Ryzen vs i7 (Mainstream); Threadripper vs i9 (HEDT); X299 vs X399/TRX40; Xeon vs Epyc

After adjusting for that, you still see amd with a slight advantage in most use cases, just not as large as otherwise.

Edit: to be fairer, this compared Epyc to v4, not SL-X, which may give the performance edge to Intel this gen. Needed to correct my statement.

Sent from my SM-G900P using Tapatalk

 

It's not a matter of making an 'adjustment' though, no matter how simple it is to do or how much performance is still there...

AMD is flat out lying to it's prospective customers from the get go.

I wouldn't partner up with them if they paid me at this point.

 

They said where the numbers came from and what adjustment was made. That is not lying. Read my edit above, that is the real point you should hit on!

Sent from my SM-G900P using Tapatalk

 

https://arstechnica.com/information...mds-zen-steps-into-the-server-room-with-epyc/
https://hothardware.com/reviews/amd-epyc-7000-series-processor-and-platform-details
http://www.pcworld.com/article/3201...chip-to-take-on-intel-in-the-data-center.html
https://www.hardocp.com/news/2017/06/20/amd_reveals_epyc_7000_server_cpus
https://nl.hardware.info/nieuws/525...ijgt-concurrentie-in-meest-lucratieve-segment

 

Yeah; technically you're right. But hiding this 'info' in the end notes doesn't build any brownie points for AMD.

Effectively lying (and hoping no one calls you out).

I'm... calling them out (for the benefit of others here...).

 

Many articles noted it, which I'm very happy for as otherwise, unless you reviewed all materials at the event, you would be caught believing something that doesn't reflect reality.

Now, of course, with time AMD may see its own c++ optimizations, but it shouldn't wholly be written off what they did here.

It also means Intel retains the crown, meaning retaining a lot of HPC and mission critical. But in cloud, all have signed on and many OEMs will have offerings from AMD (HP presented there and Dell already announced).

But let's give a hand to the journalists on this one!

Sent from my SM-G900P using Tapatalk

 

I think that by discounting Intel's 'scores' is going to hurt them - a lot.

Serve the home is comparing to circa 2012 Q1 v1 processors*** and states Epyc is 'better'.

'All' cloud have signed on? I highly doubt that.


*** 4 cores vs. 64 cores and half a decade later...

 

Boy, you can't read can you? Go to the sources say quote from the anandtech article, the last wrote in that post (#299). I'll wait...

Sent from my SM-G900P using Tapatalk

 

Thanks for waiting...

I can read. lol... But you don't provide links to backup your claims.

No offense; but I'm not believing anything you write... about the sources that AnandTech supposedly has... that states something that I just know (call it a gut feeling) to not be true...

Did they (all) just sign on to test Epyc? Sure; I'll be able to believe that if it's verified.

Did they (all) sign on to the Epyc bandwagon and ordered replacements from AMD - just because? Nah...

 

No, it doesn't take away credibility, as always (and mentioned) we need to rely more on third party benchmarks. Your last statement is comical, for if they are relying on a compiler from Intel C++ and it just is not an efficient one they can switch compilers. I did expect more of a logical argument from you.

I would never go on the boisterous claims only of an OEM. This does look promising but we now need to see the real tests!

Edit; Both Intel and AMD have their fair share of dishonesty in marketing. This has been true forever. I again just want to see the real tests to determine the strengths and weaknesses. We can then determine the real values.

Although I do think that overly generous cut into SPEC is the reason we did not get a live stream nor a Q&A session. Not sure I would want to be in that hot seat either.

 

Definitely. GCC is more widely used, which started the estimate, although we'll know more soon, hopefully. Many companies said their review systems were on the way.

Also, it needs noted only 4 chips launched today, with the other chips available in late July.

Sent from my SM-G900P using Tapatalk

EDIT:

1. Based on estimated SPECint®_rate_base2006 scores. 2P Intel Xeon E5 scores other than E5-2699Av4 were derived by AMD from the following ICC compiler-based test results published at www.spec.org, multiplied by 0.575 to convert from the ICC compiler to the GCC-02 v6.1 compiler used for EPYC testing: E5-2698v4=1620, E5-2695v4=1440, E5-2680v4=1270, E5-2650v4=1000, E5-2640v4=865, E5-2630v4=814, and E5-2620v4=683. The following EPYC 2P scores are projections made by AMD labs (measured test data for these pre-production products is not currently available): EPYC 7301=845, EPYC 7281=760, and EPYC 7251=485. All other scores based on AMD internal testing. 2P E5-2699A v4 in Intel Server System R1208WT2GSR scored 943 on 6/5/2017, with Ubuntu 16.04, GCC-02 v6.3, 512 GB (16 x 32GB 2Rx4 PC4-2666 running at 2133), 1 x 500GB SSD. 2P EPYC 7601 in Supermicro AS-1123US-TR4 scored 1390, with Ubuntu 16.04, GCC-02 v6.3, 512 GB (16 x 32GB 2Rx4 PC4-2666 running at 2400), 1 x 500GB SSD. The following EPYC 2P scores tested using AMD’s “Ethanol” reference system with Ubuntu 16.04, GCC-02 v6.1, 512 GB (16x32GB 2Rx5 PC4-2667 running at 2400), 1 x 500GB SSD: EPYC 7551=1345, EPYC 7451=1218, EPYC 7401=1120, EPYC 7351=939. Pricing ranges based on Intel recommended customer pricing per ark.intel.com, and AMD 1Ku pricing.

