Anybody planning a Ryzen build?


Recommended Posts

I did a bit more testing.  Rebooting to complete the 6.5.2 installation, I double-checked my BIOS to make sure all the settings appeared correct.  While everything looks fine, I don't see an XFR setting at all in my BIOS.  I think it is safe to assume it is enabled.

 

In Tips & Tweaks, I found that if I change "Enable Intel Turbo Boost?" to Yes (I had it on No before), then my max core frequency changes to 3.7GHz, so that's a small improvement.

 

I tried changing the "CPU Scaling Governor", and on Performance all cores basically idled at 3.7GHz.  On Demand allows other cores to scale down, normally hitting around 2.1GHz, and I can see my 1 or 2 cores hitting 3.7GHz, but no higher.  Power Saver basically forces all cores down to about 2.2GHz.

 

The odd part here is that the motherboard/cpu easily hits an 8/16 core frequency of 3.7GHz, which is perfect, but refuses to boost a single core any higher than 3.7GHz, even if only a single core is loaded.

 

3 minutes ago, david279 said:

This seems to be a ryzen 1 issue so I just oc, glad it's fixed for ryzen 2ND gen...

 

I agree that I'm glad it is fixed, I'm just baffled that I'm not finding any additional info on the web regarding the problem or the solution.

 

Paul

Link to comment
1 hour ago, david279 said:

Funny thing the all core boost for the 1800x is 3.7 but it's the xfr that doesn't seem to translate over.

 

Not sure I agree with your interpretation.

 

The 1800x  base clock is is 3.6GHz, and the single core boost is 4.0GHz.  XFR adds 0.1GHz to both of those numbers.  I'm getting 3.7Ghz all core, so I think XFR is working.

 

To me, it seems the boost isn't working.  Core Performance Boost (CPB) allows individual cores to boost up to 4GHz (4.1GHz with XFR) on the 1800x.  This works on Windows.

 

CPB support for early/first release 1800X's (which is what I have) was added to Linux kernel 4.14 (it was broken prior to that), and I've verified that the CPB flag is showing.

 

I just did some more testing.  Upgraded my BIOS to the latest, (which includes the new AGESA 1.0.0.1a) but that made no difference.  Also tried disabling SVM in the BIOS, as I read somewhere that solved another user's problem with no single-core boost in Linux.  Disabling SVM broke KVM (which I definitely need) and did not fix the boost issue anyway.

 

My temps are fine, and my cooling is sufficient.  I could easily overclock all cores to 4.0GHz, but I choose not to.  Stability, longevity, and power conservation are more important to me.  All the same, this CPU should boost a single core up to 4.1GHz, and it just won't do it under Linux.

 

 

react, since you have a 2700x, what do you get for this command?

 

root@Tower:~# cpufreq-info
cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009
Report errors and bugs to cpufreq@vger.kernel.org, please.
analyzing CPU 0:
  driver: acpi-cpufreq
  CPUs which run at the same hardware frequency: 0
  CPUs which need to have their frequency coordinated by software: 0
  maximum transition latency: 4294.55 ms.
  hardware limits: 2.20 GHz - 3.60 GHz
  available frequency steps: 3.60 GHz, 3.20 GHz, 2.20 GHz
  available cpufreq governors: conservative, userspace, powersave, ondemand, performance, schedutil
  current policy: frequency should be within 2.20 GHz and 3.60 GHz.
                  The governor "ondemand" may decide which speed to use
                  within this range.
  current CPU frequency is 2.20 GHz (asserted by call to hardware).
  cpufreq stats: 3.60 GHz:6.43%, 3.20 GHz:4.04%, 2.20 GHz:89.53%  (6269)

 

 

Paul

Link to comment

Hello all,

 

Paul running the cpufreq-info i get the following (on all cores):

 

image.png.9bdc971dc4602a7b93d3dd13975a1164.png

 

Ill try to dig into this a bit more over the weekend (post some bios screenshoots), but basically its all defaulted, yes SVN is on (or no vm), default was inactive.

