AMD CPU and Platfrom Discussion Episode IV: A New Hope is Ryzen

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > AMD CPU and Platfrom Discussion Episode IV: A New Hope is Ryzen

«‹›2 »

SamDabbers: May 26, 2003

Ryzen owns for multitasking :awesome:

Two separate x264 encodes (nice'd), each pinned to all eight threads of its own CCX, browser, irc, email, torrents, and a steam game running simultaneously.
Everything is buttery smooth.

Only registered members can see post attachments!

SamDabbers fucked around with this message at 00:08 on May 10, 2017

# ¿ May 10, 2017 00:04

Adbot: ADBOT LOVES YOU

# ¿ Apr 30, 2024 04:16

SamDabbers: May 26, 2003

I recently upgraded from an i7-4770 workstation to a Ryzen 1700. At the same 3.6GHz all-core clocks and RAM running at DDR4-3200, the Ryzen is +/- twice the throughput for x264 encoding tasks at about the same cost as I paid for the i7 setup in 2014. It's pretty much two Haswell/Broadwell i7s in a single socket, and it rules for both heavily multithreaded throughput tasks and general multitasking.

The HEDT platfroms look less appealing now that we're not limited to 4 cores on regular-rear end desktops.

# ¿ May 17, 2017 18:20

SamDabbers: May 26, 2003

NewFatMike posted:

What mobo/RAM kit are you using? I haven't had to pay much attention to RAM for builds before and I want to make sure I've got my eye on the correct kit.

I have an AsRock AB350M Pro4 and this 2x8GB EVGA DDR-3000 kit. With the latest BIOS for my board, all I had to do was enable XMP and then override to DDR-3200. The CPU overclock was similarly simple by just bumping the frequency for P-state 0 to 3.6; no voltage bumps required for anything. It could probably go higher, but I'm using the stock cooler and don't need to squeeze every last MHz out of it.

# ¿ May 17, 2017 20:11

SamDabbers: May 26, 2003

wargames posted:

Post like this make me want to get a zen machine up and find a cheap colo place to keep it at.

Hell yeah, as soon as there's a "server/workstation" AM4 board out with things like onboard IPMI and explicit ECC support. Hopefully something like that is in the works at SuperMicro or AsRock for mATX 1U applications. Intel's E3 line would basically be worthless overnight.

# ¿ May 18, 2017 18:48

SamDabbers: May 26, 2003

My kit's XMP profile says 15-16-16-35 and the BIOS on my board bumps it to 16-16-16. Maybe CL, T_RCD, and T_RP need to be set equal with this version of AGESA?

# ¿ May 19, 2017 18:05

SamDabbers: May 26, 2003

Ryzen 7 + X370 + Dual GPU would make for a sweet no-compromise multiseat setup though. It'd be the talk of the LAN party :awesome:

# ¿ May 22, 2017 01:34

SamDabbers: May 26, 2003

Some Ryzen Linux Users Are Facing Issues With Heavy Compilation Loads

# ¿ Jun 9, 2017 00:41

SamDabbers: May 26, 2003

I was browsing my local Micro Center store's open box clearance rack yesterday and was able to pick up an AsRock x370 Taichi, complete in original box, for $80.

Micro Center owns.

Fake edit: Anyone interested in a discounted AsRock AB350M Pro4?

# ¿ Jun 22, 2017 01:47

SamDabbers: May 26, 2003

Wirth1000 posted:

Is there as much unreliability and issues for the X370 as there is for the B350? I assume it's just Zen being immature but I'm curious if a step up in chipset is also a step up in build quality.

They're the same chipset with some features disabled on the B350, much like the R7 1700 and R5 1600 are the same with a couple cores disabled on the latter. So yes, the issues are basically the same if the issue is with the chipset.

That said, the build quality and auxiliary component selection on some X370 boards might be better, considering they're targeting the top end of the market segment.

