Hardware AMD Ryzen 7 Official Reviews and benchmarks

eva2000 · May 10, 2017

Ryzen 5 reviews with updated bios and software revisited Ryzen 5 review: AMD muscles in on Intel’s i5 sweet spot

The number of BIOS revisions, patches, and software updates AMD and its partners have issued since launch means that fair comparisons between Ryzen 5 and Ryzen 7 are impossible without first revisiting older benchmarks. That said, those expecting dramatic uplifts in performance versus launch should temper their expectations: there are improvements in specific games, but nothing as dramatic as frame rates that match Intel.

To ensure an even platform across tests, each Ryzen chip was benchmarked on an Asus ROG Crosshair VI Hero X370 motherboard, with the Ryzen power profile installed. This is a high-end board, but not one that's out of place alongside a chip like the 1600X. BIOS updates have helped with memory compatibility on the Asus board since launch, and it has no trouble running two 8GB sticks of Corsair Vengeance LP DDR4 memory at 2933MHz. Those are paired with a super-fast Samsung 950 Pro NVMe SSD, Corsair 400C case, and an EK-XLC Predator 240 all-in-one liquid cooler.
Click to expand...

eva2000 · May 21, 2017

AMD Ryzen coming to laptop near you Asus ROG teaser video shows the first Ryzen-powered laptop

It’s been a busy couple of weeks for AMD. First, its Ryzen 9 “Threadripper” CPU lineup was leaked, this was followed by the Radeon Vega Frontier Edition unveiling, and a few days ago the company said Ryzen-powered laptops would launch in the third quarter of this year. Now, Asus has teased the first Ryzen notebook on its Republic of Gamers YouTube page.
Click to expand...

Compared to AMD’s 7th generation Bristol Ridge APUs, Ryzen-based laptops boast a 40-percent improvement in GPU performance and a 50-percent increase in CPU performance, all while using just half the power.
Click to expand...

eva2000 · May 26, 2017

AMD Ryzen gets more memory overclocking options Ryzen AGESA 1.0.0.6 exposes more memory overclocking options

inthecloudblog · May 30, 2017

Nocix which is former datashack now offers Ryzen also.
They are not like the perfect provider to do serious stuff whee you rely on your data as many times they do provide smacked drives, at least wholesaleinternet did that until some months ago when I was with them.
WSI for those who don't recall it now operate using all the DC from Nocix but wsi provides network and staff.
Some weeks ago they had some beta scrubbing stuff as well as adding someone like Hibernia to the mix which was singlehomed cogent for v4 IIRC and he for v6.
They did not provide the same stuff as I needed like hetzner does so I left them.

NOCIX - Powerful Servers: Affordably Priced

eva2000 · May 30, 2017

Yeah read of Nocix reputation so wary. I know OVH are evaluating AMD Ryzen for Gaming servers though so looking forward to that

eva2000 · May 31, 2017

AMD Ryzen 7 Laptop ASUS announce their ROG Strix GL702ZC notebook with an 8-core Ryzen 7 CPU | Systems | OC3D News

ASUS announce their ROG Strix GL702ZC notebook with an 8-core Ryzen 7 CPU

At Computex 2017 ASUS has showcased their first Ryzen-based notebook, which surprisingly does not feature a Ryzen mobile APU but a fully fledged Ryzen 7 1700 CPU. This notebook is listed as having the desktop variant of this CPU with 8 cores, 16 threads and a 65W TDP.

On the GPU side, this notebook is also listed as having a 4GB 65W RX 580 GPU, which is a TDP that is exceptionally low for an RX 580, suggesting that this is a power optimised model with lower than average clock speeds for mobile workloads.
Click to expand...

eva2000 · Jun 4, 2017

Seems alot of AMD Ryzen related news passed me by recently

AMD Begins Cutting Prices On Ryzen CPUs

Some Ryzen Linux Users Are Facing Issues With Heavy Compilation Loads

AMD Releases Optimizing C/C++ Compiler For Ryzen (so in future for Centmin Mod looking at GCC, Clang and AMD AOCC compiler support )

AMD Ryzen AOCC 1.0 Compiler Tuning Benchmarks

Core i7 7700K vs. Ryzen 7 1800X With Ubuntu 17.04 + Linux 4.12

eva2000 · Jun 25, 2017

AMD Ryzen 7 vs Intel Kaby Lake-X Linux 4.12 Ubuntu Benchmarks With AMD Ryzen, Intel Kabylake - 12 Systems - Phoronix

For your viewing pleasure today before getting to the Core i7 7740X / Core i7 7900X results by earlier next week, the fresh Linux benchmark numbers for sharing today are of the:

1: AMD FX-8350
2: AMD Ryzen 7 1700
3: AMD Ryzen 7 1800X
4: Intel Core i7 4790K
5: Intel Core i7 5775C
6: Intel Core i7 5960X
7: Intel Core i7 6800K
8: Intel Core i5 7600K
9: Intel Core i7 7700K
10: Intel Xeon E3-1280 v5
11: Intel Xeon E5-1680 v3
12: Intel Xeon E5-2687W v3

For a nice spectrum of hardware on the Linux 4.12 Git kernel as of this week and Ubuntu 17.04. Each system was running with the memory at its maximum rated frequencies, all were using Ubuntu 17.04 with the GCC 6.3.0 compiler and manually switching to the Linux 4.12 development kernel.
Click to expand...

eva2000 · Jul 19, 2017

ouch fake Ryzen cpus Buyer Beware! Reports Of Fake Ryzen CPUs Popping Up On Retail Sites !

There are a number of reports of fake Ryzen processors being sold by third party sellers on retail sites such as Amazon and eBay.

Although this is the first time we have seen fake Ryzen processor for sale, the scam isn’t new. In the past, we have seen counterfeit Core i7 processors sold by reputable retailers, as well as fake AMD A-Series A8-7600 processors making the rounds on Amazon.
Click to expand...

The scammer buys a handful of legitimate high-dollar AMD Ryzen 7 1800X processors, scans the CPU heatspreader, and then prints a sticker and applies it to the face of the fake CPU (in this case an old Celeron processor). The scammer returns the counterfeit processor to the box, reseals it, and then sells it on eBay / Amazon. This type of scam works because the counterfeit processor looks authentic from the outside of the package.
Click to expand...

and Report: AMD Ryzen Series Customers Falling Prey To RMA Fraud On Amazon

What seems like a very well organized RMA fraud is seeping through the cracks over at Amazon. In a very short period of time, two users have so far reported receiving a fake Ryzen processor and both have indicators of being orchestrated by the same person(s). While Amazon did offer both of them refunds and even a gift card, these type of scams might turn away first timers who want to try the PC-building experience and even lead to damaged motherboards in some cases (which Amazon will probably not cover).
Click to expand...

inthecloudblog · Jul 20, 2017

I was aware because of a Telegram group but this helps spread the word

EDIT:

AMD's Ryzen processors are the hot new chips on the market, and I don't mean in terms of the heat they produce. AMD is back to offering competitive processors that undercut Intel on price. And because they are proving so popular, fake Ryzen chips are appearing for sale.
Click to expand...

Fake AMD Ryzen Chips Appear for Sale on Amazon/eBay

eva2000 · Jul 20, 2017

sneaky bastards even resealing the retail box and asking for full refunds from retailer !

eva2000 · Jul 24, 2017

More interesting compiler benchmarks on AMD Ryzen for Clang 4 vs Clang 5 vs GCC 6 vs GCC 7 vs GCC 8

Ryzen Compiler Performance: Clang 4/5 vs. GCC 6/7/8 Benchmarks - Phoronix

Centmin Mod is working on Clang 4 and 5 support Install - Compiling Newer Versions of Clang 4.0. Already has GCC 6 support but waiting on GCC 7 and well haven't even looked at GCC 8 yet

eva2000 · Jul 24, 2017

hmm gcc segmentation faults on Ryzen / Linux | Community

1. In some systems, there is instability generally when the system is idle which tends to manifest as Machine Check Exceptions, random freezes/system lock up, and random reboots. Keeping the system 'active' prevents this from occurring. This happens more frequently in Linux, but also happened to me under Windows until I switched to the Ryzen balanced power plan in Windows. This particular bug seems to be very much tied to the motherboard/RAM combination and based on everything I have read it seems to be a result of voltage instability leading to RAM errors. Changing from an MSI x370 Gaming Pro Carbon to an Asrock Taichi x370 MB completely eliminated this problem for me, but while on the MSI board, I would only experience this particular issue when the RAM was at stock JEDEC settings (1.2V, 2133MHz). Running the RAM at XMP profile settings (1.35V, 2933MHz) resolved it for me.

Fixes that have worked for others include:

Using more aggressive Load Line Calibration settings for the SOC and/or VCore.

Setting the SOC voltage to a fixed voltage. 1.1V is a safe setting and generally the default voltage the processor uses. This may be sufficient. Others have gone up to 1.15 or 1.18V successfully. I can't speak for whether those are safe for long-term use.