I believe (not sure) the boost is all done in firmware nothing to do with the OS, at least on Ryzen 2. It may well be the XFR2 because its more linear (more resolution/interpolation) it scales better on a linux OS. Windows may push XFR1 to max but not linux (just wondering) ...

 

And im using CPU Scaling Governor: Conservative on the unraid. Changed it to performance but did not notice any change (but cannot guarantee as i was just trying to spot changes using the grep command) ... 

 

Regards

 

 

Edited by react
Link to comment

Very interesting!  Thanks react!

 

I might be reading between the lines, but it looks like you don't have a software cpu frequency driver running at all, so it is defaulting to hardware frequency control. 

 

I have the acpi-cpufreq driver controlling the frequencies, which overrides the hardware control.  Perhaps the problem is with the acpi-cpufreq driver.

 

My guess is that you didn't see any change when switching from Conservative to Performance because you don't have a driver, so you aren't really controlling anything.

 

Anyone know how I can unload the acpi-cpufreq driver?

 

I also recall some settings in the BIOS that I think were related to allowing the operating system to control frequency - though I might be imagining that now.  I'll have to look again.  Maybe I can disable software/OS control.

 

Paul

Link to comment
42 minutes ago, Pauven said:

Very interesting!  Thanks react!

 

I might be reading between the lines, but it looks like you don't have a software cpu frequency driver running at all, so it is defaulting to hardware frequency control. 

 

I have the acpi-cpufreq driver controlling the frequencies, which overrides the hardware control.  Perhaps the problem is with the acpi-cpufreq driver.

 

My guess is that you didn't see any change when switching from Conservative to Performance because you don't have a driver, so you aren't really controlling anything.

 

Anyone know how I can unload the acpi-cpufreq driver?

 

I also recall some settings in the BIOS that I think were related to allowing the operating system to control frequency - though I might be imagining that now.  I'll have to look again.  Maybe I can disable software/OS control.

 

Paul

 

 

rmmod -f acpi-cpufreq to unload

 

modprobe acpi-cpufreq to load the driver

Just found that while googling...

Edited by david279
Link to comment
1 hour ago, david279 said:

rmmod -f acpi-cpufreq to unload

 

That worked, thanks!  Now I get this:

root@Tower:~# cpufreq-info
cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009
Report errors and bugs to [email protected], please.
analyzing CPU 0:
  no or unknown cpufreq driver is active on this CPU
  maximum transition latency: 4294.55 ms.

for all 16 CPU's.

 

I also see "Driver: * no driver *" for CPU Frequency Scaling in Tips & Tweaks, and changing from On Demand to Performance no longer has any impact, where before it would push all cores to 3.7GHz.  So the behavior is now very similar to react's system.

 

But I'm not seeing anything above 3.7GHz.  Top speeds looks the same.

 

I wonder if it has to boot that way.  Perhaps stripping the driver out of a running system doesn't have the same result as booting without a driver.  And I still need to look in my BIOS again to see if there is a setting related to giving the OS control over frequency scaling.

 

Paul

Link to comment
On 5/17/2018 at 4:48 PM, Pauven said:

 

I can't believe that there isn't more discussion on this, as it certainly caught my attention.

 

I have an 1800X, and my individual core's max out at 3.6GHz with unRAID, even when only a single core is loaded.

 

I always figured that this was just a limitation of Linux, as in earlier testing I saw 4.1GHz in Windows.

 

Now react comes along and shows boosting working perfectly on a 2700X, what's up with that?

 

While I know that the 2xxx series on the X470 motherboards adds Precision Boost 2 and XFR2, these simply permit more aggressive boosting when more cores are loaded, and shouldn't be required for boosting when only a single core was loaded.

  

I never see over 3.6GHz (I don't manually overclock).  Is this the same experience of other Ryzen 1 users?

 

Is there a change in Ryzen 2 or X470 that fixes this issue, or a system setting or BIOS setting I might have wrong?

