AGESA 1.2.0.8

[Taraquin]

This - but mostly you just need to not get one of the WORST motherboards, 99% of them can achieve 3600 with four ranks. Usually SoC voltage is all thats needed, but some rare boards just wont do it at all (my ROG STRIX B450-I just refused to do above 3200, no matter what)


True, but it gets into that complex math that hurts my brain with how they need to be tweaked in relation to each other

No one has examples of Hynix MJR to compare with, because no one overclocks 32GB DDR4 sticks

Example from reddit:
View attachment 294300

I found one random example on reddit, i'll give these a shot and see if they work
I dont have high hopes since my tRFC needed to be over 700 to POST, but theres a chance his other tweaks allow it to work
View attachment 294301

I would mainly focus on RRD\FAW since they affect gaming most I have tuned Hynix AJR which kinda behaves similar to MJR but can run RFC a bit lower.

Suggestion:
Try RRDS 4, RRDL 6, FAW 16 first, see if it works. If it does try RDWR 11, WRRD 3. If that works try RC 66. You already get 80% of the gains from those 3 tweaks The more advanced ones are trying lower RCDRD, SCL`s and WTR`s.

 

[Mussels]

That guy ran some insane voltage which i wont copy

As an example, C16 wont boot no matter what, needs a CMOS clear even at 2667
C17 with GDM off, is fine at 3800

It's annoying the order of timings is different in the BIOS to Zentimings, etc. Makes me need to triple check names and values of everything.

tRCDRD wont post at 19, posts with memtest errors at 20, and seems stable at 21 - with 22 being the XMP Default.
These are ran pretty close to their limits on some timings out of the box, since they dont seem to scale with MHz. Others seem to have massive headroom left over.

I would mainly focus on RRD\FAW since they affect gaming most I have tuned Hynix AJR which kinda behaves similar to MJR but can run RFC a bit lower.

Suggestion:
Try RRDS 4, RRDL 6, FAW 16 first, see if it works. If it does try RDWR 11, WRRD 3. If that works try RC 66. You already get 80% of the gains from those 3 tweaks The more advanced ones are trying lower RCDRD, SCL`s and WTR`s.

Those worked first try. nicely done.
they were all named slightly differently (usually just the t at the start) which was confusing, with so many similarly named timings. Yay for being new to this.

Unless you have a few other magic suggestions, this might be the final values - latency is definitely down. Other values have that regular random variation to them with Aida, didn't really worry about re-testing over and over for the best results - latency is down, games will like.

 

[Taraquin]

That guy ran some insane voltage which i wont copy

As an example, C16 wont boot no matter what, needs a CMOS clear even at 2667
C17 with GDM off, is fine at 3800

It's annoying the order of timings is different in the BIOS to Zentimings, etc. Makes me need to triple check names and values of everything.

tRCDRD wont post at 19, posts with memtest errors at 20, and seems stable at 21 - with 22 being the XMP Default.
These are ran pretty close to their limits on some timings out of the box, since they dont seem to scale with MHz. Others seem to have massive headroom left over.




Those worked first try. nicely done.
they were all named slightly differently (usually just the t at the start) which was confusing, with so many similarly named timings. Yay for being new to this.

Unless you have a few other magic suggestions, this might be the final values - latency is definitely down. Other values have that regular random variation to them with Aida, didn't really worry about re-testing over and over for the best results - latency is down, games will like.
View attachment 294309View attachment 294310

Nice! Since RC=RAS+RP you cpuld try RC 63. WTRS 4/WTRL 12 and RDRDSCL/WRWRSCL 4 should be easy to run You could try WR 20/RTP 10. If that works try WR 16, RTP 8. After that there is very little to be done, maybe RC could do slightly lower etc, but 90% of the gains you already have

 

[izy]

MSI (i think ASrock too) started to release AGESA ComboAm4v2PI 1.2.0.A. update, is this a new AGESA version?

New - 7C02v3G - 2023-04-27
- AGESA ComboAm4v2PI 1.2.0.A. update.
- TPM Out of Bounds Access security patch.

Last - 7C02v3G1(Beta version) - 2023-03-09
- Update to AGESA ComboAm4v2PI 1.2.0.8.
- Patch Potential Security Vulnerabilities of AMD Ryzen CPUs

 

[Mussels]

MSI (i think ASrock too) started to release AGESA ComboAm4v2PI 1.2.0.A. update, is this a new AGESA version?

New - 7C02v3G - 2023-04-27
- AGESA ComboAm4v2PI 1.2.0.A. update.
- TPM Out of Bounds Access security patch.

Last - 7C02v3G1(Beta version) - 2023-03-09
- Update to AGESA ComboAm4v2PI 1.2.0.8.
- Patch Potential Security Vulnerabilities of AMD Ryzen CPUs

There was some article put out by tomshardware that sounded like AM4's TPM could be easily hacked, this would be the fix for that

It uhh.... required a lot of effort and physical acess to be viable so the common user shouldnt care.

