1usmus Custom Power Plan for Ryzen 3000 Zen 2 Processors

[Cidious]

3700X on an MSI X570-A Pro.

I have a baseline on PassMark from 2019-09-05 to compare with (AGESA 1.0.0.3ABB)

Comparing to the latest BIOS I have access to today (AGESA 1.0.0.4B) plus the 1usmus power plan I find that my single core performance tends to be just a few points higher and it does, in fact, tend to keep heavy single threaded workloads on the best core (monitored using Ryzen Master) however it does seem to have transient moments where it tries to move it to a different core and then goes right back to the best core. However all the other tests tend to trend 1-2% lower that my baseline however that is independent from the power plan (I also checked the default AMD power plan) so apparently I lost a bit of performance from 1003abb to 1004b. I benched with both PBO set to disabled and set to Automatic and I saw a slightly better performance with PBO set to automatic but the difference was within 1% so, could just be a fluke. I rant the tests multiple times and it did tend to average higher with PBO on auto but it's a bit too close to call. Honestly I doubt you'd feel the difference in game but I can confirm that with 1usmus's power profile it WANTS to keep the heaviest load on your best core, despite the fact that it will occasionally try to move it for a second and then puts it right back on the best core. (C03 is my best core and it occasionally tried to move it to C01, then right back to C03, every time) where as the default AMD power plan tends to toss the single core test all over the place.

I hope this input helps, thanks for the work!

I can confirm these findings to be similar with my 3800X it's still hopping between threads but for me it's on the same core at least but different speeds. A bit funny. Performance wise nothing stunning to report.

1.0.0.4B didn't do anything over 1.0.0.3ABBA on my Gigabyte Aorus X570 Elite and 3900X, it lowered the max. boost a bit (from 4.650 to 4.575), but all the results stayed the same (within 1%) - processor doesn't do any work at max boost anyhow.

Tried 1usmus bios settings an power plan, no difference. Tried to push PBO overclocking, minimal increase in scores but massive amount of extra heat.

A bit underwhelming for a supposed massive bios overhaul...

Yes Agesa 1.0.0.4 B is very underwhelming for what was promised by AMD and MSI. 100+ 'improvements'... I got less performance out of it. Only useful thing was the offset options.

 

[zlobby]

P.S : Maybe we should create a petition, to claim 1usmus's immediate hiring in AMD

We better not. Once under contract, he will be required to play by their rules. His mandate often will be to drag his feet on purpose. He also won't be able to share much info as he'd be under NDA...

The community benefits more if he is a Free Willy.

 

[NC37]

Can I try it with my Ryzen 5 1600, is it gonna be an issue?

Tried it on my 1700. Scores 438 single core in CPUID before, scores 434 after. Ryzen Balanced from AMD scores 438 as well so, there is a very slight performance loss on first gen Ryzen with this powerplan.

 

[HwGeek]

We better not. Once under contract, he will be required to play by their rules. His mandate often will be to drag his feet on purpose. He also won't be able to share much info as he'd be under NDA...

The community benefits more if he is a Free Willy.

I hope that at-least he gets free samples of the new product to play with and make more mods ;-).

 

[SirMacke]

Asus Crosshair 6 Hero + 3700X
I cannot find Power Supply Idle Control anywhere.
Tested anyway, no improvement in R15/R20 and no change in them Mhz

Edit: It boosts lower now than ever. Oh well.

 

[Cresp0pt]

Hello, i cant find the cppc option on my msi b450 tomahawk and i have a 3600? Any help?

 

[Cidious]

After a while using the computer you do get to see nice boost clocks on almost all cores but those are not the sustained single core numbers, Just peak boosts.

I'll stick with the profile. It's not harming my system in any way. And it feels smooth to use. Temps are good and and performance is up to par as with 1.0.0.3 ABBA. I would recommend using this when you run 1.0.0.4 B. They promised golden mountains with 1.0.0.4 B but it's lacking and it wasn't that great at all. At least not on my MSI B450M Mortar MAX. might be different for other boards.