 

Paul

did the same test on r1700 and Asus Crosshair Hero x370

image.png.0ccf5ce22dc11876e6d22a197691b45c.png

Link to comment

Hello all,

 

Relative to the subject of Ryzen 2 oc vs Ryzen 1 oc i believe the next graph shows the reason:

 

image.png.87568f8e5e66cf91b2975700b5b9a0e7.png

 

Ryzen 1 users are reporting on unraid only getting 3.7Ghz max, that kind of make sense since unraid (when loaded with other stuff, mainly vm) will always use more than 2 threads. One way you guys can test this is to unload everything from unraid, and do cpu stress test on unraid itself and check if 1 or 2T can boost to 4GHz (if unraid does not spread the load anyway). This would have to be done as Paul proceeded in removing the Linux CPU Freq scaling driver (as i believe is useless with Ryzen). Also we would need a better way to benchmark instead of getting only instantaneous GHz reading with grep it would be great to get session min/max/avg readings, dont know if that command exists, but it can be created with the grep command in a script file running @ lets say 10ms or even less if possible (it would be handy). If this is the case then its actually a good sign Ryzen 1 users cannot reach 4GHz, meaning unraid is spreading the load evenly across multiple cores, opposite to windows ...

 

The data that i get in Ryzen 2700x both on unraid and on bare-bone windows 10 matches in perfection the previous graph and i can even see on windows (using hardware monitor) a session max close to 4.35GHz in some cores with the stock cooler. On unraid max that i could spot using grep was around 4.25GHz.

 

Regards.

 

 

 

 

Edited by react
Link to comment
36 minutes ago, react said:

One way you guys can test this is to unload everything from unraid, and do cpu stress test on unraid itself and check if 1 or 2T can boost to 4GHz (if unraid does not spread the load anyway). This would have to be done as Paul proceeded in removing the Linux CPU Freq scaling driver (as i believe is useless with Ryzen). Also we would need a better way to benchmark instead of getting only instantaneous GHz reading with grep it would be great to get session min/max/avg readings, dont know if that command exists, but it can be created with the grep command in a script file running @ lets say 10ms or even less if possible (it would be handy). If this is the case then its actually a good sign Ryzen 1 users cannot reach 4GHz, meaning unraid is spreading the load evenly across multiple cores, opposite to windows ...

 

The data that i get in Ryzen 2700x both on unraid and on bare-bone windows 10 matches in perfection the previous graph and i can even see on windows (using hardware monitor) a session max close to 4.35GHz in some cores with the stock cooler. On unraid max that i could spot using grep was around 4.25GHz.

 

I feel that I have essentially already done this test.  With everything unloaded and nothing running, unRAID reports that all 16 cores are idle at 0%.  I can then run a command to load a single core, and it never exceeds 3.7GHz.  I have also repeatedly seen 14 cores idling at 2.2GHz, and the other 2 at 3.7GHz, which seems to corroborate that I am loading only a single core.

 

I have also used other commands to test frequency, besides grepping /proc/cpuinfo.  So far, the other commands essentially matched the /proc/cpuinfo, and if anything they were slightly lower.  Sorry, I don't have all the commands in front of me, I found them during web searches (supposedly they were more accurate for Ryzen), gave them a try, and moved on.

 

I certainly agree that, based upon that graphic, the Ryzen 2xxx series will hit higher frequencies with more cores active, which is really awesome.  But that shouldn't change the fact that with only a single core loaded, a Ryzen 1800X should hit 4.0 GHz minimum, and 4.1 GHz with XFR.  Mine never goes over 3.7GHz, which is the same frequency that I can easily hit with all cores using the Performance governor.

 

Perhaps I am mistaken, and that even when it appears that only a single core is loaded, in reality more cores are active.  In which case, I agree that the graphic perfectly explains what is going on.

 

Paul

Link to comment

Hello Paul,

 

Yes i understand, is it possible for you to disable all but one core on the bios, and give it a run with unraid. That test would definitely put this doubt of hardware/software to an rest.

 

Ideally keep active your best core.

 

Also here is a command to watch the frequencies update every second:

watch -n1 "cat /proc/cpuinfo | grep MHz"

 

With the previous command i can see that even with a very small load 2 to 5% my cpu speeds on all cores over around 4GHz, dropping sometimes to 2ishGHz, that points that at minimum effort unraid does in fact spreads the load.