Nice! Since RC=RAS+RP you cpuld try RC 63. WTRS 4/WTRL 12 and RDRDSCL/WRWRSCL 4 should be easy to run You could try WR 20/RTP 10. If that works try WR 16, RTP 8. After that there is very little to be done, maybe RC could do slightly lower etc, but 90% of the gains you already have

The tighter timings booted, running some tests now

Never expected 64ns off this kit



slightly unstable, i'll work back a step

 

[Taraquin]

There was some article put out by tomshardware that sounded like AM4's TPM could be easily hacked, this would be the fix for that

It uhh.... required a lot of effort and physical acess to be viable so the common user shouldnt care.
View attachment 294400


The tighter timings booted, running some tests now

View attachment 294401View attachment 294403

Never expected 64ns off this kit



slightly unstable, i'll work back a step
View attachment 294415

What voltage are you running? WR 20/RTP 10 may fix the instability. Maybe RC needs to be higher aswell.

 

[Mussels]

I adjusted tRTP from 8 to 9 and tWR from 16 to 18 and things got a lot more stable, but not perfect yet
20 and 10 feels like where it'll stabilise, you're pretty spot on with the timings so far

1.4v, playing it safe thermally since its winter now, but summer here is spicy hot (40c+ ambients)



It's right on the edge of stability, i've been trying out a few combinations but still getting errors - interestingly since raising rtp to 10 and wr to 20, they're happening faster (maybe just heat buildup, there is a fan on the ram atm anyway)

Wish the errors helped me narrow down what timing is the issue

Edit: Oh, googling suggested to enable GDM. That didnt help. I'll work on raising the values we lowered today and see whats up, but any suggestions to speed up the process would be appreciated

 

[izy]

Anyone tested it yet? (AGESA ComboAm4v2PI 1.2.0.A)

 

[INSTG8R]

kombo strike is 100% just PBO's undervolt curve, 1.0.0.8 gave that to all x3D users on every board
MSI just added it in prior to AMD making it available to everyone

x3D behaves differently for some reason to any other Zen 3 CPU, it's like they measure the voltage and treat it as VID - My board lets me use the curve undervolt AND a voltage offset, and you can never make it unstable - it just downclocks instead so past a certain point you lose performance

Kombo strike 3/-30 UC, and -.05v is the limit before i take any performance hit at all, but -0.61 shaves a lot of power and heat for something like a 0.1% loss (that lets me run the RAM harder)


I beat w1zzards score, so i'm happy
I didnt even need a debloated OS!
View attachment 294287

Yeah that is a nice score I know how Kombo Strike works I have it at level3 plus -15 CO I am not brave enough to try any lower. This is my best score running that combo I tried all the possible options off/on. Just CO, Just Kombo Strike. Combined got me my best score
As for RAM I am running it stock timings just went from 3600 CL16 to 3800 CL18 and nothing else, voltage is 1.36V
This is the best I have gotten with all of the above options on
This a fresh R23 run since I haven't since my RAM OC my max voltage was 1.24V most cores held 4.4

 

[Taraquin]

I adjusted tRTP from 8 to 9 and tWR from 16 to 18 and things got a lot more stable, but not perfect yet
20 and 10 feels like where it'll stabilise, you're pretty spot on with the timings so far

View attachment 294426View attachment 294425

1.4v, playing it safe thermally since its winter now, but summer here is spicy hot (40c+ ambients)



It's right on the edge of stability, i've been trying out a few combinations but still getting errors - interestingly since raising rtp to 10 and wr to 20, they're happening faster (maybe just heat buildup, there is a fan on the ram atm anyway)

View attachment 294429

Wish the errors helped me narrow down what timing is the issue

Edit: Oh, googling suggested to enable GDM. That didnt help. I'll work on raising the values we lowered today and see whats up, but any suggestions to speed up the process would be appreciated

If you enable GDM CL, CWL, WR and RTP will be round up to even numbers. If not try raising RC by 1 or 2 or RCDRD or RP by 1. Sound heat reletad. Try lowering voltage to 1.35v. You may need to rause CL to 18, but try if 17 works. The only 2 ti ings that scale with voltage on Hynix DDR4 seems to be CL/CWL. Rest seems mostly unaffected.

 

[Sabotaged_Enigma]

Question: I've got a TUF B550M Plus Wi-Fi
Current AGESA is 1.2.0.3C
which agesa works best? 1.2.0.3c, 1.2.0.6, 1.2.0.7, or the latest 1.2.0.8?
anyone can help me with that? thx!