Thanks 1usmus. It's free progress. I hope AMD will at least look into it and maybe even pick it up to improve your ideas even further.

 

[ORLY]

Will it work properly on v1809 (LTSC) and 1.0.0.3abba?

 

[Crackong]

Thanks for the effort.
Sad to report that Asus Crosshair VIII Impact does not have the following settings:

Power Supply Idle Control
Cool n'quiet
PPC Adjustment

And does not benefit from your fantastic power plan.

 

[Divide Overflow]

Thanks for your efforts, 1usmus.
Unfortunately, I don't notice any difference from the Balanced power plan for AMD Ryzen.
3900x on X570 Aorus Master with 1.0.0.4B bios.
Like others, I was unable to locate many of the required bios settings.

 

[EarthDog]

Still wondering how successful this really is if users have...

1. The latest BIOS
2. The latest Windows OS version
3. Power plan set to AMD's.

It seems like for those who have at least a couple of the items listed, this doesn't do squat.

 

[VashdaNarata]

Hi, I am really interested in this, especially since I do use a lot of 1T software alongside the heavily threaded ones and the Windows 10 scheduler seems intent on putting the load on the weakest cores in the system. Unfortunately, I tried to use this power plan and it didn't affect anything. CPUz Single-thread uses two threads apparently, and one isn't even on the same CCX. The xEdit software for Fallout 4 uses just on thread on one core, and even with the power plan it still uses the weakest core, sometimes losing 100 MHz of performance which is quite significant.

I am on 1.0.0.3 ABBA and my motherboard is Asrock X570M Pro4, my CPU is Ryzen 7 3700X. Everything is at stock.

However, I couldn't find the:

• CPPC
• CPPC Preferred Cores

Options in the bios. They are simply not there, unless I missed them. So maybe that is why it is not working for me. Any advice would be appreciated. Thanks

This is assuming the bios are the same for the x470 taichi (i know the x570 taichi has them in the same place as well)

For those settings, go to Advanced\AMD CBS\NBIO Common Options\SMU Common Options

 

[nangu]

Thanks for your efforts, 1usmus.
Unfortunately, I don't notice any difference from the Balanced power plan for AMD Ryzen.
3900x on X570 Aorus Master with 1.0.0.4B bios.
Like others, I was unable to locate many of the required bios settings.

You can locate most of the required bios settings under the CBS and Additional CPU settings on the Master. The ones which are not included are Cool & Quiet and one of the two CPCC options.

Anyway, it didn't do much on my system, I only see a 25Mhz peak increase on two cores, but sustained is the same as with the Ryzen High Performance plan. CPUz and Cinebench R15/R20 ST and MT scores are well inside the error margin.

I think the 1usmus power plan will be benefitial for those systems which the Windows scheduler is not favouring the best cores. I never had that problem with any AGESA version, so that's may be why I don't see any improvement.

Also, I noticed the 1.0.0.4 Bios is not jumping frequencies all over the place. Instead the frequency stays more consistent so leveraging in a more planar curve in time. It may give us more smooth and consistent framerates in games, but I didn't test it extensively.

 

[hzy4]

Asus X570 TUF with BIOS AGESA 1.0.0.4 has added more settings like the CPPC, unfortunately still no spread spectrum setting.

 

[Dyatlov A]

Windows 7 compatible?

 

[1usmus]

Windows 7 compatible?

you can try, there is no threat to the computer

 

[Weshya]

Still wondering how successful this really is if users have...

1. The latest BIOS
2. The latest Windows OS version
3. Power plan set to AMD's.

It seems like for those who have at least a couple of the items listed, this doesn't do squat.

I had the latest BIOS 3.40 (AGESA 1.0.0.4B) on my Tomahawk MAX B450 , latest chipset drivers installed (+amd's power plan), and Windows 1909 installed. The difference on the behavior of my Ryzen 3600 was like day and night after applying 1usmus power plan and recommended BIOS settings. Finally the fastest cores are utilized at a much higher and sustained clock frequency than before. So in my case, the combination of right BIOS settings + 1usmus powerplan finally paid off .