SamDabbers fucked around with this message at 02:11 on Jun 22, 2017

# ¿ Jun 22, 2017 01:57

SamDabbers: May 26, 2003

shrike82 posted:

Thanks, I've set XMP for my ram. I'm wary about overclocking the CPU, I understand that doing so disables idling?

Use P-State overclocking if your BIOS supports it. It's probably in a menu under AMD CBS.

# ¿ Jun 22, 2017 16:56

SamDabbers: May 26, 2003

shrike82 posted:

One downside is I can't overclock my PC when running full on all 8 cores using the Wraith Spire cooler. I hit 80+C and the system shuts down for overtemp protection.
I've noticed that the CPU idle/load temps don't scale up linearly with the OC rates - I tried overclocking from 3.7GHZ down to base and there's a sharp rise in temps once you overclock at all.

I've got my 1700 running at 3.7 (P-State overclock) with a 212 EVO and it barely creeps above 50C with all threads maxed out by x264. Unless you're going to crank the voltage and try to hit 3.9+ a decent air cooler is totally viable.

# ¿ Jun 25, 2017 03:20

SamDabbers: May 26, 2003

SoftNum posted:

What kind of OS / setup are you using? I'm thinking about taking the plunge into a Ryzen setup, as I love me some AMD and cores. I was initially pondering PCIe passthrough, but I might just use Windows host and hyperV whatever else I want.

I'm running Fedora 25, and have a GTX 970 with the proprietary drivers. Everything runs buttery smooth, including native Steam games. I do use KVM with virt-manager/virsh for tinkering and it works just as well on Ryzen as on my Haswell laptop.

I'll eventually try PCIe passthrough with a Windows VM and a second GPU, but I'm waiting for the kinks to be ironed out first. Dual-booting is probably recommended for the time being if you want full performance in both Linux and Windows. I think it's likely that AMD will address the GPU passthrough/NPT issue eventually since they've bet the farm on the Zen architecture. It's just a matter of time.

# ¿ Jun 25, 2017 18:52

SamDabbers: May 26, 2003

There hasn't been any indication that it can't be fixed with microcode though. It's premature to say it can only be fixed with a new stepping until we hear from AMD engineering, who may have higher-priority issues they're working on, like the IRETQ bug. That said, NPT and IOMMU work properly, just not simultaneously on the same KVM VM. We'll just have to wait and see if a fix can be made in hypervisor, microcode, or both. Has anyone tried PCIe passthrough with NPT enabled in a hypervisor other than KVM? I know that's the favorite for GPU passthrough since it has knobs to work around NVIDIA's virtualization lockout; maybe it works better with ESXi?

If you *need* GPU passthrough *now* then Ryzen is not the platform for you. For me, it'd be nice to have, because AAA games in Windows aren't a big priority for me, and I can do everything else on this system as it is.

# ¿ Jun 25, 2017 20:15

SamDabbers: May 26, 2003

eames posted:

This reddit thread has some *very* recent info on the topic, looks like one Ryzen use could in fact confirm that it's a KVM bug and a AMD dev will look into it tomorrow.

https://www.reddit.com/r/VFIO/comments/6iomlh/apparently_the_kvm_amdnpt1_performance_bug_is/

Confirmed: it's a KVM issue. PCIe passthrough with NPT enabled on Xen does not have the problem!

# ¿ Jun 26, 2017 00:33

SamDabbers: May 26, 2003

sincx posted:

Doesn't every 1 degree rise in temperature increase hard drive failure rates by a few percent a year?

Citation needed. Google found otherwise: (see section 3.4)

Google posted:

We first look at the correlation between average tem-
perature during the observation period and failure. Fig-
ure 4 shows the distribution of drives with average tem-
perature in increments of one degree and the correspond-
ing annualized failure rates. The figure shows that fail-
ures do not increase when the average temperature in-
creases. In fact, there is a clear trend showing that lower
temperatures are associated with higher failure rates.
Only at very high temperatures is there a slight reversal
of this trend.