 

[Mussels]

level3 plus -15 CO

that cant work - they're over-riding each other.
They both change the same setting, only one of them is active

Question: I've got a TUF B550M Plus Wi-Fi
Current AGESA is 1.2.0.3C
which agesa works best? 1.2.0.3c, 1.2.0.6, 1.2.0.7, or the latest 1.2.0.8?
anyone can help me with that? thx!

with your older CPU (not an x3D) you're on a perfectly fine BIOS.
The only problem with newer is the EDC bug if you passed 90A on that CPU in the PBO settings, otherwise they behave the same.

Newer BIOS may have some improvements specific to your board, like RAM compatibility and so on - at stock values you'd see no changes

 

[INSTG8R]

that cant work - they're over-riding each other.
They both change the same setting, only one of them is active

We are getting a little off topic but if you're interested I have R23 runs of all combinations. Best result was with both

 

[freeagent]

which agesa works best? 1.2.0.3c, 1.2.0.6, 1.2.0.7, or the latest 1.2.0.8?

I think it depends on when your CPU was made, if it is an older CPU then you would want the older AGESA. But if its newer, made in the last year I think, then you would want a newer AGESA.

On my 5600X and 5900X I prefer 1203, but with X3D I prefer 1208. But my CPUs were bought right after launch, so they are old now..

 

[Mussels]

I think it depends on when your CPU was made, if it is an older CPU then you would want the older AGESA. But if its newer, made in the last year I think, then you would want a newer AGESA.

On my 5600X and 5900X I prefer 1203, but with X3D I prefer 1208. But my CPUs were bought right after launch, so they are old now..

I'll agree with those AGESA choices as well, unless you need something included from the future updates like the TPM fix (which didnt affect everyone, and RUFUS lets us say no to TPM anyway)

I run the latest on all my setups, just so they're all hot-swap compatible when something dies like my ITX rigs ROG B450-I, let me take the 5800x right out and straight to the x370

 

[olymind1]

Msi released a 1.2.0.A beta currently for b450 tomahawk containing security fixes.

 

[unwind-protect]

Msi released a 1.2.0.A beta currently for b450 tomahawk containing security fixes.

Now we also have to track security fixes, and that with the usually bad change notes?

 

[Mussels]

Now we also have to track security fixes, and that with the usually bad change notes?

There was a publicized TPM hack that required physical access, this will be to defeat that

 

[VulkanBros]

Also released for MEG X570 Unify.....

 

[JoeN]

Hi, hopefully it is ok to join this thread! I have a 5800x3d on an Aorus x570 mobo, I recently noticed a drop in performance, no specific figures, of around 5-8% following a BIOS update to an AGESA 1.2.0.8 version. The update notes mention 5800x3d improvements but I have also seen higher temps along with the performance drop.

Having seen the recent articles about burning x3d cpus I decided to try some undervolting with some success using the much maligned easytune. However it does not load on startup even if configured to do so.

I have a ticket open with Gigabyte, running for over a month, they finally came back with the suggestion that overclocking is not possible as per AMD rule partial quote:
'...also according to AMD's rules, "no overclocking/voltage reduction function"'

I can implement overclocking and higher or, supposedly, lower voltages without any warnings through one menu, and with warnings but a broken interface from another menu. Any attempt to use the PBO curves results in higher performance and much higher temps. Given the lack of support, and knowledge, I have decided to get a 5900x as the price has dropped recently and it should provide reasonable productivity performance.

Has anyone seen these issues with 5800x3d and x570 AGESA 1.2.0.8? I am using slow 3200 ECC DRAM so loosing the x3d's cache is going to be noticeable.

 

[BatRastard]

Hi, hopefully it is ok to join this thread! I have a 5800x3d on an Aorus x570 mobo, I recently noticed a drop in performance, no specific figures, of around 5-8% following a BIOS update to an AGESA 1.2.0.8 version. The update notes mention 5800x3d improvements but I have also seen higher temps along with the performance drop.

Having seen the recent articles about burning x3d cpus I decided to try some undervolting with some success using the much maligned easytune. However it does not load on startup even if configured to do so.

I have a ticket open with Gigabyte, running for over a month, they finally came back with the suggestion that overclocking is not possible as per AMD rule partial quote:
'...also according to AMD's rules, "no overclocking/voltage reduction function"'

I can implement overclocking and higher or, supposedly, lower voltages without any warnings through one menu, and with warnings but a broken interface from another menu. Any attempt to use the PBO curves results in higher performance and much higher temps. Given the lack of support, and knowledge, I have decided to get a 5900x as the price has dropped recently and it should provide reasonable productivity performance.

Has anyone seen these issues with 5800x3d and x570 AGESA 1.2.0.8? I am using slow 3200 ECC DRAM so loosing the x3d's cache is going to be noticeable.