 

[wolf]

Installed 1.0.0.4B BIOS and the power plan, seems to boost gaming clocks by 25-50 mhz on average, I'll take it.

Annoyingly with the BIOS updates, I always lose my profiles, I assume they can be backed up to HDD, if so I gotta start doing that rather that going through every screen again, especially RAM timings...

Also every time I do a new BIOS some games reset to default, tripped me out thinking I was getting MUCH lower FPS after doing this last night but DOOM had reverted to OPEN GL 4.5... eww

 

[BitsInBytes]

Many of you are putting WAY to much focus into R15 or R20 for your testing and validation​

I created an account literally just to say that; I think the biggest gains can only be seen while using your PC (I think, the way boost was intended).

I have the same PC as 1usmus, down to the motherboard even but in R20 my 3900x I was seeing max boost clock of the same old 4400 mhz and almost reverted the change. I had the new bios, correct settings, installed power plan...yet in R20 my single score sucked. But I decided to keep the power plan and run HWINFO in the background while doing things like using chrome, doing work, gaming, etc - and there is a HUGE difference in performance that R20 did not reveal. Before this power plan, most of my cores only sometimes got above 4200 mhz with just a few lucky ones doing 4400mhz.

However, now with the power profile while "just using" my PC I am seeing all my cores reaching at least up to 4500 mhz every minute with over half of my cores able to hit 4575 mhz constantly (assuming something needs the CPU and my system is not idle)!!!

Across all cores (not including rendering) I can see an all core speed of 4400 mhz! So I loaded up a memory laitencey test in AIDA64...and it detected a running speed of 4575 vs the normal 4200-4400 and my latency dropped from 68 down to 66! In gaming my FPS is improved, just using my system everything is snappier and I love seeing my CPU boosting constantly.

So for those with the new bios version, with right right settings, etc - Try just using your PC while watching core speeds and I think you will be surprised and happy!

**EDIT**

If things are not working for you, disable PBO / auto overclocking. At least for me I have discovered if I attempt to enable PBO or use the auto overclocking features like the +500 mhz per core, I am back to square one. Not sure I understand why this is happening so I need to leave PBO settings in my motherboard to [Auto]. It does not seem to matter which setting I change, the second I enable any overdrive feature, adjust EDC (or others) or overclocking I am back to 4400 mhz top and my performance drops across the board.

 

[Weshya]

The difference on the behavior of my Ryzen 3600 was like day and night after applying 1usmus power plan and recommended BIOS settings.

Many of you are putting WAY to much focus into R15 or R20 for your testing and validation​

I created an account literally just to say that; I think the biggest gains can only be seen while using your PC (I think, the way boost was intended).

I have the same PC as 1usmus, down to the motherboard even but in R20 my 3900x I was seeing max boost clock of the same old 4400 mhz and almost reverted the change. I had the new bios, correct settings, installed power plan...yet in R20 my single score sucked. But I decided to keep the power plan and run HWINFO in the background while doing things like using chrome, doing work, gaming, etc - and there is a HUGE difference in performance that R20 did not reveal. Before this power plan, most of my cores only sometimes got above 4200 mhz with just a few lucky ones doing 4400mhz.

However, now with the power profile while "just using" my PC I am seeing all my cores reaching at least up to 4500 mhz every minute with over half of my cores able to hit 4575 mhz constantly (assuming something needs the CPU and my system is not idle)!!!

Across all cores (not including rendering) I can see an all core speed of 4400 mhz! So I loaded up a memory laitencey test in AIDA64...and it detected a running speed of 4575 vs the normal 4200-4400 and my latency dropped from 68 down to 66! In gaming my FPS is improved, just using my system everything is snappier and I love seeing my CPU boosting constantly.

So for those with the new bios version, with right right settings, etc - Try just using your PC while watching core speeds and I think you will be surprised and happy!