 

Regards

Edited by react
Link to comment

Hello all, 

 

Just another quick update, if i disable all the load from my last 8 vcores, then i see the cpu frequency going down and staying down around 2ishGHz (no cpu scaling driver). Pointing to the linux scaling driver is artificiality slowing down/up the core frequencies (as Ryzen 1 users reported), overriding the Ryzen PBoost curve.

 

image.png.ebaee7440261b616aa545006661d9cbc.png

 

Regards

 

Edited by react
Link to comment

Interesting results.

 

After loading the linux scaling driver, does the behavior revert to the Ryzen PBoost curve when you unload the driver, or do you have to reboot.

 

I don't see where my driver is getting loaded, which makes me think it is automatic by the kernel.  Not sure how I would block it during the boot.

 

I just read that Ryzen microcode updates were finally added to the linux-firmware.git collection.  This makes me think my Ryzen 1800X has never received any microcode firmware updates.  Not sure if that would affect the boost or not, but regardless I posted a request for including the new microcode updates here:  https://lime-technology.com/bug-reports/stable-releases/652-amd-cpu-microcode-updates-r84/

 

Link to comment
1 hour ago, david279 said:

You could run the command at the start of the array using the scripts plugin or add it to your go file.

 

At that point, it has already been loaded, and is no different than me unloading it from the command line.

 

I want a way to prevent it from ever loading in the first place.

Link to comment

So I was originally looking into making my first unRAID server a dual Xeon E5-2650 build, which seemed fairly popular. However, the rise in prices on that old server equipment has made it less of enticing option. I'm hoping for something in the 13k+ passmark range. Do you all think a Ryzen 2600 would be a decent CPU to base a build on? My plan is to use it for the following:

  • Plex Media server - something that can handle four 1080p transcodes, and one 4k local play - all simultaneous. (Note: the one 4k local play is to an Nvidia Shield, so I don't think that will use up much CPU in the server, right?)

  • Support 6+ HDDs, and 2 SSDs.

  • Run various media grabbing services (Radarr, Sonarr, NZBGet, Deluge, VPN), run a web Server (NGINX). I would assume 12+ docker containers?

  • General file server (shared folders accessible to home PCs, and act as a general backup to data on the home PCs)

  • I'd like it to run 1-2 VMs in case I want to tinker on various things

 

My main worry is DDR4 RAM prices. I assume I want to go for ECC RAM, right? 16GB would be minimum, but 32GB would be preferred. That's probably the component that will make the Ryzen build very costly. Thoughts?

Link to comment
On 5/20/2018 at 1:46 PM, Pauven said:

Interesting results.

 

After loading the linux scaling driver, does the behavior revert to the Ryzen PBoost curve when you unload the driver, or do you have to reboot.

 

I don't see where my driver is getting loaded, which makes me think it is automatic by the kernel.  Not sure how I would block it during the boot.

 

I just read that Ryzen microcode updates were finally added to the linux-firmware.git collection.  This makes me think my Ryzen 1800X has never received any microcode firmware updates.  Not sure if that would affect the boost or not, but regardless I posted a request for including the new microcode updates here:  https://lime-technology.com/bug-reports/stable-releases/652-amd-cpu-microcode-updates-r84/

 

 

Hello,

 

Never had a cpu scaling driver, dont even know how to do it.

 

Regards. 

Link to comment
13 hours ago, Magic815 said:

So I was originally looking into making my first unRAID server a dual Xeon E5-2650 build, which seemed fairly popular. However, the rise in prices on that old server equipment has made it less of enticing option. I'm hoping for something in the 13k+ passmark range. Do you all think a Ryzen 2600 would be a decent CPU to base a build on? My plan is to use it for the following:

  • Plex Media server - something that can handle four 1080p transcodes, and one 4k local play - all simultaneous. (Note: the one 4k local play is to an Nvidia Shield, so I don't think that will use up much CPU in the server, right?)

  • Support 6+ HDDs, and 2 SSDs.