You should be seeing a BIOS update with AGESA 1.2.0.A pretty soon that reportedly fixes the EDC bug that's been a thorn in our sides for so long that the first threads about it are in hieroglyphics. Also, I believe JTC did a recent video that the voltage issues with the 7800x3d are also effecting the 5800x3d on Aorus/Gigabyte boards, and Gigabyte has stated fixes and BIOS updates are coming. I don't think they'll have it ready with AGESA 1.2.0.A but AGESA 1.2.0.B (or later) should along with support for the 5600X3D that AMD is releasing shortly ...

 

[JoeN]

You should be seeing a BIOS update with AGESA 1.2.0.A pretty soon that reportedly fixes the EDC bug that's been a thorn in our sides for so long that the first threads about it are in hieroglyphics. Also, I believe JTC did a recent video that the voltage issues with the 7800x3d are also effecting the 5800x3d on Aorus/Gigabyte boards, and Gigabyte has stated fixes and BIOS updates are coming. I don't think they'll have it ready with AGESA 1.2.0.A but AGESA 1.2.0.B (or later) should along with support for the 5600X3D that AMD is releasing shortly ...

Thanks, that's good news, searching for 'EDC bug' returned the discussions I have been looking for! I'll drop my 5800x3d into an unfinished Asrock based gaming rig when I get the time.

 

[Mussels]

1.2.0.0.A solved the EDC bug, according to reports on reddit

It was only an issue on boards that defaulted PBO to max values when it was enabled, which isn't a great idea anyway but a lot of users did this to get to other menus like the curve optimiser
Find your stock PPT/EDC/TDC values and add them to the PBO menu, and you'll find the issue resolved (and control your temps)

You should be seeing a BIOS update with AGESA 1.2.0.A pretty soon that reportedly fixes the EDC bug that's been a thorn in our sides for so long that the first threads about it are in hieroglyphics. Also, I believe JTC did a recent video that the voltage issues with the 7800x3d are also effecting the 5800x3d on Aorus/Gigabyte boards, and Gigabyte has stated fixes and BIOS updates are coming. I don't think they'll have it ready with AGESA 1.2.0.A but AGESA 1.2.0.B (or later) should along with support for the 5600X3D that AMD is releasing shortly ...

Got any info on this? The latest BIOS on the B550I aorus seems to have a default 1.09v SoC voltage (high but safe default, for AM4) but worse ram compatibility than previous BIOSes

DRAM voltage is massively higher than set, with XMP 1.35v reading as 1.41v

 

[Taraquin]

Hi, hopefully it is ok to join this thread! I have a 5800x3d on an Aorus x570 mobo, I recently noticed a drop in performance, no specific figures, of around 5-8% following a BIOS update to an AGESA 1.2.0.8 version. The update notes mention 5800x3d improvements but I have also seen higher temps along with the performance drop.

Having seen the recent articles about burning x3d cpus I decided to try some undervolting with some success using the much maligned easytune. However it does not load on startup even if configured to do so.

I have a ticket open with Gigabyte, running for over a month, they finally came back with the suggestion that overclocking is not possible as per AMD rule partial quote:
'...also according to AMD's rules, "no overclocking/voltage reduction function"'

I can implement overclocking and higher or, supposedly, lower voltages without any warnings through one menu, and with warnings but a broken interface from another menu. Any attempt to use the PBO curves results in higher performance and much higher temps. Given the lack of support, and knowledge, I have decided to get a 5900x as the price has dropped recently and it should provide reasonable productivity performance.

Has anyone seen these issues with 5800x3d and x570 AGESA 1.2.0.8? I am using slow 3200 ECC DRAM so loosing the x3d's cache is going to be noticeable.

If you updated from 1.2.0.6b it runs slightly higher voltages now, but you can use curve optimizer and possibly even an UV offset on top of that to fix it

 

[BatRastard]

1.2.0.0.A solved the EDC bug, according to reports on reddit

It was only an issue on boards that defaulted PBO to max values when it was enabled, which isn't a great idea anyway but a lot of users did this to get to other menus like the curve optimiser
Find your stock PPT/EDC/TDC values and add them to the PBO menu, and you'll find the issue resolved (and control your temps)


Got any info on this? The latest BIOS on the B550I aorus seems to have a default 1.09v SoC voltage (high but safe default, for AM4) but worse ram compatibility than previous BIOSes

DRAM voltage is massively higher than set, with XMP 1.35v reading as 1.41v

Gigabyte boards seem to overvolt and overclock by default. Just about every Gigabyte board I've seen will set the FSB or BKCLK to a few MHz over spec and the same with voltages, but the DRAM voltages on thier B550 boards are concerning. If the sensors are too close to the VRMs, Gigabyte engineers could be compensating for VDROOP. Nonetheless the "Tweaker" section of the BIOS is useless but the AMD Overclocking section in Miscellaneous has a setting for VDDG that can be tweaked as a temporary workaround to bring VDIMM down to 1.35v ...