# ¿ Jun 28, 2017 02:00

SamDabbers: May 26, 2003

nerox posted:

Hello friends, all my Ryzen stuff finally got here today.

I did a quick overclock and I got my Ryzen 7 1700x @ 1.38V, 3.9ghz and temps are hovering 60-65 Celsius. I really want to try to get to 4.0 ghz, but looking for information I am seeing max safe temp ranged as anywhere from 75-95 and safe voltage as anywhere from 1.4 to 1.45. Anyone know what it actually is?

Unfortunately memory is sitting at 2300mhz. When I tried to do the XMP profile 1, my computer went into a reboot loop and I had to cmos reset to fix it.

Ryzen overclocking presentation by an AMD employee, including recommended limits:
https://www.youtube.com/watch?v=vZgpHTaQ10k&t=236s

tl;dw: CPU Vcore not higher than 1.425V, SOC at 1.1V can help memory overclocks, RAM can go anywhere from 1.35V to 1.5V depending on the stick

My R7 1700 non-X is running rock-solid at 3.7GHz on Pstate0 with a +0.0125V Vcore offset and level 3 (middle setting) LLC. It barely gets to 50C with 16 threads of prime95 with a Hyper 212 EVO. I'm using this DDR4-3000 15-16-16-35 kit which runs at DDR4-3200 16-18-18-38 on 1.35V.

SamDabbers fucked around with this message at 23:15 on Jun 28, 2017

# ¿ Jun 28, 2017 22:52

SamDabbers: May 26, 2003

Threadripper is an objectively awesome name. Some might even say it's EPYC.

# ¿ Jun 29, 2017 22:30

SamDabbers: May 26, 2003

Cygni posted:

a dude with those huge plastic dragon computer cases

Yes, that's threadripper's target market

# ¿ Jun 30, 2017 01:17

SamDabbers: May 26, 2003

I give it a B- since they didn't paint the floppy and optical drives to match. :krad:

airbrush work on that tiger though

# ¿ Jun 30, 2017 01:41

SamDabbers: May 26, 2003

Paul MaudDib posted:

Once they get a new stepping and fix the few virtualization/NPT bugs it's good,

As it turns out, the IOMMU + NPT bug is in KVM, not the silicon or microcode. Passing through the same GPU with NPT enabled under Xen does not exhibit the performance degradation. It has been acknowledged on the KVM mailing list, but a fix is still pending.

# ¿ Jul 7, 2017 05:43

SamDabbers: May 26, 2003

incoherent posted:

Could they throw out a last min 18 core threadripper or has that left the station for this cycle?

Considering that Threadripper is physically two 8-core Ryzen dies on the same package, 16 cores and 32 threads is the limit. The EPYC chips have four of the 8-core Ryzen dies on the same package, but that's not socket-compatible with Threadripper.

# ¿ Jul 7, 2017 06:04

SamDabbers: May 26, 2003

SamDabbers posted:

Ryzen owns for multitasking

Checking in two months later; Ryzen is still cool and good and has only gotten better with each BIOS and microcode update.

Verdict: extremely satisfied :mmmhmm:

# ¿ Jul 14, 2017 20:59

SamDabbers: May 26, 2003

CapnBry posted:

Regarding IPC, if I am currently running a i5-4690K @4.2GHz all four cores, would switching to a Ryzen 1700 clocked up to, say, 3.8GHz be a lateral move, upgrade, or downgrade for 1-4 core operations? I've been on the fence since the Ryzen release. On the one hand, I can definitely use the 8 cores for software build time (current project takes about 7 minutes to build maxing out all 4 cores) and video encoding, but I don't want to take a step backward on VR gaming where I feel like I am already on the edge. I'd hate to spend a grand to upgrade my 3 year old machine only to find I've actually lost performance on some workloads.

What if you had two computers? Keep the one you have for games, and build an R7 for work and video encoding? For a grand you could do a whole new Ryzen PC.