 

AMD Epyc Info
https://www.amd.com/en/products/epyc

The Power of EPYC

AMD EPYC: Customer Testimonials

EPYC Customer Testimonial: Microsoft

EPYC Customer Testimonial: Samsung

AMD EPYC’s Secure Encrypted Virtualization (SEV) Feature Demo

EPYC Customer Testimonial: LexisNexis

EPYC Customer Testimonial: Baidu

EPYC Customer Testimonial: 1&1

 

AMD muscles in on Xeon’s turf as it unveils Epyc
Multiply a Ryzen by four and all of a sudden Intel has some real competition.
Check out web page for many more slides
https://arstechnica.com/information...-into-the-server-room-with-epyc/?sf90727933=1

"AUSTIN—Today, AMD unveiled the first generation of Epyc, its new range of server processors built around its Zen architecture. Processors will range from the Epyc 7251—an eight-core, 16-thread chip running at 2.1 to 2.9GHz in a 120W power envelope—up to the Epyc 7601: a 32-core, 64-thread monster running at 2.2 to 3.2GHz, with a 180W design power.

AMD initially revealed its server chips, codenamed "Naples," earlier this year. Since then, we've known the basics of the new chips: they'll have 128 PCIe lanes and eight DDR4 memory controllers and will support one or two socket configurations. With today's announcement, we now know much more about how the processors are put together and what features they'll offer.

The basic building block of all of AMD's Zen processors, both Ryzen on the desktop and Epyc in the server, is the eight-core, 16-thread chip. Ryzen processors use one of these; the Threadripper high-end desktop chips use two; and Epyc uses four. Each chip includes two memory controllers, a bunch of PCIe lanes, power management, and, most important of all, Infinity Fabric, AMD's high-speed interconnect that is derived from coherent HyperTransport.


Enlarge / Most of the Epyc SKUs.

FURTHER READING
AMD’s moment of Zen: Finally, an architecture that can compete
From our look at Ryzen, we already know that Infinity Fabric (IF) is used to connect two blocks of four cores (called "core complexes," CCXes) within each eight-core chip. IF is also used both to connect the chips within the multi-chip module (MCM), and, in two processor configurations, to connect the two sockets.

Within the processor, each chip has three IF links, one to each of the other three chips. Each link runs at up to 42GB/s in each direction. The speed of these links matches the 42GB/s of memory bandwidth offered by the two channels of 2,667MHz DDR4 memory that each individual chip supports, and what this means is that any one chip within the Epyc MCM can use the full memory bandwidth of the entire processor without bottlenecks. Accessing memory that's connected to a different chip will incur somewhat higher latency than accessing memory that's directly connected, but it comes at no bandwidth penalty.

In two socket configurations, there are four IF links between the sockets. Each chip in one socket is paired with a chip in the other socket, for four pairs total, with one IF link between each pair. This design means that accessing remote memory has, at most, a two-hop penalty and that there are multiple routes that data can use to move from a chip on one socket to a chip on the other. The cross-socket IF links are slightly slower than the internal ones, operating at 38GB/s bidirectional. This is because these links have higher error-checking overhead, which uses up some of their bandwidth.


Enlarge / Infinity Fabric interconnects.
Both the internal and external IF connections are power managed. If not much traffic is going across the links, the processor will cut back its performance and hence energy usage. Power not used on the links can instead be used for the cores themselves, with AMD saying that this power management can provide as much as an eight-percent improvement in performance per watt.

In total, each processor offers 128 I/O channels. In two socket configurations, 64 channels from each processor are used for Infinity Fabric connectivity, leaving an aggregate of 128 I/O channels still available. As such, both one socket and two socket configurations offer nearly identical I/O options. The main thing the I/O channels can be used for is PCIe connectivity, with up to eight PCIe 3.0x16 connections per system.

These can be subdivided all the way down to 128 PCIe 3.0 x1 links, and there's a good degree of flexibility to the possible configurations of PCIe lanes. Each chip can use eight of its links as SATA connections, too. This is one of the few areas where a two-socket system will give you more I/O capabilities; with two sockets, the chips would support a total of 16 SATA connections.