+ 1

That's exactly the behavior i noticed with my humble Ryzen 3600. More core boosting at their max, more frequently while using my PC (HWiNFO in the background)

 

[willgart]

I'm not lucky
3700X on an Asus Tuf Gaming x570, just updated with the 1.0.0.4b bios. (stock cooler for now)
and I can reach only 4250mhz maybe 25mhz more during R20 single core tests
and the frequency chart still have up / down movement, maybe less when all the recommended setup activated. but still there.

because I dont see the temperature raising so much, I dont think that the problem is related to the default cooler, or is it?

with manual overclocking, I can have 4200 to 4225 to all cores (sometimes R20 crashes), no change, its what I had before.
and my temperature go up to 93deg in this case.
I plan to get a dark rock pro 4 to stay cool and quiet all the time.

 

[BitsInBytes]

Could very well be temp limited, I know ryzen 3000 boost is extremely temp dependent. I am running a water cooling system so unless I am rendering, all core load is only 45-55C. Rendering is about 70-75C.

 

[Cidious]

Many of you are putting WAY to much focus into R15 or R20 for your testing and validation​

I created an account literally just to say that; I think the biggest gains can only be seen while using your PC (I think, the way boost was intended).

I have the same PC as 1usmus, down to the motherboard even but in R20 my 3900x I was seeing max boost clock of the same old 4400 mhz and almost reverted the change. I had the new bios, correct settings, installed power plan...yet in R20 my single score sucked. But I decided to keep the power plan and run HWINFO in the background while doing things like using chrome, doing work, gaming, etc - and there is a HUGE difference in performance that R20 did not reveal. Before this power plan, most of my cores only sometimes got above 4200 mhz with just a few lucky ones doing 4400mhz.

However, now with the power profile while "just using" my PC I am seeing all my cores reaching at least up to 4500 mhz every minute with over half of my cores able to hit 4575 mhz constantly (assuming something needs the CPU and my system is not idle)!!!

Across all cores (not including rendering) I can see an all core speed of 4400 mhz! So I loaded up a memory laitencey test in AIDA64...and it detected a running speed of 4575 vs the normal 4200-4400 and my latency dropped from 68 down to 66! In gaming my FPS is improved, just using my system everything is snappier and I love seeing my CPU boosting constantly.

So for those with the new bios version, with right right settings, etc - Try just using your PC while watching core speeds and I think you will be surprised and happy!

Although I do agree with most parts you are writing. There is a reason we use Cinebench and other tools to confirm a gain. Just seeing clocks rise in HWINFO doesn't equal better performance. That's just placebo. If you're happy with that, that's ok for me. But I would like to actually find out if my processor performs better in applications that I use. Which is difficult to do with gaming when my GPU (2070 Super) will be the bottleneck at 1440p. A quick way to test if you get more performance is using a tool that utilizes does cores like Cinebench. And then my Single cores boost slightly higher but no where near those peak speeds. Those peaks speeds you read in HWINFO are useless. Your processor is not doing actual work at those speeds. It's nice to see yes. But it's nicer to actually know it's making a difference. Which from my findings it barely does. Don't get me wrong I am using the profile because just like you I feel the system is snappy and works very well. But so did the Ryzen Balanced profile. I see no harm in the profile but the benefit for me is not really there also (mind you 3800X). You can say we rely too much on Cinebench. But I think in this case you rely too much on some core speed numbers in your monitoring software which doesn't translate into real performance gains in any way.

So thank you for making this post and signing up. Happy you have the same system as 1usmus that it works for. But you don't have to judge all of us for trying to figure out if we actually get something out of it or not in BOLD CAPITAL LETTERS. Seems a bit rude to me.

 

[willgart]

Could very well be temp limited, I know ryzen 3000 boost is extremely temp dependent. I am running a water cooling system so unless I am rendering, all core load is only 45-55C. Rendering is about 70-75C.

do you know the bios options to increase the supported temperature?

 

[Cidious]

Here is the difference between power profiles on my system. Same test ran over again. so I didn't switch power profiles during the test. Screenshotted at the same moment. Same power limit values. Same temps. Same core hopping. Same clockspeeds.. well.. spot the differences.

The boost in performance people are noticing might be more from finally using the correct bios settings than the profile. But I own a 3800X. Things might be different on the 3900X