server market brings about understandable concerns about the ecosystem. At the end of the day, most will purchase systems from OEM providers, and administrators expect rock-solid support with enterprise-class applications. AMD's been hard at work on the enablement front and has amassed a solid set of launch partners for both hardware and software.

AMD has also developed a robust set of features that should further its objectives in the data center. The company noted that it designed the architecture from the ground up for data center workloads, which isn't a surprising admission. AMD's reentrance into the server market will be a long process, which company representatives have repeatedly acknowledged, but it does look promising.


Four of the high-end SKUs, along with several OEM systems, are available today. The remainder of the product stack, and further expansion of OEM's servers, comes in July.

Quote:AMD has robust industry relations with the game-console manufacturers and has monopolized semi-custom SoC manufacturing for that market segment. It also pushed the software industry for favorable yet open standards, which its client hardware is inherently good at. The company must now direct its efforts towards building similar industry relationships with the most popular server OEMs, who not only make products based on EPYC, but also allocate their own marketing budgets to promoting EPYC-powered solutions to their longstanding or prospective customers. The onus is also on AMD to ensure that EPYC chips live up to the reliability standards the industry has come to expect of competing Xeon chips, and to provide a robust after-sales support net to customers.

AMD must also nurture a vibrant ecosystem built around EPYC, which includes not just hardware and OEMs marketing them, but also the matter of pioneering software standards in the enterprise segment. This must include not just open- and open-source standards, but proprietary ones from the likes of Microsoft or Novell. Fortune 500 companies need to know that EPYC-powered machines can be relied upon to power their businesses. AMD should also build the brand in the high-performance compute (HPC) arena with turnkey solutions that combine its enviable GPGPU leadership with its homebrew EPYC processors. We're inclined to think NVIDIA isn't fooling around by investing billions of dollars into machine-learning, and AMD is in a position to combine its CPU, GPU, and software resources into uniform and reliable solutions driven by open standards. The PC and server CPU industry dominated by Intel is finally getting a much needed shakedown, and it will only benefit the industry at large.

Quote:This confirms some of the rumors we’ve seen around Ryzen 3 as well, assuming AMD keeps parity between the pro and consumer variants. The Ryzen 3 chips will be native quad cores compared with the Core i3’s dual-core + Hyper-Threading configuration.

It’s less clear how this will impact performance. Intel’s single-threaded perf is still a bit higher than AMD’s, clock-for-clock, but AMD’s SMT implementation has generally given it a larger boost than Intel gains from Hyper-Threading. It’ll be quite interesting to see how the two compare here, since AMD will still have a physical core advantage, but not the SMT implementation that has made chips like the Ryzen 5 1600X able to punch well above their weight class.