Epyc is designed as a system-on-chip. Many features that would typically need additional components on the motherboard have been integrated into what AMD calls the Server Controller Hub (SCH) within the Epyc processor itself. This includes four USB 3.0 controllers, serial port controllers, clock generation, and low-speed interfaces such as I2C. The one notable I/O component not in the processor is Ethernet; for that, you'll need a PCIe card or motherboard-integrated interface.

A scaled-up Ryzen?
In most other regards, Epyc is little different from a scaled-up Ryzen—not altogether surprising, given the common heritage. Ryzen features, such as individually adjusting the voltage on a per-core basis, are found in Epyc, for example.

Some of these features have an Epyc twist, however. Like Ryzen, Epyc can boost clock speeds depending on usage levels. The top-end 7601 part, for example, has a base speed of 2.2GHz, with an all-cores boost of 2.7GHz and a maximum boost of 3.2GHz. Ryzen's maximum boost is very limited, only applying with one or two cores active. Epyc's is a bit more versatile; that 3.2GHz can be reached with up to 12 cores active.

Epyc chips also offer two modes, set at boot time, that let you pick between consistent performance and consistent power usage. In performance mode, the chip will offer repeatable, consistent clock speeds and boosting, drawing more power as required. In power mode, the chip will tightly stick to an upper bound for power usage and cut performance, if necessary, to stay within that envelope. This isn't available on the desktop chips, where power constraints are relatively lax and governed more by the cooling system than anything else. But it is valuable in densely packed server racks, where the overall power draw of a rack is often constrained.

The on-chip power management will also strive to detect certain workload patterns and reduce clock speed accordingly. In workloads that cause bursts of activity followed by idle periods, Epyc will reduce the clock speed during those activity bursts. This will make them take a little longer and cut down the idle time. Because power usage tends to scale with the cube of the clock speed, AMD argues that this behavior can cause a net reduction in power usage; at maximum speed, any power saved during idle is more than offset by the extra power used during the activity bursts. So cutting that peak power draw will lead to an overall reduction in power usage, even if a core is idling less.

This stands in contrast to the normal "race to idle" behavior that's often used to reduce power usage, wherein a processor runs as fast as it can as briefly as it can, because idling is so overwhelmingly superior from a power-usage perspective. It might also have some latency impact since each burst of work will take a little longer to complete.

For Epyc, AMD is also promoting some features that appear to also be available in Ryzen (at least, there are firmware options to control them) but only make a great amount of sense in server configurations. For example, Epyc supports encrypted system memory. Each memory controller has an encryption engine, and it can transparently decrypt and encrypt everything it reads and writes from RAM. This can operate in two modes; a global mode, in which all memory is encrypted using keys generated by the processor, and a software-controlled mode that enables, for example, memory belonging to different virtual machines to use different encryption keys.

Epyc also supports data poisoning. Typically, when ECC memory finds an uncorrectable error, the default operating-system behavior is to bring down the entire machine. With data poisoning, the operating system can instead choose to crash only the process or virtual machine that contained the error, leaving the rest of the machine unaffected.


Enlarge / Epyc has tons of I/O.
Compared to the Broadwell-based Xeons that are currently on the market, Epyc looks very compelling. It offers considerably more I/O than Intel's chips (which only offer 40 PCIe lanes per chip), and it offers considerably more cores per socket. AMD's line-up is also much more consistent, with the same set of features available across the entire range (with some small exceptions; the company will have three single-socket-only chips, with model numbers ending in P).

The low-end parts don't omit any reliability or security features found in the high-end parts, making the only choice the number of cores and clock speeds that you need or can afford. In the very limited benchmarks AMD has demonstrated, Epyc 7601 handily beats a pair of Xeon E5-2699A v4 processors, Intel's fastest two-socket Xeons.

But Intel's new generation of Xeons built around the Skylake SP core are right around the corner. AMD says that it built Epyc not simply to beat Broadwell, but also Skylake. That comparison looks like it's going to be far more complex. AMD will certainly offer more memory bandwidth—Skylake-SP has only six memory channels to Epyc's eight—and AMD will likely offer more cores and threads per socket than Intel. But Skylake-SP's single-threaded performance is better than Zen's, and Intel's use of monolithic dies, rather than multi-chip modules, should give Intel's chips lower latency access to memory. Skylake-SP also includes new features such as AVX512, which may provide a healthy boost to number-crunching applications.

How this will all turn out remains to be seen; until Skylake-SP hits the market we have no benchmarks between the two, and we might well expect to see different winners depending on the workload being run.

Either way, though, one thing is clear: Intel has a level of competition that it hasn't had for a while. Epyc may not be the best choice for every workload, but it's sure to be the right option for many. While pricing hasn't been announced yet, we expect AMD to continue its trend of undercutting its larger competitor. Just as Ryzen has done on the desktop, Epyc is creating options in the server room."