Along with the SOC voltage, setting the VCore voltage to a fixed setting has helped some people. AMD's tech support recommended to me that I try settings from 1.3625V up to 1.4250V (with SOC voltage set at 1.1V) so I would assume those voltages are considered OK by AMD (assuming heat isn't an issue).

Changing the CPU governor to performance (this will basically keep your processor always at the highest frequency p state).

What doesn't appear make a difference for this issue:

Disabling OpCache, SMT, or ASLR

RAM timings (assuming they are not overly aggressive to begin with)

2. There is a problem that arises in highly multi-threaded workloads under very specific scenarios. This particular problem manifests as segfaults or trapped illegal instruction. Typically this is encountered when compiling software packages in Linux using gcc and happens more frequently when using large amounts of RAM and many concurrent processes (e.g. make -j16). It should be noted that stress tests like prime95, or stress-ng are stable and do not crash, although they may exacerbate the problem if run concurrently with the compilation. For me, using the "ryzen_segv_test" compiled with GCC 6.3 is the most effective way to trigger the segfaults which happen with minutes (up to 15 max. in my case). This is very specific though, as running the same code compiled with GCC 6.4 only crashed once in 8 hours, and never crashed in over 12 hours of tests when the code was compiled with Clang 4.0. This particular issue seems related to both multithreading and the opcache, and seems to be also dependent on the particular location in memory where the code is running. Disabling ASLR will significantly reduce the frequency of the segfaults (but not eliminate them in my case). Disabling SMT will completely fix this problem for me. Others have reported that disabling OpCache completely eliminated the problem for them.

What others have reported as helping:

A new processor (e.g received through RMA). Unfortunately this didn't help for me.

Using kernel version 4.11.X, with the kernel timer frequency at 1000Hz (low-latency kernel). This reduced the frequency of segfaults for me, but did not eliminate them.

Better cooling (though I'm not convinced this was the same problem)

Using higher Vcore and SOC voltages (but this did nothing for me, and I think it's conflating problem #1 with this one).

Using Clang rather than GCC

Some Gentoo users have noted improvements based on the compiler optimizations used when compiling their system and/or versions of binutils. (No particular combination ever resulted in stability for me)

As an aside, I would really like to find a way to reproduce issue #2 in Windows if I can - if only to rule out a Linux kernel issue (although it also happens in BSD...). The problem has been demonstrated to happen in Windows Subsystem for Linux, and I have also confirmed that it occurs in a Linux virtual machine running on a Windows host, but I have not been able to get a Windows 10 system to exhibit the same issue. The ryzen_segv_test compiled with Visual Studio 2017 ran for >24 hours without error for me.
Click to expand...

inthecloudblog · Jul 24, 2017

love how the 1700x is going down
EDIT:
I mean in price

eva2000 · Jul 24, 2017

indeed, though from previous post - choosing the right ryzen motherboard pairing with cpu is important too

eva2000 · Aug 5, 2017

hmmm Ryzen-Test & Stress-Run Make It Easy To Cause Segmentation Faults On Zen CPUs - Phoronix

With running a number of new Ryzen Linux tests lately, a number of readers requested I take a fresh look at the reported Ryzen segmentation fault issues / bugs affecting a number of many Linux users. I did and still am able to reproduce the problem.

For those that missed our earlier article on the matter from early June, heavy workloads can cause problems on Ryzen, in particular segmentation faults while there have also been reports of some stability problems.

This Google Doc remains among the resources trying to track this issue on Linux while on the Gentoo Forums, AMD Forums, and elsewhere are more reports of various problems encountered under extreme workloads -- like a ton of code compiling for hours on end, but can also happen in other scenarios.

AMD hasn't publicly commented on the problem and as of Linux 4.13 the issue is still happening. If carrying out the same tests on Intel CPUs, the segmentation faults do not occur. There is even ryzen-testto easily try reproducing the issue. The ryzen-test script will build GCC in parallel loops from a compressed ramdisk, in order to easily stress the CPU. In my day-to-day benchmarking of Ryzen CPUs, however, I haven't hit this problem or even on my main production desktop with using Ryzen 5. The problem really comes to light just under very heavy and continuous workloads it seems.
Click to expand...

eva2000 · Aug 7, 2017

ouch 50+ Segmentation Faults Per Hour: Continuing To Stress Ryzen - Phoronix

In direct continuation of yesterday's article about easily causing segmentation faults on AMD Zen CPUs, I have carried out another battery of tests for 24 hours and have more information to report today on the ability to trivially cause segmentation faults and in some cases system lock-ups with Ryzen CPUs.
Click to expand...

haven't read entire thread 50+ Segmentation Faults Per Hour: Continuing To Stress Ryzen - Phoronix Forums

Unfortunately, kill-ryzen.sh still segfaults under Antergos and gcc 7.1.1 toolchain. It takes much longer than Ubuntu, always around 10.500s. I don't understand this. But is is very consistent. I am trying to compile a kernel with 1000 hz now.