# ¿ Aug 1, 2017 22:46

SamDabbers: May 26, 2003

wargames posted:

2 cores per die, since half of the ~~ccx~~ dies are dummies for chip rigidity.

Fixed. One die has two 4-core CCXes. TR has two real dies and two dummy dies.

# ¿ Aug 4, 2017 03:45

SamDabbers: May 26, 2003

What you want is P-state overclocking, which leaves all the power management and idle-throttling capabilities enabled, but lets you tweak the top-end speed and voltage. It should be under an "AMD CBS" submenu in your bios, and there are easy to find guides on various sites.

BF1 is actually pretty decently multithreaded and can take advantage of Ryzen's ample cores. Enjoy your buttery smooth multitasking too!

Edit; not apropos of the above: I had to RMA my 1700 for the "performance marginality" segfault problem. AMD's RMA department took a few days to start the process once I opened the support ticket. They had me test a few times with various BIOS settings and send them a picture of the inside of my case to make sure I didn't put it together incorrectly, and then issued a prepaid FedEx label. The new chip arrived about a week after I shipped out the bad one, and it's a good sample which goes to 3.8GHz at stock voltage and doesn't segfault under heavy load. Thanks AMD! Ryzen satisfies, and so did AMD support.

SamDabbers fucked around with this message at 21:26 on Sep 24, 2017

# ¿ Sep 24, 2017 21:17

SamDabbers: May 26, 2003

Paul MaudDib posted:

I'd really like to see the segfault bug squashed before relying on it, especially under Linux/BSD...

I had an affected early production R7 1700, and the warranty replacement AMD sent me has been rock solid running at 3.8GHz on stock voltage (1.1825V). AMD's warranty service took about 2 weeks from ticket open to the replacement arriving. They had me send them a photo of the inside of my case to make sure I didn't have airflow issues, and test a few times with various BIOS settings before approving the RMA, but they did get me a good chip in the end.

# ¿ Nov 23, 2017 04:43

SamDabbers: May 26, 2003

I got an R7 1700 at launch and paid $110 for 16GB (2x8GB) of DDR4-2933, which I thought was expensive at the time. How much higher will it go before prices get back to reasonable levels again? I really want to upgrade to 32GB, but definitely can't justify it at these prices, even though my machine has started swapping at times when working with multiple VMs.

# ¿ Jan 26, 2018 19:39

SamDabbers: May 26, 2003

Kazinsal posted:

It's never going back down. Welcome to the idiot hellfucker dimension where the cost of building a PC is going back to 1980s levels.

On the other hand, while RAM prices are bad and stupid, they're the least of the worries we have in this bleak hellscape timeline.

# ¿ Jan 26, 2018 19:49

SamDabbers: May 26, 2003

The new BIOS for the X370 Taichi is fantastic. You can now tweak every P-state setting (FID, DID, and VID) in the AMD CBS menu and it will actually work.

I got my R7 1700 up to 4.0GHz at 1.2675V with no LLC, and it appears to be stable under load testing! Previously, I was unable to get it above 3.8GHz without running into the "multibug" and having it stuck in P1 at 2.7GHz.

Also, every single PCIe device attached directly to the CPU lanes is in its own IOMMU group now. The PCH still does not do PCIe ACS so all the devices connected to it still show up in one big group.

# ¿ Feb 9, 2018 21:12

SamDabbers: May 26, 2003

With low enough expectations, one can only be pleasantly surprised.

# ¿ Mar 6, 2018 21:40

SamDabbers: May 26, 2003

There's a new BIOS out for the X370 Taichi that contains a new AGESA. It appears to do little else but update the microcode on Raven Ridge to support IBPB instructions for Spectre V2 mitigation.

# ¿ Mar 8, 2018 05:01

SamDabbers: May 26, 2003

Khorne posted:

I'm not going to scam myself by buying a side grade to my 6 year old processor.

It may be a side grade in single threaded performance, but double the cores. Multitasking is buttery smooth on Ryzen.