Check out web page for many more slides

 

AMD EPYC 7000 Series Benchmarks Versus Intel Xeon - HotHardware


AMD Launches EPYC Enterprise CPUs Available Today
https://www.hardocp.com/article/2017/06/20/amd_launches_epyc_enterprise_cpus_available_today

AMD EPYC Launch
"It has been a long time since I went to an AMD CPU Launch and came away with a feeling that everything was looking up for AMD. While we all know that AMD's Ryzen 7 and Ryzen 5 CPUs have been a success in the enthusiast market, where the rubber meets road when it comes to making the big bucks is the datacenter. Gaining ground on Intel in the datacenter is what will "save" AMD. I know it, you know it, AMD knows it.

I was live-Tweeting and put up a news post with the highlights during the EPYC launch event so you can catch up there if you wish, as this is not going to be an article that is a regurgitation of specs and marketing lines. This is an editorial."

 

AMD Launches EPYC for the Datacenter
https://videocardz.com/press-release/amd-launches-epyc-for-the-datacenter

AMD EPYC™ Datacenter Processor Launches with Record-Setting Performance, Optimized Platforms, and Global Server Ecosystem Support

Dell, HPE, Lenovo, Mellanox, Samsung Electronics, Supermicro, VMware, Xilinx, and many othersform strong global ecosystem for EPYC™ processors

AMD Launches EPYC for the Datacenter

AUSTIN, Texas, June 20, 2017 (GLOBE NEWSWIRE) — AMD (NASDAQ:AMD), and a global ecosystem of server partners, today marked a new era in the datacenter with the launch of AMD EPYC™ 7000 series high-performance datacenter processors. AMD was joined by multiple customers and partners at the global launch event in presenting a wide array of systems, performance demonstrations, and customer testimonials. The innovative, record-setting AMD EPYC design, with up to 32 high-performance “Zen” cores and an unparalleled feature set, delivers greater performance than the competition across a full range of integer, floating point, memory bandwidth, and I/O benchmarks and workloads.

“With our EPYC family of processors, AMD is delivering industry-leading performance on critical enterprise, cloud, and machine intelligence workloads,” said Lisa Su, president and CEO, AMD. “EPYC processors offer uncompromising performance for single-socket systems while scaling dual-socket server performance to new heights, outperforming the competition at every price point. We are proud to bring choice and innovation back to the datacenter with the strong support of our global ecosystem partners.”

The world’s largest server manufacturers introduced products based on AMD EPYC 7000-series processors at today’s launch, including HPE, Dell, Asus, Gigabyte, Inventec, Lenovo, Sugon, Supermicro, Tyan, and Wistron. Primary hypervisor and server operating system providers Microsoft, Red Hat, and VMware showcased optimized support for EPYC, while key server hardware ecosystem partners Mellanox, Samsung Electronics, and Xilinx were also featured in EPYC-optimized platforms.

Leading Server OEM Platforms

“The EPYC processor represents a paradigm shift in computing and will usher in a new era for the IT ecosystem,” said Antonio Neri, EVP and general manager Enterprise Group, HPE. “Starting with the Cloudline CL3150 and expanding into other product lines later this year, the arrival of EPYC in HPE systems will be welcomed by customers who are eager to deploy the performance and innovation EPYC delivers.”

“As an industry leader, we are committed to driving IT transformation for our customers,” said Ashley Gorakhpurwalla, president, server solutions division at Dell EMC, “Our next generation of PowerEdge servers are the bedrock of the modern data center that are designed to maximize business scalability and intelligent automation with integrated security. The combination of PowerEdge and the AMD EPYC performance and security capabilities will create unique compute solutions for our customers to accelerate workloads and protect their business.”

Cloud Datacenter and Enterprise Customers

Datacenter and cloud service providers also welcomed EPYC to the market today. Members of the “Super 7” datacenter services providers, including Baidu and Microsoft Azure, as well as 1&1, Bloomberg, Dropbox and LexisNexis, all voiced their support at launch.

“As the world’s largest Chinese language search engine and leading AI-Tech company, Baidu prides itself on simplifying a complex world through technology,” said By Dr. Zhang Ya Qin, president of Baidu. “The AMD EPYC processor powered one-socket server can significantly increase our datacenter computing efficiency, reduce TCO and lower energy consumption. We will start deploying with the launch of AMD EPYC and I look forward to our cooperation leading to scaled EPYC adoption this year, and ongoing innovations.”

“We’ve worked to make Microsoft Azure a powerful enterprise grade cloud platform, that helps guide the success of our customers, no matter their size or geography,” said Girish Bablani, corporate vice president, Azure Compute, Microsoft Corp. “To power Azure, we require the most cutting-edge infrastructure and the latest advances in silicon which is why we intend to be the first global cloud provider to deliver AMD EPYC, and its combination of high performance and value, to customers.