  • Run various media grabbing services (Radarr, Sonarr, NZBGet, Deluge, VPN), run a web Server (NGINX). I would assume 12+ docker containers?

  • General file server (shared folders accessible to home PCs, and act as a general backup to data on the home PCs)

  • I'd like it to run 1-2 VMs in case I want to tinker on various things

 

My main worry is DDR4 RAM prices. I assume I want to go for ECC RAM, right? 16GB would be minimum, but 32GB would be preferred. That's probably the component that will make the Ryzen build very costly. Thoughts?

 

I cannot help you with the load requirements, relative to the mem i dont think ECC would be reasonable for a Ryzen 2. To take full advantage of all those cores you need fast mem (>2.6GHz). As far im aware fast ECC DDR4 does not exist, if it does, it will be bloody expensive !

 

Edited by react
  • Like 1
Link to comment
On 5/16/2018 at 3:00 PM, react said:

The system is currently up and running for more than 3 days, no issues.

 

Hi I am ready to Upgrade my old AMD Phenom 4C CPU with a new one and new MB!!  I just want to know if with the Ryzen 2 Family including the Ryzen 5 2600X the issues with Freezes and disabling C Sate is resolve?  I just want a rock solid 24/7 system with low power consumption without any freezes and be able to handle a few VM's a Plex Media Server with a few Dockers.  

Link to comment
22 hours ago, afc said:

 

Hi I am ready to Upgrade my old AMD Phenom 4C CPU with a new one and new MB!!  I just want to know if with the Ryzen 2 Family including the Ryzen 5 2600X the issues with Freezes and disabling C Sate is resolve?  I just want a rock solid 24/7 system with low power consumption without any freezes and be able to handle a few VM's a Plex Media Server with a few Dockers.  

 

The Ryzen 2xxx range does now seem to be stable with the introduction of 'Power Supply Idle' motherboard BIOS option. Changing that from the default of 'auto' to 'Typical' does seem to fix the lockup problem, without the need to modify C-states. To help you decide whether that makes things stable enough for you I would recommend the following kernel thread. Lots of good information earlier, but I've bookmarked comment 318 as a reasonable starting point for browsing the thread.

 

https://bugzilla.kernel.org/show_bug.cgi?id=196683#c318

 

As much as I'm enjoying my new (and so far, stable) MSI X470 mobo/Ryzen 2700x combo, if stability is absolutely paramount to you then you might want to consider Intel. It does seem like Ryzen is over the hump, but I'm not sure if it is yet at the 99.999+% reliability that we would hope for.

 

Link to comment

I’m also planning a Ryzen build. Want to get a 1600(X) and I need about 6 SATA Ports (at least if it could a problem to use an additional SATA PCIE Card?). My biggest concern is IOMMU, because I want to run a Mac OS VM on it. First I thought it should be capable of gaming as well, but it would be enough if I can use the Mac OS VM for daily Office work and some photoshop from time to time, so nothing fancy...

 

Can someone recommend a board, so I can separate the GPU as well as a PCIE WiFi card and at least one USB port?

Link to comment
16 minutes ago, lixe said:

Can someone recommend a board, so I can separate the GPU as well as a PCIE WiFi card and at least one USB port?

 

I can't recommend a specific board, but I can perhaps narrow down your choices. I went for the MSI Gaming Pro Carbon AC (AC being the model that has integrated wifi/bluetooth). That's a nice X470 board, with eight SATA ports, but the AC model's integrated wifi is stuck in the same IOMMU group as the network card (with or without ACS override). One to drop from your list (though the non-AC model with an add-on wifi card would be viable, and just as nice a board).

 

I'm assuming that you don't want to run your Mac OS VM and Windows Gaming VM simultaneously, in which case you won't need the multi-gpu support that the X370/X470 motherboards provide, and can save some money (and reduce the list of boards that you need to consider) by focusing on the B350 (single GPU) range. The new B450 range is due for release any day now, but if the X370/X470 pattern is anything to go by then you won't miss out on anything significant by going for the B350 and will save some money.

 

Good hunting

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.