# ¿ Apr 2, 2018 02:51

SamDabbers: May 26, 2003

I guess if your main use case is gaming and your current machine is CPU bottlenecking the games you want to play, then Ryzen might not be the best option for you.

# ¿ Apr 2, 2018 02:59

SamDabbers: May 26, 2003

Yeah I have no complaints with my 4.0GHz 1700, but my monitors are 1080p/60 and I do more with it than just gaming. If only that sweet Samsung B-die wasn't so outrageously expensive...

Khorne posted:

The only other annoying thing to upgrading is picking between win7+wufuc even though support drops in 2020, using linux on the desktop and being unable to play certain games, or using win10 and being forced to play games in fullscreen (+other issues, due to how it handles >60Hz screens and doing stuff on secondary monitors). I don't like any of those options. I mean I like linux, but I don't like not being able to play whatever I feel like and I already have linux on my work laptop. I wish MS weren't lazy assholes who have left win8/win10 busted for gaming. The MS store is just another reason to boycott win10 for me, too.

My dude, Ryzen is excellent for virtualization including PCIe passthrough. Run the best OS for each game, at essentially native speeds!

SamDabbers fucked around with this message at 03:38 on Apr 2, 2018

# ¿ Apr 2, 2018 03:28

SamDabbers: May 26, 2003

Methylethylaldehyde posted:

Buy 4 sticks of this stuff: Samsung branded memory, Part # M391A2k43BB1-CRC. It's basically OEM brand memory, so it takes a bit of hunting to find it, and it's not cheap when you do, each stick cost me ~$250 when I bought it. If you hunt around you can find other sticks with the same dies in them, but unbuffered ECC is hard to find.

ECC also lets you overclock the memory subsystem really easily, since it actually gives you feedback on how close to the ragged edge you're pushing it, via making GBS threads a ton of errors into your syslog once you push it hard enough to start getting single bit errors.

You can pretty consistently get those 2400 CL17 sticks to 2800-2933 16-16-16-36-1T, depending on phase of the moon, motherboard, and all the usual factors.

Thanks for this. I grabbed two sticks to use with my R7 and was able to get them to DDR4-2933 16-16-16-34-1T without bumping the voltage above the stock 1.2V, and the ECC function is recognized on my x370 Taichi. They were kinda pricey at $220/stick, but that's what the 2x16GB G.Skill Non-ECC B-die kits are going for right now on Newegg.

I bought my sticks from these guys, and they came quickly and well-packed.

# ¿ Apr 6, 2018 19:44

SamDabbers: May 26, 2003

Combat Pretzel posted:

Regarding B-die memory, what are the chances I will be able to push it a lot, regardless of what the sticks say (e.g. pushing a DDR4-2400 stick to 3200)? Mostly interested about this to get DDR4-3200 ECC sticks.

I got a pair of 16GB B-die EUDIMMs rated for DDR4-2400 CL17 at 1.2V up to DDR4-2933 CL16 at stock voltage on my 1700 without any effort. If it's possible to drive these to 3200, it'll require more voltage and/or relaxed timings, but I haven't spent the time to fiddle.

tl;dr: B-die is the best for Ryzen

# ¿ Apr 23, 2018 20:57

SamDabbers: May 26, 2003

Measly Twerp posted:

So an update on my virtual machine issues. The newer BIOS versions for my motherboard did break something, specifically when overclocking. Here's the thread over on the Level1Techs forum.

This turned out to be correct and running at CPU stock clocks fixes it. Of course the older BIOS versions did not have this issue so... gently caress ASRock?

The latest BIOS for my X370 Taichi (4.60, PinnaclePI 1.0.0.1) exhibits the same problem with unstable TSC when overclocking. In my case, any OC via the BIOS also only seems to affect core 0, and all the other ones top out at stock frequencies under load. I reverted to 4.40 and everything works properly again. The newer AGESA introduced IBPB instructions for Spectre v2 mitigation, so hopefully the next release fixes things so we can have our overclocks and IBPB too.