Record-Setting EPYC Performance

The excitement around EPYC is driven by multiple record-setting server benchmarks achieved by EPYC-powered one-socket and two-socket systems.

AMD EPYC processors set several performance records, including:

• Two-Socket Server

• AMD EPYC 7601-based system scored 2360 on SPECint®_rate2006, higher than any other two-socket system score1

• One-Socket Server

• AMD EPYC™ 7601-based system scored 1200 on SPECint®_rate2006, higher than any other mainstream one-socket x86-based system score2
• AMD EPYC 7601-based system scored 943 on SPECfp®_rate2006, higher than any other one-socket system score3

All EPYC processors combine innovative security features, enterprise class reliability, and support a full feature-set. An AMD EPYC™ 7601 CPU-based one-socket system shifts expectations for single socket server performance, helping lower total-cost-of-ownership (TCO), providing up to 20% CapEx savings compared to the Intel Xeon E5-2660 v4-based two-socket system4. At every targeted price point for two-socket processors, EPYC outperforms the competition, with up to 70% more performance in the eight hundred dollar price band and up to 47% more performance at the high-end of the market of four thousand dollars or more5.

EPYC Product Overview

• A highly scalable System on Chip (SoC) design ranging from 8-core to 32-core, supporting two high-performance threads per core
• Industry-leading memory bandwidth across the line-up, with 8 channels of memory on every EPYC device. In a two-socket server, support for up to 32 DIMMS of DDR4 on 16 memory channels, delivering up to 4 terabytes of total memory capacity
• Unprecedented support for integrated, high-speed I/O with 128 lanes of PCIe® 3 on every product
• A highly-optimized cache structure for high-performance, energy efficient compute
• AMD Infinity Fabric coherent interconnect linking EPYC CPUs in a two-socket system
• Dedicated security hardware

EPYC Product Lineup

Model Core / Thread Base Freq. Max Boost TDP
EPYC™ 7601 32 / 64 2.2 GHz 3.2 GHz 180W
EPYC™ 7551P 32 / 64 2.0 GHz 3.0 GHz 180W
EPYC™ 7501 32 / 64 2.0 GHz 3.0 GHz 155/170W
EPYC™ 7451 24 / 48 2.3 GHz 3.2 GHz 180W
EPYC™ 7401P 24 / 48 2.0 GHz 3.0 GHz 155/170W
EPYC™ 7351P 16 / 32 2.4 GHz 2.9 GHz 155/170W
EPYC™ 7301 16 / 32 2.2 GHz 2.7 GHz 155/170W
EPYC™ 7281 16 / 32 2.1 GHz 2.7 GHz 155/170W
EPYC™ 7251 8 / 16 2.1 GHz 2.9 GHz 120W

 

So, a person that created a throw away account had this to say about ICC:
"Throwaway account...
I'm a software engineer working on the code optimizer in one of the major C++ compilers out there and I think what AMD did with the results is justified. The Intel compiler is indeed "optimized for SPEC" - it employs optimizations that are either illegal in a language such as C++ or not applicable to pretty much any real program outside the SPEC benchmarks that is larger than a few hundred lines of code. You have to use a magical combination of flags to get anywhere close to the numbers they publish, and if you try that on other programs you either don't see any improvements or might introduce runtime bugs. GCC is overall the most suitable compiler for systems and server software right now.

Every compiler is in a way or another optimized for certain benchmarks because that's usually the code used to test how good the optimizations are, but Intel does seem to go a bit too many steps in this direction. In a way I can understand that they want to make their platform look better by any means..."

 

And just so everyone knows, including @tilleroftheearth , the following shows an Intel 2P was actually tested using the same compiler as AMD:

1. Based on estimated SPECint®_rate_base2006 scores. 2P Intel Xeon E5 scores other than E5-2699Av4 were derived by AMD from the following ICC compiler-based test results published at www.spec.org, multiplied by 0.575 to convert from the ICC compiler to the GCC-02 v6.1 compiler used for EPYC testing: E5-2698v4=1620, E5-2695v4=1440, E5-2680v4=1270, E5-2650v4=1000, E5-2640v4=865, E5-2630v4=814, and E5-2620v4=683. The following EPYC 2P scores are projections made by AMD labs (measured test data for these pre-production products is not currently available): EPYC 7301=845, EPYC 7281=760, and EPYC 7251=485. All other scores based on AMD internal testing. 2P E5-2699A v4 in Intel Server System R1208WT2GSR scored 943 on 6/5/2017, with Ubuntu 16.04, GCC-02 v6.3, 512 GB (16 x 32GB 2Rx4 PC4-2666 running at 2133), 1 x 500GB SSD. 2P EPYC 7601 in Supermicro AS-1123US-TR4 scored 1390, with Ubuntu 16.04, GCC-02 v6.3, 512 GB (16 x 32GB 2Rx4 PC4-2666 running at 2400), 1 x 500GB SSD. The following EPYC 2P scores tested using AMD’s “Ethanol” reference system with Ubuntu 16.04, GCC-02 v6.1, 512 GB (16x32GB 2Rx5 PC4-2667 running at 2400), 1 x 500GB SSD: EPYC 7551=1345, EPYC 7451=1218, EPYC 7401=1120, EPYC 7351=939. Pricing ranges based on Intel recommended customer pricing per ark.intel.com, and AMD 1Ku pricing.
   http://www.amd.com/en-us/press-releases/Pages/amd-epyc-datacenter-2017jun20.aspx
   