Good news is that in AMD support forum one of the guys just reported that he got a good processor through RMA. We already had reports of people getting bad processors after RMA, but now we have a case of good RMA. This may indicate that maybe that are Ryzen batches with problems, or that newer batches are better after tweaks in the production line. This new, good, processor was fabricated in June. Brutalix when did you buy your processor? Is is recent?

So there seems to be good and bad processors in wild. Which is a good sign. It may indicate that RMAs for the affected may fix the problem. Most people won't require a RMA, for example if they use Windows or even Linux but do not compile a lot. I am in the process of RMA of my processor. My fingers are crossed.
Click to expand...

eva2000 · Aug 8, 2017

AMD Confirms Linux Performance Marginality Problem Affecting Some, Doesn't Affect Epyc / TR - Phoronix

This morning I was on a call with AMD and they are now able to confirm they have reproduced the Ryzen "segmentation fault issue" and are working with affected customers.

AMD engineers found the problem to be very complex and characterize it as a performance marginality problem exclusive to certain workloads on Linux. The problem may also affect other Unix-like operating systems such as FreeBSD, but testing is ongoing for this complex problem and is not related to the recently talked about FreeBSD guard page issue attributed to Ryzen. AMD's testing of this issue under Windows hasn't uncovered problematic behavior.

With the Ryzen segmentation faults on Linux they are found to occur with many, parallel compilation workloads in particular -- certainly not the workloads most Linux users will be firing off on a frequent basis unless intentionally running scripts like ryzen-test/kill-ryzen.
Click to expand...

AMD was also able to confirm this issue is not present with AMD Epyc or AMD ThreadRipper processors, but isolated to these early Ryzen processors under Linux. We will also now be receiving Threadripper and Epyc hardware for testing to confirm their Linux state. Their analysis has also found that these Ryzen segmentation faults aren't isolated to a particular motherboard vendor or the like, contrary to rumors/noise online due to the complexity of the problem.
Click to expand...

eva2000 · Aug 25, 2017

Good news for new stock of AMD Ryzens New Ryzen Is Running Solid Under Linux, No Compiler Segmentation Fault Issue - Phoronix

As a quick update to the AMD Linux "Performance Marginality Problem" affecting some early Ryzen processors under heavy load, today I received a new Ryzen 7 processor and indeed it's been running strong now for the past few hours under demanding load and has yet to hit the compiler segmentation fault bug.
Click to expand...

AMD has not provided an official public explanation of the fundamental problem, but from those in our forums and elsewhere, it appears to affect Ryzen CPUs manufactured prior to week 25. This Ryzen 7 1800X was from "week 30" where as my original Ryzen 7 1800X review sample was from "week 5" as well as the Ryzen 7 1700. The Ryzen 5 1400 that I bought via retail channels in June was manufactured in week 14. The Ryzen 3 1300X review sample was from week 20 and the Ryzen 3 1200 from week 25
Click to expand...

eva2000 · Aug 29, 2017

Looks like Hetzner's AMD Ryzen cpus are older batches with issues [Info] Hetzner AX60-SSD currently uses Ryzen's older than CW25..30 | Web Hosting Talk

Just tried out two of Hetzner's AX60-SSD and found out that both are suffering from the segmentation fault issue.
When building a Yocto based Linux distribution, segmentation faults occurred after approx. 10mins.
The widely spread ./kill-ryzen.sh test lead to failures on both machines (after ~150s on the first and ~30s on the replacement machine which I received from the kind support guy of Hetzner).
If anyone is considering a Ryzen based Hetzner Server to reliably perform heavy compilation tasks using Linux, I strongly advise to hold back for now.

Hetzner confirmed that any Ryzen CPU's they currently use (as of 28. August) are older than CW25..30 and therefore might suffer from the segmentation fault issue.
PS: I do not want in any way to blame Hetzner for this as this is an error on AMD's side.
Click to expand...

Log in / Register

Forums

Discover Centmin Mod today

Hardware AMD Ryzen 7 Official Reviews and benchmarks

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

inthecloudblog Active Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

inthecloudblog Active Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

inthecloudblog Active Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

.

Log in / Register

Useful Searches

Discover Centmin Mod today

Hardware AMD Ryzen 7 Official Reviews and benchmarks

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

inthecloudblog Active Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

inthecloudblog Active Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

inthecloudblog Active Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

eva2000 Administrator Staff Member

.