One thing I haven't yet tried is overclocking once the OS has loaded a la Ryzen Master. I'm running Linux though, so it takes a little more fiddling. There's a ZenStates.py script out there that can adjust the frequency and voltage MSRs, but it needs the msr kernel module which isn't built in the stock Fedora kernels, and I've been too lazy to take the time to fiddle.

# ¿ Apr 24, 2018 12:51

SamDabbers: May 26, 2003

I can't be the only one who wants to see a workstation-class ARM chip, right? Will ARM always be relegated to low power implementations?

# ¿ Apr 27, 2018 15:47

SamDabbers: May 26, 2003

Quoting for relevance:

Methylethylaldehyde posted:

Buy 4 sticks of this stuff: Samsung branded memory, Part # M391A2k43BB1-CRC. It's basically OEM brand memory, so it takes a bit of hunting to find it, and it's not cheap when you do, each stick cost me ~$250 when I bought it. If you hunt around you can find other sticks with the same dies in them, but unbuffered ECC is hard to find.

ECC also lets you overclock the memory subsystem really easily, since it actually gives you feedback on how close to the ragged edge you're pushing it, via making GBS threads a ton of errors into your syslog once you push it hard enough to start getting single bit errors.

You can pretty consistently get those 2400 CL17 sticks to 2800-2933 16-16-16-36-1T, depending on phase of the moon, motherboard, and all the usual factors.

I can confirm that this part number goes up to 2933 CL16 without effort, but I haven't been able to get beyond that. My machine (X370 Taichi, R7 1700) throws ECC errors or outright crashes if I try to either tighten the timings or increase the clock rate. Still, good for 2x16GB sticks on that platfrom. Zen+ might do even better.

They're still ~$220 per stick: https://www.ebay.com/itm/322375046134

# ¿ May 25, 2018 17:03

Adbot: ADBOT LOVES YOU

# ¿ Apr 30, 2024 04:16

SamDabbers: May 26, 2003

mdxi posted:

In the lead-up to the Ryzen 2X00 release, I had planned to replace my 1700 with a 2700, and put the 1700 in a partial build that I had put on hold. But some life stuff came up, and I had to put that cash elsewhere for a couple months. Yesterday I decided to finish out the build, but to just keep my desktop as-is and slot a 1X00 into the second machine. My local microcenter had restocked, and had 10+ R5 1600s at $150, so I snagged one of those.

Now I have 28 threads, in two machines, crunching WorldComputeGrid workunits 24/7. Two years ago that would have cost me, what, $2500 in Intel CPUs? Total CPU outlay here: $450. Since neither of these machines is a gaming rig, I'll probably keep their mobos and memory and swap out CPUs when the Zen2s arrive.

Well? Show us some sweet graphs of these monsters in action

mdxi posted:

Finally, it looks like the continued BIOS and/or kernel upgrades have improved performance under Linux. When I first rebuilt my desktop with the 1700, all cores were running at their base clock of 3.0GHz. Yesterday after initial install, I checked the speeds of the 1600s cores, and the kernel was reporting 3.4GHz. I thought I remembered (and Wikipedia confirmed) that the base clock of a 1600 was 3.2GHz -- but the all-cores boost clock is 3.4GHz, and the CPU is continuing to run at that speed even after about 18 hours of 100% utilization. The reason I think this is due to BIOS/kernel improvements is that my 1700 is now running all threads at 3.2GHz, which it definitely was not doing before. I'll take it.

Edit: interestingly, the 1600 is running all cores at 3393MHz, with just fractional clock differences. The 1700 has a bit more of a spread: 6 cores at 3193MHz, 1 core at 3187MHz, and 1 core at 3184MHz.