 

The overall issue i we need to see how these stack up against the new Sky-lake X Xeons, and then of course pricing. Over time this should start to play out.

Each OEM will chose the benchmark that shows their product in a good light. This is to be expected. Instead of undermining the outcome it should just be explained upfront and a proper suite of benchmarks showing not just good the CPU is but also how off that other benchmark really is and it should not be used. Arbitrary and meaningless statements are useless.

 

My last two posts are to address why they had to do a scaling at all with Intel's scores and that they DID perform a head to head, which showed a massive improvement over Intel. I agree that we need it against SL-SP AND for the tests to be independent, but that should come in short order.

 

These same SPEC / compiler settings disagreements have been going on for decades for most all processors.

They are only for ballpark acceptance of "it's good enough to expend the resources to run a testbed for evaluation".

Then test using your own software builds in an evaluation test bed for acceptance and scaling it against your needs.

.

 

Different Compilers;

 

I'm suprised a name like "EPYC" got past the brass.

 

Here's the video news report, it's the first segment - looking forward to the full review.

HW News: 7900X Runs Hot, Xbox One X, Threadripper Release

 

A New Era for the Datacenter
Published on Jun 20, 2017
Welcome to the start of a new era for the datacenter with the launch of AMD EPYC. Footage taken from the launch event in Austin, Texas on June 20th, 2017.

Learn more at www.amd.com/EPYC

 

http://www.tweaktown.com/news/58113/amd-epyc-32c-64t-flagship-cpu-costs-4200-monster-perf/index.html

Sent from my SM-G900P using Tapatalk

 

Bodes well for ThreadRipper 16c/32t being under $1000, maybe right around the rumored price of $849

 

Now this is more like it, instead of just having a guestimate.

And a video;

 

Benchmarking the Core i9 7900X, i7 7820X, and i7 7800X

 

Love Ryzen is keeping up with the 8-core 7820x in some of those (or beating it).

Sent from my SM-G900P using Tapatalk

 

They should have at least used a 1800x. Wouldn't have been overall a faster CPU but a better showing. Overall though the 1800x is not supposed to beat the Intel offering but be a good overall alternative. This especially considering not only the added cost of the CPU but the x299 board as well.

 

But, even with that, the one video does show it win in a couple cases. But, this guy used 3200MHz ram, likely his computer had the AGESA update and the software updates, etc.

My complaint has been the numbers by many sites are using the 8 core numbers from March, which are not accurate anymore. But, because I hold that vote, that also means Intel desires the same in August/Septembet, after all, every honest reviewer has said this felt rushed, just like Ryzen...

Sent from my SM-G900P using Tapatalk

 

I figures that's the angle they were going for in the first place. Right now enterprise is the best way to break back into the market in a big way and the lower cost will overcome any small difference in overall performance in that sector.

 

Who created this 'throw away' account? AMD, perhaps?

Intel uses the SPEC scores as it's meant to be used for the target market of their customers - AMD is trying to get into that market... uhm... instead of saying the scores are favoring Intel, why don't they just shut up and beat them at their own game? Is it because if AMD used the BM as Intel is using it they will look inferior? Probably.

AMD just tested one system and extrapolated the results to every other system they compared Epyc to. NOT the same thing.

I'm not defending Intel or AMD here; I am defending 'right'.

When I have an interview with a new client I don't dismiss what the previous contractor did - I simply state what I can do for them.

You can spin this anyway you want - in the end; AMD lost all credibility with me.

 

'Every honest reviewer'? Just because you happen to agree with that point of view doesn't make it right.

I guess we'll just ignore the facts ( and the post I made pages back) that SL-X has been in the works since at least 2013...

I'll repeat this again: Processors and Platforms of this calibre do not spring up over night nor are they pulled from someone's ...

 

To show what the current offerings are, not irrelevant numbers, is honest. The fact it changed since initial March tests, giving vast improvements, means if you do not present UPDATED numbers, you are giving FALSE info, without often posting the caveat explaining when it was benched and conditions. If you give all info or a disclaimer, that makes you honest, even if not accurate for current offerings.

You can also say they lost credibility with you all you want. You are an Intel cheerleader. So you discredited your opinion long ago as impartial and neutral. First fix your credibility, then address theirs.

As to Intel, the chipset and support was rushed. MANY REVIEWERS SAID THIS AND SAID IT WAS LIKE WITH RYZEN: ROUGH! So move on IF you will accept no bad information regarding your team--Intel. They moved the release from August back to June because of AMD. That is fine, IF THE PLATFORM IS READY! It wasn't. Accept it!

Sent from my SM-G900P using Tapatalk

 

Whoa, hold on there...

I never said that the Intel launch was smooth. But they are still ahead of AMD even with all the 'rough' in there...

My credibility isn't in question here; AMD's is. I'm not asking for anyone's $$$$, after all. AMD is doing that.

I totally agree with giving the most current information possible when comparing; but that extends further into giving accurate information too and not mangled info that is hidden in the end notes of a 33 page slideshow... Not to mention the testing of a single setup and then extrapolating it to all other Intel platforms.

Intel publishes and shares BM 'scores' that are directed to their customers for the most appropriate use of their hardware... who is AMD to judge that isn't right (because; more than likely, their processors wouldn't look as good if they use the same BM's)?

 

I cleaned this up a bit, we do not need duplication of information in two or more threads. Also pricing specific info on just one platform of CPU release belongs not in a comparison thread unless it is pricing comparisons to the other platform. Even then please post once in the more appropriate thread.

And please all, let's be civil!!!!!

 

sorry i'd have to disagree with you there. by stating this, you are basically acting the same way you claim him to be doing.

hes making a sound argument and making valid points, not saying you're not making your points however its becoming more personal as each statement goes on.

 

The difference is, I still acknowledge Intel's strengths, he marginalizes everything about AMD. That is a significant difference. I also consistently share information on Intel that is positive. He does not do the same for AMD. Do you see a pattern of behavior?

 

It wasn't mentioned whether the Ryzen CPU's were OC'd or stock, but I'd assume stock.

 

I bow down to you and your ninja skills for usually sharing AMD and Intel info (first). Thank you.

But your quick draw does not make me the person you state me to be. Notice that difference?

All the positives of AMD's platform, at this time, can be summed up in one sentence. AMD is the value champion if we only compare processor costs. I've stated this many times before; that is a positive, btw. Not everyone can have the budget, workflow or desire to need best performance (period), but for those that do? Intel is still in the lead as of today.

That 'value' that AMD is offering even helps Intel users. With performance pushed by both companies, prices dropping and options opening up for everyone, the benefits are broad and plentiful.

Clearly, I nor anyone else wishes AMD anything but to continue on this great path they've started.

But you're missing the forest for the trees if you think Intel is a step behind in any aspect that a workstation user cares about. That is just simple truth at this point. Not fanboyism, not marginalizing AMD (without cause) and hopefully gives the readers of this thread a balanced point of view instead of AMD! AMD! AMD! (which isn't bad in itself, btw) when; for the people that want/need to buy a system right now (I suggest they wait too...) may end up making the wrong decision because the assumptions, guesses and actions that AMD may move with it's platform(s) is a lot different than what you are evangelizing.

 

Now, if both were tested at stock, then Intel may pull away (after both are overclocked, if the Intel wasn't overclocked), taking the few wins away from Ryzen. I'll need to review the numbers presented in that video again...

Sent from my SM-G900P using Tapatalk

 

See, Intel fails in one aspect, I/O for graphics acceleration, which IS important to prosumers for workstations. Good misstating facts to prove my point. Thank you! ;-)

Sent from my SM-G900P using Tapatalk

 

You're assuming that when I say 'workstation' it means I need a GPU? Stop assuming and try asking for clarification next time.

I keep (gently) pointing out the fault in your conversational skills; you keep ignoring those points and repeating them.

 

You said it as a blanket statement that Intel wins in every category and you didn't want to mislead people reading. But, with such generalized statement, you then go to a personalized statement on what is needed for you, which is Photoshop related, and I agree, does not apply to your workload. Photoshop still has some of the best scores on the 7700K, while having minor benefit by GPU acceleration up to a 1060 gpu.

But once we go to premiere and davinci, among other photo editing software that benefits from GPU acceleration, as well as certain other modeling that benefits from it, the discussion is more complex. There, you need a fast enough cpu to deal with the multiple GPUs, as well as the ability to scale in number of GPUs. Intel offers the speed and IPC, but not the ability to scale as far as TR on graphics support. So it is a mixed bag there that will depend on Intel having cards, some of which are limited to 8 pcie lanes, which is a hit, but NOT a major one, as compared to the lower clock rate and lower IPC of TR, but with the support of more cards or providing the full 16 lanes.

Sent from my SM-G900P using Tapatalk

Edit: There is also a question of when the programs will support better core scaling, but...