Have you tried overclocking? You're leaving a significant amount of performance on the table if you don't. My 1700 gets to 4.0GHz on 1.275V with 2x16GB ECC B-die at DDR4-2933 CL16. That's about ~30% more clock speed, to say nothing of the effects of high speed memory. The ECC RAM gets noisy throwing errors when I try to drive the DIMMs or the memory controller too hard, so it's easy to figure out how far you can push it.

A cool dude wrote a script (https://github.com/r4m0n/ZenStates-Linux) that lets us overclock at runtime on Linux like Windows users can with the Ryzen Master app. I find it both easier and more reliable than trying to work around some of the overclocking-related UEFI bugs on my system. It won't do anything with RAM speed or timings though, so get those settings dialed in with the UEFI.

code:

$ sudo zenstates.py -l
P0 - Enabled - FID = A0 - DID = 8 - VID = 2C - Ratio = 40.00 - vCore = 1.27500
P1 - Enabled - FID = B4 - DID = A - VID = 3A - Ratio = 36.00 - vCore = 1.18750
P2 - Enabled - FID = D8 - DID = 10 - VID = 50 - Ratio = 27.00 - vCore = 1.05000
P3 - Disabled
P4 - Disabled
P5 - Disabled
P6 - Disabled
P7 - Disabled
C6 State - Package - Enabled
C6 State - Core - Enabled
$ sudo dmidecode -t memory | grep -E '(Manufacturer|Part|Speed)' | grep -v Unknown
	Speed: 2934 MT/s
	Manufacturer: Samsung
	Part Number: M391A2K43BB1-CRC    
	Configured Clock Speed: 1467 MT/s
	Speed: 2934 MT/s
	Manufacturer: Samsung
	Part Number: M391A2K43BB1-CRC    
	Configured Clock Speed: 1467 MT/s

I have a systemd unit that'll enable my overclock settings at boot:

code:

$ cat /etc/systemd/system/overclock.service 
[Unit]
Description=Configure CPU overclock

[Service]
Type=oneshot
RemainAfterExit=true
ExecStart=/usr/local/sbin/overclock --enable
ExecStop=/usr/local/sbin/overclock --disable

[Install]
WantedBy=multi-user.target
$ cat /usr/local/sbin/overclock
#!/bin/bash

ZENSTATES="/usr/local/sbin/zenstates.py"

while [[ $# -gt 0 ]]
do
case $1 in
  -e|--enable)
# Overclocked values:
# P0 - Enabled - FID = A0 - DID = 8  - VID = 2C - Ratio = 40.00 - vCore = 1.27500
# P1 - Enabled - FID = B4 - DID = A  - VID = 3A - Ratio = 36.00 - vCore = 1.18750
# P2 - Enabled - FID = D8 - DID = 10 - VID = 50 - Ratio = 27.00 - vCore = 1.05000
  P0="-p0 -f a0 -d 8  -v 2c"
  P1="-p1 -f b4 -d a  -v 3a"
  P2="-p2 -f d8 -d 10 -v 50"
  shift
  ;;
  -d|--disable)
# Original values:
# P0 - Enabled - FID = 78 - DID = 8  - VID = 3A - Ratio = 30.00 - vCore = 1.18750
# P1 - Enabled - FID = 87 - DID = A  - VID = 50 - Ratio = 27.00 - vCore = 1.05000
# P2 - Enabled - FID = 7C - DID = 10 - VID = 6C - Ratio = 15.50 - vCore = 0.87500
  P0="-p0 -f 78 -d 8  -v 3a"
  P1="-p1 -f 87 -d a  -v 50"
  P2="-p2 -f 7c -d 10 -v 6c"
  shift
  ;;
  *)
  echo fart!
  exit 1
  ;;
esac
done

$ZENSTATES $P0
$ZENSTATES $P1
$ZENSTATES $P2

SamDabbers fucked around with this message at 13:14 on Jun 5, 2018

# ¿ Jun 5, 2018 13:06

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > AMD CPU and Platfrom Discussion Episode IV: A New Hope is Ryzen

«‹›2 »