 

Note what you're (continuing...) to do: I have never made a blanket statement that Intel wins in every category. Assume much?

I have also made a hundred posts or more stating 'in my workflow/workloads'. I thought that by now it goes without saying? I know I was tired of repeating myself - I'm sure others were too...

Again; workstation workloads (mine and many others) do not necessarily need a GPU to run optimally. What they need is a CPU to run as fast as they can. Even when a GPU is needed for a certain workstation type workflow; a single GPU is enough and two are more than likely overkill.

Beyond that? You're entering territory that is not defined as workstation class usage. I'm not talking about that; you are.

 

I have a friend that is a videographer that works for professional studios that would beg to differ one GPU is enough and that two is overkill. In fact, most of the people, like Jay Two-Centz, use two cards, with PCworld saying the majority has been moved off of a powerful CPU onto graphics, meaning the scaling of the ability for graphics cards is becoming more and more important as time goes on. Now, I'll admit support beyond one card, even to two cards, is limited and at times even forced, such as for Premiere. But, considering the shift in the industry, we will likely see a shift in this regard in the near future, as software grows and changes to keep up with hardware.

I'll even acknowledge that this may take a year or two to adjust to AMD now offering a significant change in PCIe lanes, which means for many, the benefit may be better to base purchasing decisions on around 2018 or 2019, which will be Intel's new Ice Lake-X platform and Zen 2, one or both of which may or may not support DDR5. Has to be said. But that is not to say it is currently wholly irrelevant, considering some people skip 1-3 generations as there is no real benefit to upgrade on the CPU side, which you have admitted and stated you were doing yourself at times. But, that also means this consideration is relevant if they believe they will need more cards in the future during that period. So...

 

Sounds like for the moment the focus is single, more powerful GPUs, but yeah, I can see the need returning for multi-GPU down the road. But at least one is needed in a lot of cases.

 

Premiere only officially sports single, but says two or more can be used. They just do not guarantee performance. Many, if not most, use 2 titans or 1080 Tis. That is largely because two cards took 32 of the 40 lanes from Intel in workstations. A three way setup reduces two cards to X8, which would limit speeds to a degree, and increase troubleshooting issues involved. Now, Intel is pushing three way sli and crossfire. AMD, though, can offer 3-way at X16 without issue, leaving plenty of room for other peripherals. But, there are some drawbacks. One is whether MB manufacturers will support TB3. AMD already said it isn't on their end to support it. In fact, amd boards incorporate Intel components, like networking, all the time. Intel cannot withhold the license unreasonably. TB3 is important for faster transfers from some external storage devices, which is a boon (they use it because of Mac and Intel support). So, there are other platform and compatibility considerations to factor, which should not be ignored. If you need seamless flow between your hardware, that can effect the decision. Not going to lie. But, to marginalize when that is where the market is headed is to stick one's head in the ground. These professionals want more. Now they have an opportunity to get more. Do you think Intel will let the increased PCIe support go unanswered after people figure out better ways to utilize it and software extends support? I doubt it.

Sent from my SM-G900P using Tapatalk

Edit: You have to remember, if a company says they support SLI or crossfire, they have to provide support for it! I'm sure you can understand that part of it!

Edit 2: "Your GPU is everything! It is more important than your CPU or system RAM (both of which should not be skimped on either).

This should be a dedicated GPU just for image processing in addition to the graphics card running your desktop GUI (user interface). In the case that you are using a laptop or any system with a single, or integrated GPU you can still run Resolve, but performance will be compromised compared to a dual or multi-GPU system."
https://www.dcinema.me/davinci-resolve-system-requirements-a-reality-check/

Edit 3: https://forum.blackmagicdesign.com/viewtopic.php?t=42104
"Curious if anyone is running 3 GPU's for processing in Resolve."
Response:
"Testing here seems to confirm that is true. 4 GPUs in fact."

 

Someone in another comment section on another website posted alleged EPYC SiSoftware Sandra benchmarks.
They were initially low, but subsequently, an update was released for Sandra which supported Ryzen and future hardware, resulting in a really nice uplift... well over double.
http://ranker.sisoftware.net/show_r...d5e3daeed8e0d4f280bd8dabceab96a680f3cefe&l=en

And here's the summary for it:
http://ranker.sisoftware.net/show_s...e5d8edcba39eab8df5c8f9dfbadfe2d2f487ba8a&l=en

The Xeon Platinum 8180 reportedly 'scores' 5000... but Sandra might yet receive more updates for AMD.
Although there's that issue with 8180 costing $12 000, whereas EPYC would cost $4000 (3x price differential).
So, even if EPYC is slower than Platinum (which remains to be seen until both are actually tested and see whether there is a gap and how big)... then an organisation could still get same as or slightly more powerful system for 2 thirds of the price if they go with AMD (at least if you take those Sisoftware benches into account).